[All] A Way to Get Rid of Garbage Items?
heronymus
abundance of "useless" items, and was trying to think of a
solution to this problem. I didn't succeed. But I think I have
thought out at least a partial solution.
How about some lovely person programs a menu where a player
can choose between using a white- or black-ball system (or
neither) primarily affecting the visibility of items on the
dungeon floor? White-balling allows the player to select which
items to show, black-balling allows the player to select which
items to hide.
The unwanted garbage items are not only invisible but also
incapable of being picked up or activated / ingested. They
should also not be displayed in the "You see" list. As far as I
can tell, there are three issues concerning either of these
systems:
(1): Should this affect unidentified items?
Of [expletive] course not! Otherwise the character can
choose to hide things like potions of poison, mushrooms of
hallucination, etc. without ever identifying (or unfortunately
trying) one. This eradicates a large part of the game-play, and
frankly, makes me nauseous. Besides, only identified items
should appear in either system's list so that it's not too much
of a spoiler.
(2): But there are too many items!
There are so many items in *band that selecting them
individually in either system would be very time-consuming.
Therefore, the items should be listed in levelled categories (or
individually; player's choice). I'm not a greatly experienced
*bander, so the categories I've listed might be a little shabby:
- All
- Armour (All)
- [too many types to list now]
- Edibles
- Magical Edibles
- Other Edibles
- Magical Items
- Amulets
- Books
- Potions
- Rings
- Rods
- Scrolls
- Staves
- Weapons
- Primary
- Sling/Bow/Crossbow
- Etc.
Each subheading (except obviously weapons and armour) should
also enclose these sub-subheadings:
- Beneficial
- Detrimental
The reason I thought of including weapons and armour is that
a player may have an item that s/he thinks is (temporarily or
permanently) irreplaceable, such as those boots (Feanor?) I've
read are the best, making any other pair of boots redundant
(random artefacts aside).
(3): Isn't this all a bit... fiddly?
Yeah, it's quite fiddly - too fiddly for some. Personally I
wouldn't put either system into action until I'd reached approx.
2500'. Updating the list every time I find an item to categorise
would stretch my patience too far.
However, I still feel that this method would slough off
enough garbage to justify its implementation. I *hate* sifting
through and destroying garbage, and I'm 100% sure I'm not alone
there!
Side-notes:
- There is nothing "immoral" about using either system IMO;
things that are known to be useless are basically treated like
they don't exist anyway, except to be destroyed when on top of
other items in a pile.
- These systems should definitely not stop the -creation- of
the undesired items, i.e. they come back into existence once
they're taken off a black-ball list or placed back on a
white-ball list.
- I think hidden items should still be able to be destroyed,
since a player can -choose- to disregard a mushroom of poison
whereas a Scroll of *Destruction*'s effects cannot.
- Since blocking the visibility of weapons / armour will only
occur in rare circumstances, I guess this means that there
will still be a fair amount of garbage using either of these
systems. But something's better than nothing, and although
I'm not a programmer, I can't see this being too difficult to
implement (changing a visibility flag for the affected items
to false, and preventing them from being involved in any
'activity' such as being picked up, activated or ingested,
etc.).
- I may *very* wrong about that last point, maybe it requires
fundamental code-changes / rewriting large portions of code.
If that's the case, well, damn.
So... I think that's about it, sorry 'bout the length of
this post. If this idea is somehow fundamentally flawed, then
apologies again; I've a terrific hangover!
--
heronymus
Jason Willoughby
; There are so many items in *band that selecting them
; individually in either system would be very time-consuming.
; Therefore, the items should be listed in levelled categories (or
; individually; player's choice).
Fortunately, this bit is already done. See the debug command Create Item
screen.
; - Since blocking the visibility of weapons / armour will only
; occur in rare circumstances, I guess this means that there
; will still be a fair amount of garbage using either of these
; systems. But something's better than nothing, and although
; I'm not a programmer, I can't see this being too difficult to
; implement (changing a visibility flag for the affected items
; to false, and preventing them from being involved in any
; 'activity' such as being picked up, activated or ingested,
; etc.).
I think it would be easier just to auto-destroy. Treat it exactly as if
the 'k' had been pressed on it, but do it automatically in the create item
or identify functions. This has some drawbacks, but it's much more
efficient since you don't have to check every item every turn.
The proper thing to do for weapons and armor, by the way, is to list both
there base type and the egos separately, and then only hide if the player
has checked off both of them.
It's also a giant list of spoilers. You could only list items that have
been identified, but this doesn't work for weapons and armor. Maybe. I
wish I had the code in front of me.
Of course the biggest problem is the save file. You need to save the kill
list in the save file, breaking compatibility. Breaking save files in a
patch looks to be very unpleasant. You end up with files that can be read
by some, but not all, version of 2.8.3, and no easy way to tell which save
file is which. Again, I don't have the code right now, but I think you'd
need an external utility to do the save file conversions.
heronymus
<jwil...@navajo.gate.net> wrote:
>I think it would be easier just to auto-destroy. Treat it exactly as if
>the 'k' had been pressed on it, but do it automatically in the create item
>or identify functions. This has some drawbacks, but it's much more
>efficient since you don't have to check every item every turn.
True. I was keeping my mind open to the possibility that if for
some reason a player desperately needs a hidden item and takes
it off a black-ball list s/he'd want it to turn up on the floor
if it were "really" there. But I've no idea regarding the
implications on the speed of calculating this check for each
item, each turn. I guess it would be very CPU-munching. :-(
>The proper thing to do for weapons and armor, by the way, is to list both
>there base type and the egos separately, and then only hide if the player
>has checked off both of them.
Egos for weapons and armour wouldn't come into it, would
they? Since you have to identify a weapon / piece of armour to
find out which ego-type it has, if any. If someone uses a
white-ball system to show, say, Power items only, then things
like cursed swords would be hidden - I don't think that's fair,
since the player doesn't know what an unidentified sword is yet.
I consider this (put yourself in the dungeon) a "player sees
that red spotted mushroom on the dungeon floor, knows exactly
what it is, and wants nothing to do with it"-type deal. If the
player sees a sowrd, s/he doesn't know what it is yet, so s/he
has to identify it. That's one type of junk this method cannot
deal with, at least not (IMO) without losing some "moral
integrity."
>It's also a giant list of spoilers. You could only list items that have
>been identified, but this doesn't work for weapons and armor. Maybe. I
>wish I had the code in front of me.
It's not a list of spoilers, because it wouldn't go into
gloves, cesti, or gauntlets-type levels of detail. It would only
check for "on-hand armour," "on-foot armour," etc. in cases
where the player thinks s/he has an unbeatable piece of armour
for a particular body-part, or an unbeatable weapon.
>Of course the biggest problem is the save file. You need to save the kill
>list in the save file, breaking compatibility. Breaking save files in a
>patch looks to be very unpleasant. You end up with files that can be read
>by some, but not all, version of 2.8.3, and no easy way to tell which save
>file is which. Again, I don't have the code right now, but I think you'd
>need an external utility to do the save file conversions.
I'm not a programmer so I can't really comment here. Maybe
put it in an external file, and if it doesn't exist, create one
with "White-Balled = All" in it?
--
heronymus
Jason Willoughby
; <jwil...@navajo.gate.net> wrote:
; >The proper thing to do for weapons and armor, by the way, is to list both
; >there base type and the egos separately, and then only hide if the player
; >has checked off both of them.
; Egos for weapons and armour wouldn't come into it, would
; they? Since you have to identify a weapon / piece of armour to
; find out which ego-type it has, if any.
As I said in a bit I snipped, the autokill would also be killed on identify.
Just to save me having to find the 'k' key and maybe accidently deleting
the wrong item.
; >It's also a giant list of spoilers. You could only list items that have
; >been identified, but this doesn't work for weapons and armor. Maybe. I
; >wish I had the code in front of me.
; It's not a list of spoilers, because it wouldn't go into
; gloves, cesti, or gauntlets-type levels of detail. It would only
; check for "on-hand armour," "on-foot armour," etc. in cases
; where the player thinks s/he has an unbeatable piece of armour
; for a particular body-part, or an unbeatable weapon.
Hmm. There's no such thing as an unbeatable weapon. Who wouldn't want a
Holy Avenger Blade of Chaos? Even if you're not going to use it immediately,
you would want to save it at home, just in case. And I can't get rid of
average Robes without also losing Power Dragon Scale Mails?
No, for this to be the most useful, we need to give the player control over
individual wearables.
; >Of course the biggest problem is the save file. You need to save the kill
; >list in the save file, breaking compatibility. Breaking save files in a
; >patch looks to be very unpleasant. You end up with files that can be read
; >by some, but not all, version of 2.8.3, and no easy way to tell which save
; >file is which. Again, I don't have the code right now, but I think you'd
; >need an external utility to do the save file conversions.
; I'm not a programmer so I can't really comment here. Maybe
; put it in an external file, and if it doesn't exist, create one
; with "White-Balled = All" in it?
You could, but an external file only really makes sense for something not
connected to a particular character (e.g. the score file), but you most
certainly do not want an autokill from your last winner to be inherited by
your new character.
Plus I'd hate having to carry around yet another file whenever I want to
switch machines.
James W Sager Iii
You're 'ignore list' will change over time.
First it will be 'ignore stix'
Later 'ignore potions of minimal worth'
Later 'ignore wands that you'd sell during an earlier phase'
Later still 'Ignore anything but what will give a premium cost'
I vote to save it to a different file, but have it tagged like macros
to load up to your most recent version depending on character.
James W Sager Iii
Way to Get Rid .. by hero...@com.hotmail
> they? Since you have to identify a weapon / piece of armour to
> white-ball system to show, say, Power items only, then things
> like cursed swords would be hidden - I don't think that's fair,
> since the player doesn't know what an unidentified sword is yet.
If you figure it out, you prolly just identified it on the ground.
Thusly you can destroy it with 1 keypress too. (if you haven't macroed
id to item on ground and kill top intem on ground it is extremely useful
when you start digging deeper in the dungeon)
The other thing is pseudo-id, in which case its in your inventory and
you need to deal with it anyway.
Also you mentioned about CPU crunching if too many items are on the floor:
A) You can use the auto item cruncher(not sure how it worx, and this
concerns with memory)
B) Maybe you need to make it 'auto destroy'(easier to code)
C) Dramatic recoding of functionality of checking floor items if the
speed concern matters.
Julian Lighton
Jason Willoughby <jwil...@navajo.gate.net> wrote:
>Of course the biggest problem is the save file. You need to save the kill
>list in the save file, breaking compatibility. Breaking save files in a
>patch looks to be very unpleasant. You end up with files that can be read
>by some, but not all, version of 2.8.3, and no easy way to tell which save
>file is which.
Take a look at save.c, wr_xtra(). There's room. (You get to cheese
Ben off by using space that he may consider to be reserved, but it's
there.)
--
Julian Lighton jl...@fragment.com
Death, destruction, and plastic forks
William Tanksley
>Jason Willoughby <jwil...@navajo.gate.net> wrote:
>>Of course the biggest problem is the save file. You need to save the kill
>>list in the save file, breaking compatibility. Breaking save files in a
>>patch looks to be very unpleasant. You end up with files that can be read
>>by some, but not all, version of 2.8.3, and no easy way to tell which save
>>file is which.
>Take a look at save.c, wr_xtra(). There's room. (You get to cheese
>Ben off by using space that he may consider to be reserved, but it's
>there.)
Another solution is to keep your killfile in a format similar to the macro
files -- in fact, perhaps right IN the macro file.
>Julian Lighton jl...@fragment.com
--
-William "Billy" Tanksley, in hoc signo hack
heronymus
<jwil...@dakota.gate.net> wrote:
>; Egos for weapons and armour wouldn't come into it, would
>; they? Since you have to identify a weapon / piece of armour to
>; find out which ego-type it has, if any.
>As I said in a bit I snipped, the autokill would also be killed on identify.
>Just to save me having to find the 'k' key and maybe accidently deleting
>the wrong item.
Sounds good. But the player *must identify weapons and armour
beforehand.* A player can't instantly tell what a sword is just
by looking at it (well maybe except in rare cases of pseudo-ID),
and I want this to be realistic.
>Hmm. There's no such thing as an unbeatable weapon. Who wouldn't want a
>Holy Avenger Blade of Chaos? Even if you're not going to use it immediately,
>you would want to save it at home, just in case.
What if you already had one, two, three? Sure, you can get
better ones, but maybe you just don't feel that there's enough
chance so you don't want to deal with swords anymore. (shrug)
Just a thought. Maybe not likely, but very possible.
>And I can't get rid of
>average Robes without also losing Power Dragon Scale Mails?
>No, for this to be the most useful, we need to give the player control over
>individual wearables.
That's too spoilery. It's unfortunate, and a clear downfall
of this method, but I just can't see a way around it unless a
flag is set for each type of ego, each type of "on-foot" armour,
etc. which declares that it's been seen before. If that's
feasible, great.
>; I'm not a programmer so I can't really comment here. Maybe
>; put it in an external file, and if it doesn't exist, create one
>; with "White-Balled = All" in it?
>You could, but an external file only really makes sense for something not
>connected to a particular character (e.g. the score file), but you most
>certainly do not want an autokill from your last winner to be inherited by
>your new character.
That's right. What I meant to say is have the info in a
character-specific external file. This is definitely related to
each individual character.
>Plus I'd hate having to carry around yet another file whenever I want to
>switch machines.
If that's more annoying than sifting through oodles of crap
every time you enter the dungeon then I guess this isn't
something that'd interest you. Where the save-file is "GROD",
you could have the item-file as "GROD.ITM" or something, so that
you can use wildcards to copy 'em over.
--
heronymus
Lev Zakrevski
1) Interface suggestion - when you "k"ill an item, you can "K"ill it
instead - it will add it to black list.
2) Give high level players (at least, warrior-type) ability to
automatically identify "average" armor/weapons (maybe "good" items too)
and add them to black list. So, player will see only excellent items,
selected wands/potions/scrolls/books etc. (if he wants).
Lev Zakrevski
Jason Willoughby
> <jwil...@dakota.gate.net> wrote:
>>As I said in a bit I snipped, the autokill would also be killed on identify.
My, that's awkward. I meant either "autokill invoked" or "item killed"
and managed to use both.
Anyway...
>>Hmm. There's no such thing as an unbeatable weapon. Who wouldn't want a
>>Holy Avenger Blade of Chaos? Even if you're not going to use it immediately,
>>you would want to save it at home, just in case.
> What if you already had one, two, three? Sure, you can get
> better ones, but maybe you just don't feel that there's enough
> chance so you don't want to deal with swords anymore. (shrug)
> Just a thought. Maybe not likely, but very possible.
I still think there should be a fine enough distinction that I can start
cutting out average daggers long before I'm sure I'll never need another
non-artifact weapon.
>>No, for this to be the most useful, we need to give the player control over
>>individual wearables.
> That's too spoilery. It's unfortunate, and a clear downfall
> of this method, but I just can't see a way around it unless a
> flag is set for each type of ego, each type of "on-foot" armour,
> etc. which declares that it's been seen before. If that's
> feasible, great.
Well, we already have the aware flag for all objects, which is how the
game tells the difference between "A Blue Potion" and "A Potion of
Experience". Currently it defaults to true for weapons and armor
("unflavored item") and is mostly unused. I believe it could be pressed
in to service without too much trouble.
There is no such flag for egos, but there are only about a quarter as
many ego items as items, so adding an extra byte isn't too bad.
All-in-all, it should take about a K of extra memory, and do almost
nothing to performance.
>>Plus I'd hate having to carry around yet another file whenever I want to
>>switch machines.
> If that's more annoying than sifting through oodles of crap
> every time you enter the dungeon then I guess this isn't
> something that'd interest you. Where the save-file is "GROD",
> you could have the item-file as "GROD.ITM" or something, so that
> you can use wildcards to copy 'em over.
Save file names change from platform to platform, and mine
("1000.Jwilloug") bears no relation to the character name. But anyway,
I agree with you. The external file, while not optimal, is the best we
can do. For a patch, at least. I'd recommend that variant authors use
the save file.
So is anybody going to do this? I'd take a crack at it, but I've got
terms projects and finals for the next couple weeks, and won't be able
to get to it until around Christmas. But if no one else beats me to the
punch, I'll do it.
James W Sager Iii
Way to Get Rid .. by Lev Zakr...@bu.edu
> automatically identify "average" armor/weapons (maybe "good" items too)
> and add them to black list. So, player will see only excellent items,
> selected wands/potions/scrolls/books etc. (if he wants).
This is an actual change to the game. I think that would be a cool
addition though. It really is rediculous: Buying 100+ id scrolls. A
warrior fights. He can tell by the days he's been down in the dungeon a
blade of exceptional quality. I still find it kinda perverse how a
warrior can use pretty much all the utility spells a mage can. There is
no sense in spending your life in arcane magic if you can buy the same
and even greater effect for pennies on the dollar. I don't know how to
propose a change to fix the arcane thing, cuz if you make warriors not
use scrolls, you loose teleport, and warriors become weaker.
James W Sager Iii
> better ones, but maybe you just don't feel that there's enough
> chance so you don't want to deal with swords anymore. (shrug)
> Just a thought. Maybe not likely, but very possible.
LOL! I just thought of something cool
So you have troubles with DEATH SWORDS and CLOAKERS?
Well pal, I got the solution for you.
Just never look for cloaks and swords, and the only thing you'll
see is enemies.
Then again. If Topi gets his hands on this code, he'll turn cloakers
and death swords invisible.
LOL.
The sword that you didn't think was worth looking at hits you -more-
hits you -more-
beats you about the face and neck -more-
You die, but no one obviously cares because they have elf corpses on ignore.
Ben Harrison
My *intention* with the following lines in "load2.c"
/* Later use (always zero) */
rd_u32b(&tmp32u);
/* Later use (always zero) */
rd_u32b(&tmp32u);
has always been to set up some kind of "savefile extension system",
in which instead of "zero" and "zero", there would instead be a pair
of numbers, the first being an "extension id" and the second being
a "size", and then they would be followed by that many bytes of data,
which, if the game understood the extension, would be processed, and
if not, would be ignored. This would repeat until a pair of zeros
was encountered.
However, I also have intended to rip out the whole savefile hackery
and replace it with some rewrite of the code in "save.c" wrapped in
"#ifdef FUTURE_SAVEFILES", which, as should be obvious from the
comments below, needs some work. Oh, and all the occurances of
"2.8.0" should be "2.9.0". And the comments about splitting the
uniques out of the monsters were mainly to make it easier to add
more monsters to the game without breaking the memory limitations.
I would imagine that I should actually extend the header to include
a "subtype", increased by one each time the savefile format of the
given "type" changes, to remove the need for annoying checks on the
actual version numbers. Plus, it could be argued that the "type"
could be broken into a "user" (four bytes) and "type" (two bytes),
to allow people to acquire a unique user value, and then be able
to make up a whole bunch of extensions, without worrying about any
collisions (aside from the obvious one inherent in picking a user
id, but there are all kinds of ways to deal with that issue). Also,
the "check" and "stamp" fields could be combined into a "csum" field,
since it is only there to detect gratuitous file corruption, or, in
fact, just removed entirely, since the file itself would have its own
checksum of some kind. Also, the header should contain a tiny textual
description of the extension itself, so error messages can be slightly
less silly looking.
Thus, the header could be 32 bytes, of the following form:
user (4 bytes) -- The user id
type (2 bytes) -- The extension id
vers (2 bytes) -- The extension version
size (2 bytes) -- The size of the data
desc (20 bytes) -- The description of the extension
There are a variety of (somewhat obvious) ways to assign "user"
ids, and each user would simply assign "type" and "vers" values
in ascending order, starting with zero. The "size" only needs
two bytes, because the support code for creating the savefile
will only allow 65535 bytes for the temp space anyway (the whole
block will be generated in memory and then written out to disk
as a whole). If you need more than "size" bytes, simply break
your extension into multiple pieces, each with a different "type".
User "0" would be used for Vanilla Angband Extensions, with type "0"
being used to indicate the end of the file, with the "data" including
some kind of "checksum" information for the rest of the file.
This is, of course, only a suggestion for future work.... :-)
/*
* XXX XXX XXX Ignore this for now...
*
* The basic format of Angband 2.8.0 (and later) savefiles is simple.
*
* The savefile itself is a "header" (4 bytes) plus a series of "blocks",
* plus, perhaps, some form of "footer" at the end.
*
* The "header" contains information about the "version" of the savefile.
* Conveniently, pre-2.8.0 savefiles also use a 4 byte header, though the
* interpretation of the "sf_extra" byte is very different. Unfortunately,
* savefiles from Angband 2.5.X reverse the sf_major and sf_minor fields,
* and must be handled specially, until we decide to start ignoring them.
*
* Each "block" is a "type" (2 bytes), plus a "size" (2 bytes), plus "data",
* plus a "check" (2 bytes), plus a "stamp" (2 bytes). The format of the
* "check" and "stamp" bytes is still being contemplated, but it would be
* nice for one to be a simple byte-checksum, and the other to be a complex
* additive checksum of some kind. Both should be zero if the block is empty.
*
* Standard types:
* TYPE_BIRTH --> creation info
* TYPE_OPTIONS --> option settings
* TYPE_MESSAGES --> message recall
* TYPE_PLAYER --> player information
* TYPE_SPELLS --> spell information
* TYPE_INVEN --> player inven/equip
* TYPE_STORES --> store information
* TYPE_RACES --> monster race data
* TYPE_KINDS --> object kind data
* TYPE_UNIQUES --> unique info
* TYPE_ARTIFACTS --> artifact info
* TYPE_QUESTS --> quest info
*
* Dungeon information:
* TYPE_DUNGEON --> dungeon info
* TYPE_FEATURES --> dungeon features
* TYPE_OBJECTS --> dungeon objects
* TYPE_MONSTERS --> dungeon monsters
*
* Conversions:
* Break old "races" into normals/uniques
* Extract info about the "unique" monsters
*
* Question:
* Should there be a single "block" for info about all the stores, or one
* "block" for each store? Or one "block", which contains "sub-blocks" of
* some kind? Should we dump every "sub-block", or just the "useful" ones?
*
* Question:
* Should the normals/uniques be broken for 2.8.0, or should 2.8.0 simply
* be a "fixed point" into which older savefiles are converted, and then
* future versions could ignore older savefiles, and the "conversions"
* would be much simpler.
*/
--- Ben ---
jl...@nowhere.fragment.com (Julian Lighton) writes:
> In article <81rbvp$n88$1...@news.gate.net>,
Screwtape
Ben Harrison schrieb:
>Thus, the header could be 32 bytes, of the following form:
>
> user (4 bytes) -- The user id
> type (2 bytes) -- The extension id
> vers (2 bytes) -- The extension version
> size (2 bytes) -- The size of the data
> desc (20 bytes) -- The description of the extension
Is this the header of the entire save file, or of "extra chunks" added on the
end?
If it's for the entire save file, might I ask that a field be added containing
constant text (say, "Angband Save File") to make the lives of utilities like
find(1) easier?
Screwtape
<lurk>
--
,------------------------------------------------- ------ ---- -- - - -
| Screwtape | Reply-To: is munged on Usenet | members.xoom.com/thristian
|--------------------------------------------- ---- ---- --- -- - - - -
|
| I wish I could change my .sig without it being retroactive.
|
Gwidon S. Naskrent
wrote:
>If it's for the entire save file, might I ask that a field be added containing
>constant text (say, "Angband Save File") to make the lives of utilities like
>find(1) easier?
You must mean file(1) and some magic bytes at the beginning?
Screwtape
Oops, yes.
--
,------------------------------------------------- ------ ---- -- - - -
| Screwtape | Reply-To: is munged on Usenet | members.xoom.com/thristian
|--------------------------------------------- ---- ---- --- -- - - - -
|
| "But the shape of a lemon, that's subjective..."
|
James W Sager Iii
Way to Get Rid .. by Scre...@ferd2.thristia
utilities >>like
I think the idea of writing the code to interact with the save file
better, but I am a strong advocate of not having garbage items squelched
tagged to characters.
Problems:
1) You start a new character over a dead character.
Now you need to untag.
2) You need to rewrite which ones you want untagged every time you play
the game.
Anyone have any more ideas on this? I think if you saved squelched
items with a character, you run into more depression when your character
dies than normal. You need to do more 'code' work to make a new guy :(
If it was saved separately, you could just load up your previous squelch list.
One more problem/bug I forsee.
Character A ids a mushroom of paranoia.
Squelches it.
Dies
Character B loads up squelch file.
Un id mushroom should be seen.
Easy way to deal: if not id then do not squelch, unless equipment.
If someone wants to write autosquelching for good/excellent -x,-x whatever
go for, but I think that would just be a bonus.
Ed Cogburn
>
> Excerpts from netnews.rec.games.roguelike.angband: 1-Dec-99 Re: [All] A
> Way to Get Rid .. by Scre...@ferd2.thristia
> > >>constant text (say, "Angband Save File") to make the lives of
> > >>utilities like
>
> I think the idea of writing the code to interact with the save file
> better, but I am a strong advocate of not having garbage items squelched
> tagged to characters.
>
> Problems:
>
> 1) You start a new character over a dead character.
> Now you need to untag.
Have these squelch options be reset at the death of a player. Maybe
this should be done in a generic way, since there might be other cases
in the future where some options should automatically reset
themselves, when a char dies.
>
> 2) You need to rewrite which ones you want untagged every time you play
> the game.
Well just how many options will there be? Imagine two columns with
the space bar acting as the toggle, and arrow keys to move up and
down. This wouldn't be so bad, I think.
>
> Anyone have any more ideas on this? I think if you saved squelched
> items with a character, you run into more depression when your character
> dies than normal. You need to do more 'code' work to make a new guy :(
Don't understand this.
>
> If it was saved separately, you could just load up your previous squelch list.
How is this different from saving the squelch list with the
character? When the character dies, and the program saves the reduced
size savefile, which includes the squelch list, the program would
reset the switches.
>
> [snip]
>
> If someone wants to write autosquelching for good/excellent -x,-x whatever
> go for, but I think that would just be a bonus.
To truly be useful, it would have to distinguish a number of items:
items of worthless value (anything cursed for example, except
artifacts), torches/lanterns/oil flasks, food, bad stuff thats been
ID'd ("Potion of Poison"), normal weapons ("Dagger (+0,+0)"), good
weapons ("Dagger (+5,+5)"), great weapons ("Dagger of Westernesse
(+12,+14)"), normal armor, good armor, great armor. DSM's should be
counted as great, you think?
Plus probably a few more.
--
Ed C.
-----------== Posted via Newsfeeds.Com, Uncensored Usenet News ==----------
http://www.newsfeeds.com The Largest Usenet Servers in the World!
------== Over 73,000 Newsgroups - Including Dedicated Binaries Servers ==-----
James W Sager Iii
> > items with a character, you run into more depression when your character
> > dies than normal. You need to do more 'code' work to make a new guy :(
>
>
> Don't understand this.
By manual coding. I mean every time you make a new character.
At least I for one would do this:
broken stix/skulls/blah junked
As I progress to different levels of play, I want different things squelched.
At start, I do want potions of cure serious, and wands.
Mid way through the game my list changes completely.
And deep down, again my list changes.
But with every character I play, the list will always be the same at
different points in my characters life. You want items that sell for a
lot or are useful, and don't want ones that waste space for their
monetary value.
As you get rich, you don't want certain things anymore.
What I mean of recoding is the user playing the game would have to
change his list ALL the time if it couldn't be saved. Picture having 4
levels of wealth in squelch lists. As you hit a 'level of wealth' you
load that squelch file.
now picture instead redoing the set list every time you load a character...
It helps some, but it would be extra work for the player.
Joseph Oberlander
> change his list ALL the time if it couldn't be saved. Picture having 4
> levels of wealth in squelch lists. As you hit a 'level of wealth' you
> load that squelch file.
> now picture instead redoing the set list every time you load a character...
> It helps some, but it would be extra work for the player.
I don't see why a manual list wouldn't work - you can change the menu
like any other prefs menu. Just save the config file like you do macros/
layouts/etc. if you wish.
I find it easy to get to a point and no longer CARE about anything other
than exceptional weapons(say I find the Glaive of Pain) and better.
A few keypresses and I am set. As simple as any other method and actually
not that hard to code as the auto-destroy worthless items code is already
in place.
Ed Cogburn
>
> Way to Get Rid .. by Ed Cog...@greene.xtn.ne
> > > Anyone have any more ideas on this? I think if you saved squelched
> > > items with a character, you run into more depression when your character
> > > dies than normal. You need to do more 'code' work to make a new guy :(
> >
> >
> > Don't understand this.
>
> By manual coding. I mean every time you make a new character.
Everytime you lose a character the squelch list is reset to its
defaults.
>
> At least I for one would do this:
> broken stix/skulls/blah junked
This could be the default state, so starting out there is no need to
change anything.
>
> As I progress to different levels of play, I want different things squelched.
>
> At start, I do want potions of cure serious, and wands.
Only potions or wands of zero value (we rely on the monetary price
that the game calculates to tell us if the thing is worthwhile) are
squelched. The good stuff remains.
>
> Mid way through the game my list changes completely.
Ok, hit '=', select 'Squelch Junk', a page with toggle options down
the page shows up, use arrow keys to move down with the space bar to
toggle the flags on/off. How hard can that be?
>
> And deep down, again my list changes.
I realize that things will change as you go deeper, but unless you
are talking about 3 or 4 flags for every specific item
(cursed/mundane/good/great) in the game, then there probably be no
more than a dozen flags; it wouldn't be a big deal to go thru that
list every once in a great while.
If you *are* talking about 4 flags for every specific item, I have a
strong suspicion that no one will go to that much trouble to solve the
problem at hand; its just asking too much.
>
> But with every character I play, the list will always be the same at
> different points in my characters life. You want items that sell for a
> lot or are useful, and don't want ones that waste space for their
> monetary value.
>
> As you get rich, you don't want certain things anymore.
Of course. A squelch routine that works on catagories of things,
would prevent some things out of a catagory to show up that you may
want to see, but the alternative is serious in its scope and
complexity. Count up all the different items (sans artifacts), and
never mind all the extra items created in the variants, and see what
number you come up with. My guess is close to a hundred flags and
most certainly over a hundred in many of the variants. If we base
things on the categories, we have no more than a dozen flags to worry
about, at most, and I'm probably overstating the real number. Since
character development tends to slow down as the game progresses, its
not as if you'll be needing to manage these switches every hour of
playing.
Keep in mind, that most people, I believe, will be quite happy with a
routine that squelches the natural junk (sticks, skeletons, skulls,
etc), bad potions, bad scrolls, bad wands/staffs/rods, cursed weapons,
and cursed armor. That alone would make the routine worth it. Notice
that *all* of the above can be taken care of just by checking for its
monetary value. For later on, deeper in the dungeon, we could also
squelch food items, light items, (torch/lantern/oil flask) mundane
weapons/armor, and even later, good weapons/armor. This adds up to
about 7 flags, thats it. Look what it leaves: good scrolls, potions,
wands, staffs, rods, ego weapons/armor and artifacts. That would at
least be a heck of an improvement over the status quo.
> [snip]
James W Sager Iii
>
>
> Ok, hit '=', select 'Squelch Junk', a page with toggle options down
> the page shows up, use arrow keys to move down with the space bar to
> toggle the flags on/off. How hard can that be?
Look at it in a computer science point of view.
If you have 5 states of wealth, each state you need to squelch N items.
Then for each character that hits the a state you must spend N steps
squelching an item.
If you could save the list externally as opposed to on the character,
then you could save squelch lists. Either way it saves ALOT of VERY
BORING work to the band player. There is a real minor amount of boring
work involved if coded onto player lists instead of external files like
macros.
Macro's don't give a lot of flexibility in changing over time. Your auto
light might go from a magic spell to the phial. But it is REALLy nice
to load up a generic macro file from another player. Imagine having to
rewrite basic rest, identify, destroy item and lighting macros for every
player
when you died and you get a hint of what it would be like to tack
squelch lists player specific.
James W Sager Iii
kinda petty arguing over details especially since I am not doing the
physical work to actually code it. My suspicion is though that if I
code this, and play 3 characters to completion that the time spent
coding would be much less than the time I would sift through stuff on
the floor ^^
I'm just discusssing though because it will help whoever undertakes this
coding chunk make an informed decision on how he/she wants it like. Its
always good to have options in front of you when you go to sit into code.
Excerpts from netnews.rec.games.roguelike.angband: 2-Dec-99 Re: [All] A
Way to Get Rid .. by Ed Cog...@greene.xtn.ne
> Keep in mind, that most people, I believe, will be quite happy with a
> routine that squelches the natural junk (sticks, skeletons, skulls,
> etc), bad potions, bad scrolls, bad wands/staffs/rods, cursed weapons,
> and cursed armor. That alone would make the routine worth it.
Yes :)
>Notice
> that *all* of the above can be taken care of just by checking for its
> monetary value.
That is a nobel idea.
Important things you want wouldn't show up if a money only squelch is
enabled for any significant $. But to say: if not worth 10$ then don't
show would be a good squelch as it would ignore 'junk' as anyone sees it.
> For later on, deeper in the dungeon, we could also
> squelch food items, light items, (torch/lantern/oil flask) mundane
> weapons/armor, and even later, good weapons/armor.
See I originally didn't think to go as far to even worry about flags on
weapons/armor.
Though it would kick ass if there was an auto throw out ability for warriors
with good/mundane items. They drop it from inventory and ignore.
Or ignore upon identify on the ground.
> This adds up to
> about 7 flags, thats it.
You are generalizing the code more than I would have the foresight to be
able to... I was thinking more along the line of an item by item yes/no
list.
> Look what it leaves: good scrolls, potions,
> wands, staffs, rods, ego weapons/armor and artifacts. That would at
> least be a heck of an improvement over the status quo.
I agree, but I would think the yes/no item to item would be easier to
code, although I don't know what kind of processing time it would
take(I'm thinking very nominal)
Peter Ellis
>
>Look at it in a computer science point of view.
>
>If you have 5 states of wealth, each state you need to squelch N items.
>Then for each character that hits the a state you must spend N steps
>squelching an item.
>
But to be fair, it's just *not* down to states of wealth. If I'm a
priest, and don't have fire immunity, I want to keep picking up extras of
the priest books even if they're monetarily worthless to me, for at least
some of the game.
Depending on my equipment, I may have only a tacky set of boots of
stealth, but an artifact sword -- so I want to see any ego boots, but can
ignore all but artifact swords... that's a distinction that can't be made
merely monetarily.
And to be honest, the main boring bit isn't *seeing* the items... I can
just ignore them. It's having to _identify_ the bloody things in the
first place. When you're at level 80, searching through all the drops
finding <good> item after item, when you only give a shit about ego items
or better is a *pain*
_Eventually_ you get to a state where you just trash everything, looking
for artifacts only; but the untermediate stage, where you want to be
easily able to distinguish between ego items and <good> items is the
awkward stage.
For me, the easiest way to solve this would be a progression like this:
Level 15: Can detect cursed items on sight
Level 25: Can detect <good> or better items on sight ( = weak pseudo-ID)
Level 35: Can detect all grades of item on sight ( = strong pseudo-ID)
Sure, if you want, you can then add code to stop them even being visible,
but for me a radical revamp of the pseudo-ID system would be by far the
most preferred option.
(In which case a flag for each inventory slot saying whether you want to
ignore cursed, normal, good, excellent would be sufficient. Add one
more flag for cruft like pottery, one for food. With scrolls, potions,
wands etc. there's so much variety that it'll be futile trying to
discriminate, so just leave them all visible)
>If you could save the list externally as opposed to on the character,
>then you could save squelch lists. Either way it saves ALOT of VERY
>BORING work to the band player.
At 12 items in the inventory, plus cruft, plus food, that's only 14 flags
to set. That would take me about 14 seconds. You think that's boring?
>
>Macro's don't give a lot of flexibility in changing over time. Your auto
>light might go from a magic spell to the phial. But it is REALLy nice
>to load up a generic macro file from another player. Imagine having to
>rewrite basic rest, identify, destroy item and lighting macros for every
>player when you died and you get a hint of what it would be like to tack
>squelch lists player specific.
>
Yes, but a macro you have to use *every* time you want to perform an
action. With a squelch list, you'd set it up once, adjust it maybe
half-a-dozen times during the course of a complete game, taking maybe
thirty seconds each time, and save yourself *so* much hassle. I really
don't see what you're aiming at with this comparison.
Peter
Jason Willoughby
> If you could save the list externally as opposed to on the character,
> then you could save squelch lists. Either way it saves ALOT of VERY
> work involved if coded onto player lists instead of external files like
> macros.
Keep in mind that the two aren't mutually exclusive. Options, for
instance, are in the save file but can also be dumped and loaded from
files. And how do you plan to autoload these external files? Remember
that the charcater name doesn't necessarily change for every new
character. Or would we have to load them by hand every time?
Keith Willoughby
, Peter Ellis <pj...@hermes.cam.ac.uk> writes
>On Thu, 2 Dec 1999, James W Sager Iii wrote:
>>If you could save the list externally as opposed to on the character,
>>then you could save squelch lists. Either way it saves ALOT of VERY
>>BORING work to the band player.
>
>to set. That would take me about 14 seconds. You think that's boring?
Seeing as IDing and destroying is by far and away the most tedious part
of *band, any scheme to make it easier gets my vote. Ben/Zang DevTeam -
please please please implement something of this sort.
--
Keith Willoughby
James W Sager Iii
> >Then for each character that hits the a state you must spend N steps
> >squelching an item.
> >
>
> But to be fair, it's just *not* down to states of wealth. If I'm a
> priest, and don't have fire immunity, I want to keep picking up extras of
> the priest books even if they're monetarily worthless to me, for at least
> some of the game.
My bad, I'm miscommunicating to you.
You squelch x individual items.
This list of squelched items is saved to a file.
I call this file a wealth state:
Example State 1) broken stix squelched
State 2) scrolls of confuse monster squelched
State 3) wands squelched
State 4) Cure critical potions squelched
State 5) all but the best items squelched
If you save to a character, you need to redo each state in the editor
when you reach them. If they are saved to a seperate file you could
just load/save like wealth_1.whatever
Sorry about not being more clear. I hope I communicate this rcorrectly,
we're on different thought paths more than once.
James W Sager Iii
> > then you could save squelch lists. Either way it saves ALOT of VERY
> > work involved if coded onto player lists instead of external files like
> > macros.
>
> Keep in mind that the two aren't mutually exclusive. Options, for
> instance, are in the save file but can also be dumped and loaded from
> mind that the charcater name doesn't necessarily change for every new
> character. Or would we have to load them by hand every time?
LOL atta a beast of pure hatred in your post
Anyways:
Macros are autoloaded with character though they save to an external file.
You would have to load them by hand only when making a new character or
adjusting the current one.
Ie:
GUG the fighter is loaded.
GUG.squelch list is auto loaded and contains the information for a deep
level squelching.
GUG the fighter gets hit by a gravity hound and tossed into a Raal's
tome room then dies to a creeping silver coin.
You load up Gug again, but the old file is tagged to him.
You make a new guy.
You simply load 1_wealth then save as GUG. whatever
1_wealth
2_wealth
verses:
Load up Gug
make new guy
re-edit squelch list.
There is a larger degree of work for the player in the second instance.
Ed Cogburn
>
> > > Mid way through the game my list changes completely.
> >
> >
> > Ok, hit '=', select 'Squelch Junk', a page with toggle options down
> > the page shows up, use arrow keys to move down with the space bar to
> > toggle the flags on/off. How hard can that be?
>
>
> Then for each character that hits the a state you must spend N steps
> squelching an item.
>
> then you could save squelch lists. Either way it saves ALOT of VERY
> BORING work to the band player. There is a real minor amount of boring
> work involved if coded onto player lists instead of external files like
> macros.
>
> light might go from a magic spell to the phial. But it is REALLy nice
> to load up a generic macro file from another player. Imagine having to
> rewrite basic rest, identify, destroy item and lighting macros for every
> player
> when you died and you get a hint of what it would be like to tack
> squelch lists player specific.
If we tried to give fine-grained control over what is and isn't
squelched, meaning we had switches for every single item type, then we
would need to do as you suggest.
I'm just trying to point out that this is a *major* programming
effort that you are describing, and I would be quite happy with
something much simpler, yet useful as well since it would at least
eliminate the obvious junk. Besides, it could be a good 'first step'
that would be improved upon as time goes on.
Jason Willoughby
> LOL atta a beast of pure hatred in your post
Well, 'tis the season
http://www.angryflower.com/christ.gif
(An unfortunate name, given the recent discussion. It's actually short
for "Christmas")
> Anyways:
> Macros are autoloaded with character though they save to an external file.
But macros aren't dangerous, squelching is. Autoloading a 4000' squelch
list for character with a recall depth of 0 is just stupid.
> You load up Gug again, but the old file is tagged to him.
> You make a new guy.
> You simply load 1_wealth then save as GUG. whatever
And if you forget, and go down the stairs? What if you start next to
Maggot and squelch his drop?
> verses:
> Load up Gug
> make new guy
> re-edit squelch list.
> There is a larger degree of work for the player in the second instance.
Only if you don't allow them to be dumped to file (*without*
autoloading the file, they're in the save, remember). But why would you
do that?
Besides, who needs a squelch list at 50'? You don't start drowning in
junk till past stat gain.
Andrew Tait
I'm replying to something just to get into this thread.
Funny how you can just 'see' things in your head. How about this:
When you destroy any object (floor or backpack), instead of 'Are you sure
(y/n)?' you get an extended menu, say 'Destroy Wand of Light (y/n/d/p/q)?'
'y' Destroys the Wand. All future WoLight will get the prompt to destroy
(normal behaviour)
'n' Doesn't destroy the Wand. Normal behaviour.
'd' Destroys the Wand and sets a flag to destroy all future Wands of Light
automagically.
'p' Destroys the Wand and sets a flag to always ask you immediately if you
want to destroy the Wand if you walk over it. [A bit like autopickup, but
you 'k' instead of 'g' if you hit 'y']
'q' Destroys the Wand and sets a flag to remove the prompt in future, so
that 'k' will destroy the object the way 'bad items' are destroyed in Z just
now. You need to press 'k', but no prompt thereafter.
There should be a menu which lets you set the flag [I'm guessing it would be
either 'never', 'noprompt', 'prompt' or 'always'] for all the pseudo-ID
classes. {worthless}->{special}. With pseudo-ID, the behaviour would need
to along the lines of "You feel the Gauntlets in your backpack are cursed.
You destroy the Gauntlets {cursed}"
Now the nibbly bit. Suddenly you want to pick up a Wand of Light, and it's
been set to 'd', so that you destroy them on sight. The easiest way to
implement this would be to have a togglable option to temporarily switch all
's' objects to behave as 'p'. "Destroy Wand of Light (y/n/c/p/q)? [n] Pick
up Wand of Light (y/n)? [y]"
The 'c' option will 'c'lear the Autodestroy flag completely on the object.
Lastly, the 'd' mode takes up a turn if you walk over the object on the
floor (maybe not? Autopickup doesn't, so perhaps Autodestroyfloor
shouldn't), so it /must/ be disabled if you are in a 'disturbed' state. I
don't know how easy/hard this is to determine.
Totty <8^)
Andrew Tait
> In article <828s14$q...@singer.cent.gla.ac.uk> you wrote:
>
> : When you destroy any object (floor or backpack), instead of 'Are you sure
> : (y/n)?' you get an extended menu, say 'Destroy Wand of Light (y/n/d/p/q)?'
>
> Sounds interesting.
>
> : Now the nibbly bit. Suddenly you want to pick up a Wand of Light, and it's
> : been set to 'd', so that you destroy them on sight. The easiest way to
> : implement this would be to have a togglable option to temporarily switch all
> : 's' objects to behave as 'p'. "Destroy Wand of Light (y/n/c/p/q)? [n] Pick
> : up Wand of Light (y/n)? [y]"
>
> : The 'c' option will 'c'lear the Autodestroy flag completely on the object.
>
> This sounds a bit hairy, however. Why not make c an option on the look
> command, so you could just change it by looking at it?
I'd say both. In Z, if you don't have extended stacking list thingy
on, you can only see the top item, and the ultimate aim of the
AutoDestroy would be to make standing on a stack of objects go like:
"Pick up Gauntlets [y/n]? -more- You destroy the Broken Skull -more-
You destroy the Wand of Light -more-" etc. Switching all AutoDestroys
to 'p'rompt mode would change the behaviour to a tolerable level (In
the above you'd see "... Destroy Broken Skull? (y/n/c/p/q) [y] -more-
Destroy Wand of Light? (y/n/c/p/q) [c] -more- Pick up Wand of Light?
(y/n) [y]) Now you can switch AutoDestroy back to "Deadly" mode, and
still pick up WoL.
But I agree that being able to switch modes on a item at /any/ point
could be very useful.
What you really want is TrashMonster The Borg[TM], who is an
undestroyable wraithlike, intent on destroying the stuff local to you
which you don't want. TMTB[TM] is always on the lookout for Potion of
Dispel Master, but never finds it ;)
Totty <8^)
Ed Cogburn
>
> [snip]
>
> Besides, who needs a squelch list at 50'? You don't start drowning in
> junk till past stat gain.
This is a good point. I think this feature should be held back until
the player is 25-35 levels or so, when he really *needs* it. Having
this feature in the beginning could be construed as an advantage. We
don't want to make the game easier, we just want to alleviate some of
the tedium faced by high level folks deep in the dungeon.
Ed Cogburn
>
> Hi all,
>
> I'm replying to something just to get into this thread.
>
> Funny how you can just 'see' things in your head. How about this:
>
> (y/n)?' you get an extended menu, say 'Destroy Wand of Light (y/n/d/p/q)?'
>
> (normal behaviour)
> 'n' Doesn't destroy the Wand. Normal behaviour.
> 'd' Destroys the Wand and sets a flag to destroy all future Wands of Light
> automagically.
> 'p' Destroys the Wand and sets a flag to always ask you immediately if you
> want to destroy the Wand if you walk over it. [A bit like autopickup, but
> you 'k' instead of 'g' if you hit 'y']
> 'q' Destroys the Wand and sets a flag to remove the prompt in future, so
> that 'k' will destroy the object the way 'bad items' are destroyed in Z just
> now. You need to press 'k', but no prompt thereafter.
>
> There should be a menu which lets you set the flag [I'm guessing it would be
> either 'never', 'noprompt', 'prompt' or 'always'] for all the pseudo-ID
> classes. {worthless}->{special}. With pseudo-ID, the behaviour would need
> to along the lines of "You feel the Gauntlets in your backpack are cursed.
> You destroy the Gauntlets {cursed}"
>
> been set to 'd', so that you destroy them on sight. The easiest way to
> implement this would be to have a togglable option to temporarily switch all
> 's' objects to behave as 'p'. "Destroy Wand of Light (y/n/c/p/q)? [n] Pick
> up Wand of Light (y/n)? [y]"
>
> The 'c' option will 'c'lear the Autodestroy flag completely on the object.
>
> floor (maybe not? Autopickup doesn't, so perhaps Autodestroyfloor
> shouldn't), so it /must/ be disabled if you are in a 'disturbed' state. I
> don't know how easy/hard this is to determine.
>
> Totty <8^)
You are describing a really complex implementation, whose value
probably doesn't exceed the difficulty in coding it. As for handling
each object individually, do you know how many objects there are in
the game? In [GSN]'s defines.h, it lists a huge number of SV_<item>
defines. I stopped counting at 400... Take away a hundred for the
artifacts, and another hundred for good measure, and you're still
talking about 200+ items that would have to be dealt with on an
individual basis. Can you just imagine the size of the switch
statement it would take to deal with all that? :-)
Andrew Tait
> Andrew Tait wrote:
> > When you destroy any object (floor or backpack), instead of 'Are you sure
> > (y/n)?' you get an extended menu, say 'Destroy Wand of Light (y/n/d/p/q)?'
> You are describing a really complex implementation, whose value
> probably doesn't exceed the difficulty in coding it. As for handling
Is there such a thing? ;) I've coded all sorts of contingency stuff
for programs at work, some of which can be 5 pages or more, and give a
totally pathetic result in terms of user interface. But a /useable/
result. No-one who's playing cares about the difficulty in coding,
just the difficulty in destroying all those Scrolls of Light lying
amonst the swords and boots outside an Orc Pit. :)
> each object individually, do you know how many objects there are in
> the game? In [GSN]'s defines.h, it lists a huge number of SV_<item>
> defines. I stopped counting at 400... Take away a hundred for the
> artifacts, and another hundred for good measure, and you're still
> talking about 200+ items that would have to be dealt with on an
> individual basis. Can you just imagine the size of the switch
> statement it would take to deal with all that? :-)
Surely the whole thing can be chucked into an array. I'm gritting
away at starting coding, but all you need is an array of object
type/object number, surely. No 'special cases'. Everything starts
off unflagged, and if /you/ flag every MHDSN you see to get
destroyed on sight, it's your fault :)
You could be right. I was suggesting the interface from a users point
of view, not a coders.
Totty <8^)
James W Sager Iii
Way to Get Rid .. by Ed Cog...@greene.xtn.ne
> > junk till past stat gain.
>
>
> This is a good point. I think this feature should be held back until
> the player is 25-35 levels or so, when he really *needs* it. Having
> this feature in the beginning could be construed as an advantage. We
> don't want to make the game easier, we just want to alleviate some of
> the tedium faced by high level folks deep in the dungeon.
How could it be an advantage?
Some people like broken stix for theif fodder.
I personally don't like dropping them.
I don't see how this could make the game easier or harder.
I only see it as less tedious.
Now when it comes to detection. There are serious issues there.
I think in the detection case one needs to concern with not
allowing detection to be abused.
A way around this is to allow all items to show up, until looked at.
When looked at, it will say, "nothing important here", or when you
walk over them you get the same message.
Or possibly if items are in line of sight to do this too.
This way significant processing time seems like it could be sured up too.
Jason Willoughby
> You are describing a really complex implementation, whose value
> probably doesn't exceed the difficulty in coding it. As for handling
> the game? In [GSN]'s defines.h, it lists a huge number of SV_<item>
> defines. I stopped counting at 400... Take away a hundred for the
> artifacts, and another hundred for good measure, and you're still
> talking about 200+ items that would have to be dealt with on an
> individual basis. Can you just imagine the size of the switch
> statement it would take to deal with all that? :-)
There are 421, 461, and 508 item types in V, O and Z, respectively (not
counting INSTA_ARTs), but there's no need for a switch. Each item is
specified uniquely by a tval,sval pair, and the "obvious" interface is a
two level list of all tvals (potion, wand, armor...) and then all svals
(water, apple juice...). The work is already done, see the wizard mode
Create Item command.
I would suggest a list of all ego items on top of that, to allow for a
finer grain. And while I don't really see a need for modifications to
the 'k'ill command (I don't think you'll really spend all that much time
manipulating the squelch list), it can't really hurt.
heronymus
<ecog...@greene.xtn.net> wrote:
> This is a good point. I think this feature should be held back until
>the player is 25-35 levels or so, when he really *needs* it. Having
>this feature in the beginning could be construed as an advantage. We
>don't want to make the game easier, we just want to alleviate some of
>the tedium faced by high level folks deep in the dungeon.
How can it be an advantage, if all they can squelch are things
they've already ID'd?
I liked (forgets who it was)'s idea of having a
gradually-increased list of instantly-recognisable items, like
knowing what's mediocre at L25, cursed at L30, good and
worthless at L35, excellent and terrible at L40, and all ego
types at L45 or something.
--
heronymus
Johan Kullstam
> Hi all,
>
> I'm replying to something just to get into this thread.
>
> Funny how you can just 'see' things in your head. How about this:
>
> When you destroy any object (floor or backpack), instead of 'Are you sure
> (y/n)?' you get an extended menu, say 'Destroy Wand of Light
> (y/n/d/p/q)?'
how about this? press O. would you like to destroy all worthless
objects on the entire level?
that's what i want. i'd just like to be able to clobber everything
i've identified and shops will not buy in one fell swoop. kind of
compacting objects upon demand. Z (whatever version i have) seems to
have a problem compacting objects.
i think it's cheating if useless dross isn't created and strewn across
the floor potentially blocking other items from falling. however, i
want some way for crap to get wiped out.
--
J o h a n K u l l s t a m
[kull...@ne.mediaone.net]
Don't Fear the Penguin!
Ed Cogburn
>
> On Sat, 04 Dec 1999 11:10:43 +0000, Ed Cogburn
> <ecog...@greene.xtn.net> wrote:
>
> > This is a good point. I think this feature should be held back until
> >the player is 25-35 levels or so, when he really *needs* it. Having
> >this feature in the beginning could be construed as an advantage. We
> >don't want to make the game easier, we just want to alleviate some of
> >the tedium faced by high level folks deep in the dungeon.
>
> How can it be an advantage, if all they can squelch are things
> they've already ID'd?
I'm talking about suppressing certain types of junk *at creation*
when the player hasn't ID'd them. Now for those things known only by
a flavor until ID'd you have to ID the first one, but after that (if
its on your suppress list) it never shows up again (eg. Wand of Heal
Monster).
The "problem" here is with weapons and armor. If you require the
user to ID them first to find out if its something special, then you
haven't solved the problem everyone is talking about: having to sift
through mountains of weapons, armor, rings, etc. looking for things
which interest you.
That's why there needs to be a minimum limit. By the time a
character gets to level 30 and beyond (the exact level is debatable,
anywhere from 25 to 40?) he has no problem with ID'ing stuff. Magic
users have the spell, and everyone else, by this time, has more than
enough money to buy as many ID scrolls or Perception staffs as they
can carry.
Beyond this point, dealing with masses of junk really becomes
*tedious*, and that is the thing I'm trying to eliminate.
>
> I liked (forgets who it was)'s idea of having a
> gradually-increased list of instantly-recognisable items, like
> knowing what's mediocre at L25, cursed at L30, good and
> worthless at L35, excellent and terrible at L40, and all ego
> types at L45 or something.
Huh? What's the level got to do with this? A cursed item is junk
whether its on level 25 or 40, and what is an "instantly-recognizable"
item?
How do you program the game to "understand" what is "mediocre" by our
standards at a given depth? Your on the virge of basically asking for
a non-trivial AI routine to make all these decisions. Now who's going
to write that just to suppress junk items? We've got plenty of room
for improvement in monster AI, but progress is slow, since this is a
difficult thing to do, and few people are able to do it. I'll bet
there are less than a dozen people who really understand the AI code
we've got now, and improving on it isn't an easy thing to do.
Ed Cogburn
>
> Way to Get Rid .. by Ed Cog...@greene.xtn.ne
> > > Besides, who needs a squelch list at 50'? You don't start drowning in
> > > junk till past stat gain.
> >
> >
> > the player is 25-35 levels or so, when he really *needs* it. Having
> > this feature in the beginning could be construed as an advantage. We
> > don't want to make the game easier, we just want to alleviate some of
> > the tedium faced by high level folks deep in the dungeon.
>
It would be an advantage for early characters. Once a char has
reached lvl 25 or so, they are more than able to identify everything
they come across. The mage types have the spell, and others have
enough money to by as many ID scrolls as they can carry. At this
point, having the game suppress stuff without you having to read an ID
scroll, is not much of an advantage, but is very appealing at this
point since you are now sifting through lots of junk on every level.
> Some people like broken stix for theif fodder.
> I personally don't like dropping them.
Fine, then don't suppress them.
>
> I don't see how this could make the game easier or harder.
> I only see it as less tedious.
That's the point! This idea is *not* supposed to change the
difficulty level, its only purpose is to alleviate some of the tedium
of searching mounds of junk every time you clear out a troll or orc
pit.
>
> Now when it comes to detection. There are serious issues there.
> I think in the detection case one needs to concern with not
> allowing detection to be abused.
By detection, you mean ID'ing stuff? If you restrict this feature to
high level characters, the "advantage" of having things ID'd for you
is negligible. The payoff is the time reduced for having to comb
through piles of stuff looking for good things.
>
> A way around this is to allow all items to show up, until looked at.
> When looked at, it will say, "nothing important here", or when you
> walk over them you get the same message.
> Or possibly if items are in line of sight to do this too.
Why not suppress the creation of these things in the first place?
There's nothing fundamentally different here, I ID the item at
creation, you ID it when they walk over it. End result is the same,
except for the tedious part of getting a half dozen of your messages
when a char walks over a large stack of stuff.
>
> This way significant processing time seems like it could be sured up too.
I don't think so, my routine doesn't seem to slow anything down, I
have a 200Mhz pentium-clone (AMD), so the difference isn't noticeable.
Peter Ellis
>
>I'm talking about suppressing certain types of junk *at creation*
>when the player hasn't ID'd them.
...
>The "problem" here is with weapons and armor. If you require the
>user to ID them first to find out if its something special, then you
>haven't solved the problem everyone is talking about:
...
>By the time a character gets to level 30 and beyond (the exact level is
>debatable, anywhere from 25 to 40?) he has no problem with ID'ing stuff.
> Beyond this point, dealing with masses of junk really becomes
>*tedious*, and that is the thing I'm trying to eliminate.
>
Exactly. So when you reach the point where ID is no longer a problem;
allow items to be generated already pseudo-ID'ed, and give the option to
squelch anything non-{good} or non-{excellent} as desired.
At level 15 (say), you get a pseudo-ID at item creation that tells you if
something's cursed or not. The item's generated with an inscription that
says {cursed}, or not, depending. Now you can filter out cursed items.
At level 25, you get a weak pseudo-ID (that is, {good}, {average} or
{cursed}, like mage pseudo-ID). Now you can filter out all except good
items.
At level 35, you get full pseudo-ID (that is {special}, {excellent},
{good}, {average}, {cursed}, {worthless}, like a warrior's pseudo-ID).
Now you can squelch whatever you like.
All that's required is to have the game pipe items through the appropriate
pseudo-ID filter as they're generated, so that they're dropped already
flagged as {cursed}, or {excellent} or whatever. As the character
advances, the type of pseudo-ID given to items at generation improves.
Concurrently with this, you institute a squelch system so that you can
tell it to automatically destroy inscribed {cursed} items at generation
time, or whatever.
Better make it at generation time, otherwise you get the following...
The <foo> points at you and curses! <more>
You feel a malignant black aura surrounding your Robe of Permanence!
<more>
-- squelching Robe of Permanence -- <more>
At lower levels, the items don't get pseudo-ID'ed, so they don't get
squelched. At higher levels, it's generated with the pseudo-ID
inscription already attached, and can be squelched if the player wants.
> Huh? What's the level got to do with this? A cursed item is junk
>whether its on level 25 or 40,
Yes, but as the _character_ gets to higher levels, they learn to
*recognise* whether an item is cursed or not without having to take the
bother of identifying it. At character level 1, you know nothing. At
level 40, you can tell a crap sword without having to identify it.
>and what is an "instantly-recognizable" item?
An item that's generated with its pseudo-ID inscription already created.
All this suggestion is, is a beefing up of the pseudo-ID system, combined
with a squelch system that works by looking at the pseudo-ID inscriptions.
>Your on the virge of basically asking for
>a non-trivial AI routine to make all these decisions.
No, it *is* trivial in concept at least, since pseudo-ID already exists.
You just say that
"If character level is >25, subject all items to weak pseudo-ID at
creation"
"If character level is >35, subject all items to strong pseudo-ID at
creation"
Then you add an array which registers for each inventory slot whether you
want to kill items inscribed with {cursed}, {normal}, {good}, {excellent}
at creation.
Peter
Ed Cogburn
>
> Ed Cogburn, a beast of pure hatred with purpose malign, wrote:
> > You are describing a really complex implementation, whose value
> > probably doesn't exceed the difficulty in coding it. As for handling
> > each object individually, do you know how many objects there are in
> > the game? In [GSN]'s defines.h, it lists a huge number of SV_<item>
> > defines. I stopped counting at 400... Take away a hundred for the
> > artifacts, and another hundred for good measure, and you're still
> > talking about 200+ items that would have to be dealt with on an
> > individual basis. Can you just imagine the size of the switch
> > statement it would take to deal with all that? :-)
>
> There are 421, 461, and 508 item types in V, O and Z, respectively (not
> counting INSTA_ARTs), but there's no need for a switch. Each item is
> specified uniquely by a tval,sval pair, and the "obvious" interface is a
> two level list of all tvals (potion, wand, armor...) and then all svals
> (water, apple juice...). The work is already done, see the wizard mode
> Create Item command.
The create item code isn't doing the same thing we're talking about,
so suggesting that all the work is already done is a little
premature. :-)
You also forgot to mention that there is a third level below the
first two, because you need to be able to say "let me see good/great
objects of this type, but suppress the mundane versions". So we are
talking about at least a 16 bit flag for every item, which would
require a different interface than the one provided by the wizard
create item code.
The thing that really blows my mind though is would anybody
*seriously* want to go through 200-300 toggle switches, setting things
you want to see and others to be suppressed? Someone talked about
having "sets" that preinitialized all these switches to some setting,
then changing "sets" as your char develops and discovers new needs and
suppressing things that the char is no longer interested in. Which
sounds good, until you ask yourself, if everyone is going to use these
pre-built sets because nobody really wants to set all those flags
manually, then why in the world would you go to all the trouble to
allow for flags on each individual item in the first place?
Breaking things down by their usefulness still sounds better than a
flag for every item. Suppressing items by categories like 'mundane',
'good', and 'great' is more useful then trying to say the same thing
by setting 200 flags, and only requires one flag. Most people would
like to see 'great' items regardless of the specific item, since a
'great' weapon will probably prove desirable regardless of its
specific weapon type. I'll take a Short Sword of Westernessee over a
Battle Axe (+5,+5) any day.
>
> I would suggest a list of all ego items on top of that, to allow for a
> finer grain. And while I don't really see a need for modifications to
> the 'k'ill command (I don't think you'll really spend all that much time
> manipulating the squelch list), it can't really hurt.
Your absolutely right, people aren't going to spend all that much
time setting 200 flags with at least three possible settings for each
flag, in which case you should ask yourself "Why do it that way
then?".
Ed Cogburn
>
> "Andrew Tait" <A.T...@surrc.gla.ac.uk> writes:
>
> > Hi all,
> >
> > I'm replying to something just to get into this thread.
> >
> > Funny how you can just 'see' things in your head. How about this:
> >
> > When you destroy any object (floor or backpack), instead of 'Are you sure
> > (y/n)?' you get an extended menu, say 'Destroy Wand of Light
> > (y/n/d/p/q)?'
>
> how about this? press O. would you like to destroy all worthless
> objects on the entire level?
How about having all worthless items suppressed when the dungeon
level is created, or at the moment when a monster drops something?
>
> that's what i want. i'd just like to be able to clobber everything
> i've identified and shops will not buy in one fell swoop. kind of
> compacting objects upon demand. Z (whatever version i have) seems to
> have a problem compacting objects.
I've already done this, unfortunately its a patch for only [GSN202].
Since [GSN] is a [Z] variant, getting it working with [Z] might be
relatively easy.
>
> i think it's cheating if useless dross isn't created and strewn across
> the floor potentially blocking other items from falling. however, i
> want some way for crap to get wiped out.
Angband goes to major extremes to drop something nearby. That's
really a moot point now that we have stacking in squares. I've
*never* seen the "The <whatever> disappears" message, indicating it
couldn't find a drop spot. I did see it in older versions, around the
2.7.9 mark or earlier, I think. I'd rather it eliminate the junk, so
I only have to deal with potentially useful things on the floor, or
after a monster drop.
>
> --
> J o h a n K u l l s t a m
> [kull...@ne.mediaone.net]
> Don't Fear the Penguin!
--
Johan Kullstam
> Johan Kullstam wrote:
> > i think it's cheating if useless dross isn't created and strewn across
> > the floor potentially blocking other items from falling. however, i
> > want some way for crap to get wiped out.
>
>
> Angband goes to major extremes to drop something nearby. That's
> really a moot point now that we have stacking in squares. I've
> *never* seen the "The <whatever> disappears" message, indicating it
> couldn't find a drop spot. I did see it in older versions, around the
> 2.7.9 mark or earlier, I think. I'd rather it eliminate the junk, so
> I only have to deal with potentially useful things on the floor, or
> after a monster drop.
while the option has been there since angband 2.8.1, i don't use it.
i don't allow stacking in squares. well, i'll allow stuff that will
stack in the inventory to stack on the floor too, but no hetrogenous
stacks. i think it's so against the balance of angband. there should
be a trade-off between fighting a summoner (like some of the wyrms) in
a little corridor and fighting in a large room.
Jonathan Ellis
Johan Kullstam wrote in message ...
>Ed Cogburn <ecog...@greene.xtn.net> writes:
>
>> Johan Kullstam wrote:
>
>> > i think it's cheating if useless dross isn't created and strewn
across
>> > the floor potentially blocking other items from falling. however,
i
>> > want some way for crap to get wiped out.
>>
>> Angband goes to major extremes to drop something nearby. That's
>> really a moot point now that we have stacking in squares. I've
>> *never* seen the "The <whatever> disappears" message, indicating it
>> couldn't find a drop spot.
that's full of traps in a Greater Traps Vault. I have... And I've also
mindwaved a multi-hued dragon pit (or, in fact, *ultrablasted* it with
Psiband's power of that name, which does 100d(level) damage and is
easily capable of taking down an AMHD with one shot. It does, however,
cost 255 mana and is for level 50 characters only.) and seen hundreds of
"Too many objects! Compacting objects..." messages. Objects do still
disappear if there is no place for them to go - apart from artifacts and
the set of stairs dropped by Sauron and Morgoth, which basically
eliminate whatever is underneath them instead. (I've seen the stairs
generated in the place where a wall used to be.)
Jonathan.
Jason Willoughby
>> counting INSTA_ARTs), but there's no need for a switch. Each item is
>> specified uniquely by a tval,sval pair, and the "obvious" interface is a
>> two level list of all tvals (potion, wand, armor...) and then all svals
>> (water, apple juice...). The work is already done, see the wizard mode
>> Create Item command.
> The create item code isn't doing the same thing we're talking about,
> so suggesting that all the work is already done is a little
> premature. :-)
Obviously it can't be used as is, but the hard part is giving the user a
decent way of specifying a particular k_idx. It's also what people keep
complaining about, "You mean I have to go through a list of 500 items to
squelch something! What a waste of time!" You don't. You only have to
go through two lists of <40.
> You also forgot to mention that there is a third level below the
> first two, because you need to be able to say "let me see good/great
> objects of this type, but suppress the mundane versions".
Ego items was the subject of the very next sentence after that bit you
quoted, and by the time this is useful, the difference betwen
cursed/average/good is negligible.
> So we are talking about at least a 16 bit flag for every item, which
> would require a different interface than the one provided by the
> wizard create item code.
It needs one bool on each object_kind and ego_item_type. That's nothing
more than a checkbox.
> Which sounds good, until you ask yourself, if everyone is going to use
> these pre-built sets because nobody really wants to set all those
> flags manually, then why in the world would you go to all the trouble
> to allow for flags on each individual item in the first place?
Because you can never give the user too much control.
> Breaking things down by their usefulness still sounds better than a
> flag for every item. Suppressing items by categories like 'mundane',
> 'good', and 'great' is more useful then trying to say the same thing
> by setting 200 flags, and only requires one flag.
But those catagories are huge! "Mundane" covers everything from Rings
of Searching to Rings of Damage, "good" goes from Robes [0,+1] to Power
Dragon Scale Mail, and "great" covers Boots of Feathering Falling and
Boots of Speed.
A squelch mechanism like that is practically useless. I, for one,
always want to see the high-end of each catagory.
Jason Willoughby
> while the option has been there since angband 2.8.1, i don't use it.
> i don't allow stacking in squares. well, i'll allow stuff that will
> stack in the inventory to stack on the floor too, but no hetrogenous
> stacks. i think it's so against the balance of angband. there should
> be a trade-off between fighting a summoner (like some of the wyrms) in
> a little corridor and fighting in a large room.
This, like preserve mode, is one of my major peeves with modern Angband.
I used to love having to draw Wyrms into the middle of a room, and
there's nothing cheezier than an anti-summoning corridor. But I think
we more or less have to accept that stacking is here to stay.
Not that I'm not bursting with compromises. For instance, how if items
will only stack with items dropped by another monster? You'll still
only get 5 items off a D in a hallway, but orc pits won't be nearly so
tedious. I can see a way to do this with squelching, too... But it
means a complete rewrite of the item placement code.
Andrew Tait
> Johan Kullstam wrote:
> >
> > "Andrew Tait" <A.T...@surrc.gla.ac.uk> writes:
> >
> > > Hi all,
> > >
> > > I'm replying to something just to get into this thread.
> > >
> > > Funny how you can just 'see' things in your head. How about this:
> > >
> > > When you destroy any object (floor or backpack), instead of 'Are you sure
> > > (y/n)?' you get an extended menu, say 'Destroy Wand of Light
> > > (y/n/d/p/q)?'
> >
> > how about this? press O. would you like to destroy all worthless
> > objects on the entire level?
>
>
> How about having all worthless items suppressed when the dungeon
> level is created, or at the moment when a monster drops something?
There's a basic difference between an autodestroy system and
suppression. With the autodestroy, the item gets generated and
destroyed (by PLAYER), so the no. of game turns required to sift
through the drop from an Orc Pit is identical, but you don't need to
'knnnyy' your way to distraction all the way through the game.
Suppressing bad items may make the game easier, but it may knock the
balance off kilter a bit, especially if you can control the
suppression of un-ID'd weapon and armour (eg a Dagger of Morgul is
suppressed as it is created, rather than when you ID/pseudo-ID it)
> > i think it's cheating if useless dross isn't created and strewn across
> > the floor potentially blocking other items from falling. however, i
> > want some way for crap to get wiped out.
I agree completely. Perhaps an autodestroying scroll of some sort?
After all, OoD 'knows' about bad items before they're ID'd by the
player, so it would be within the games limits to have a Staff/Rod of
Destroy Junk, and a foo of Destroy Average Item for later.
> Angband goes to major extremes to drop something nearby. That's
> really a moot point now that we have stacking in squares. I've
> *never* seen the "The <whatever> disappears" message, indicating it
> couldn't find a drop spot. I did see it in older versions, around the
> 2.7.9 mark or earlier, I think. I'd rather it eliminate the junk, so
> I only have to deal with potentially useful things on the floor, or
> after a monster drop.
It would be nice, but IMO too Munchkinish. A bit like having a
keypress to destroy all Giant Fleas once you get past the point where
they go from dangerous to nuisance. The game has good stuff and bad
stuff, but I think that issues like dropped junk have to be dealt with
through the user interface, and not through game modification.
YM, of course, MV ;)
Totty <8^)
James W Sager Iii
Way to Get Rid .. by Ed Cog...@greene.xtn.ne
> really a moot point now that we have stacking in squares. I've
> *never* seen the "The <whatever> disappears" message, indicating it
Unless you go out of your way in a place with doors filling the room,
doesn't disappear.
Ed Cogburn
The pseudo-ID you are talking about already exists. Its the "easy
sense" option in GSN, and probably other variants. It tries to sense
all objects on the level at creation time, and marks them with an
inscription.
I won't argue your basic point, a graduated approach would be better.
I still can't get why everyone wants to make this harder than it
needs to be, though. The thing about using your inventory to
auto-destroy things is possibly dangerous as your example above, but
worst of all, it forces you to do the things we're trying to
eliminate: pick up stuff, if ID'd then destroy bad stuff, else wait
for player to ID the stuff himself, then destroy the bad stuff. Why
not destroy stuff at creation, so the player doesn't have to deal with
any of this.
You know, an extension of the "easy sense" code could do something
similar to what you want (and what I want). When the code goes to ID
everything on the level, have it auto-destroy bad stuff which were
sensed. Remember, the "easy sense" thing is graduated, it gets better
as you improve your character, so there isn't really an advantage to a
low-level character. I like this idea..... [wanders off looking for
the file with the "easy sense" code in it....]
Ed Cogburn
>
> Johan Kullstam wrote in message ...
> >Ed Cogburn <ecog...@greene.xtn.net> writes:
> >
> >> Johan Kullstam wrote:
> >
> across
> >> > the floor potentially blocking other items from falling. however,
> i
> >> > want some way for crap to get wiped out.
> >>
> >> really a moot point now that we have stacking in squares. I've
> >> *never* seen the "The <whatever> disappears" message, indicating it
> that's full of traps in a Greater Traps Vault. I have... And I've also
> mindwaved a multi-hued dragon pit (or, in fact, *ultrablasted* it with
> Psiband's power of that name, which does 100d(level) damage and is
> easily capable of taking down an AMHD with one shot. It does, however,
> cost 255 mana and is for level 50 characters only.) and seen hundreds of
> "Too many objects! Compacting objects..." messages. Objects do still
> disappear if there is no place for them to go - apart from artifacts and
> the set of stairs dropped by Sauron and Morgoth, which basically
> eliminate whatever is underneath them instead. (I've seen the stairs
> generated in the place where a wall used to be.)
>
> Jonathan.
I've obviously not gone deep enough to see that kind of situation.
Which *band were you playing? I'm going by my experience with
orc/troll pits for example, where junk is piled up 6-7 deep in squares
around the entrance to the pit, yet I never saw the disappearing item
message. I've played [V] and [O] deep (not as deep as you though),
and [GSN] shallow.
Ed Cogburn
>
> Ed Cogburn wrote the following about Re: [All] A Way to Get Rid of Garbage Items?:
>
> > Johan Kullstam wrote:
> > >
> > > "Andrew Tait" <A.T...@surrc.gla.ac.uk> writes:
> > >
> > > > Hi all,
> > > >
> > > > I'm replying to something just to get into this thread.
> > > >
> > > > Funny how you can just 'see' things in your head. How about this:
> > > >
> > > > When you destroy any object (floor or backpack), instead of 'Are you sure
> > > > (y/n)?' you get an extended menu, say 'Destroy Wand of Light
> > > > (y/n/d/p/q)?'
> > >
> > > how about this? press O. would you like to destroy all worthless
> > > objects on the entire level?
> >
> >
> > How about having all worthless items suppressed when the dungeon
> > level is created, or at the moment when a monster drops something?
>
> There's a basic difference between an autodestroy system and
> suppression. With the autodestroy, the item gets generated and
> destroyed (by PLAYER), so the no. of game turns required to sift
> through the drop from an Orc Pit is identical, but you don't need to
> 'knnnyy' your way to distraction all the way through the game.
>
> Suppressing bad items may make the game easier, but it may knock the
> balance off kilter a bit, especially if you can control the
> suppression of un-ID'd weapon and armour (eg a Dagger of Morgul is
> suppressed as it is created, rather than when you ID/pseudo-ID it)
>
> > > the floor potentially blocking other items from falling. however, i
> > > want some way for crap to get wiped out.
>
> After all, OoD 'knows' about bad items before they're ID'd by the
> player, so it would be within the games limits to have a Staff/Rod of
> Destroy Junk, and a foo of Destroy Average Item for later.
>
> > really a moot point now that we have stacking in squares. I've
> > *never* seen the "The <whatever> disappears" message, indicating it
> > 2.7.9 mark or earlier, I think. I'd rather it eliminate the junk, so
> > I only have to deal with potentially useful things on the floor, or
> > after a monster drop.
>
> It would be nice, but IMO too Munchkinish. A bit like having a
> keypress to destroy all Giant Fleas once you get past the point where
> they go from dangerous to nuisance. The game has good stuff and bad
> stuff, but I think that issues like dropped junk have to be dealt with
> through the user interface, and not through game modification.
>
> YM, of course, MV ;)
>
> Totty <8^)
Actually, I've got an idea better than mine. How about extending the
"easy sense" feature so junk you successfully ID'd is auto-destroyed?
No advantage here, if you can't ID it when created, it doesn't get
destroyed. "easy sense" works for both stuff at level creation and
monster drops.
I guess we're not likely to agree about this. I would never consider
an auto destroy for monsters, since for Fleas, they are supposed to be
a nuisance. Its when you meet them with other monsters where their
abilities create a situation far more dangerous then facing them, or
those other monsters, alone. Reducing the amount of junk being
generated is not, to me, either an advantage of some kind, or
munchkinish. As you said though, YMMV.
Peter Ellis
>
>Peter Ellis wrote:
>>
>> Concurrently with this, you institute a squelch system so that you can
>> tell it to automatically destroy inscribed {cursed} items at generation
>> time, or whatever.
>>
>
>sense" option in GSN, and probably other variants.
>
I don't think this option is in Z, and certainly not in Vanilla.
>
>Why
>not destroy stuff at creation, so the player doesn't have to deal with
>any of this.
>
I point you to the quoted section of my post above...
> You know, an extension of the "easy sense" code could do something
>similar to what you want (and what I want).
That's *exactly* what I was suggesting. Having an array to store the
squelch setting for each equipment slot also makes sense, since that
allows squelching of useless light items once you have the Phial, etc.
etc.
Add another couple of elements to the array for junk like pottery/spikes,
one element for food items, one for missiles, and this eliminates 90% of
the hassle. Wands, rods, potions etc. are too diverse to squelch, so leave
those as-is. And for those who disapprove of the option, they can
leave it unset. Perhaps a notice on the main character screen "you are
squelching unwanted items" to alert readers of character dumps.
>When the code goes to ID
>everything on the level, have it auto-destroy bad stuff which was
>sensed. Remember, the "easy sense" thing is graduated, it gets better
>as you improve your character, so there isn't really an advantage to a
>low-level character. I like this idea.....
Ditto. I don't know if the current "easy-sense" implementation is
graduated, but it seems as if it would be simple to make it so, though I'm
not a coder.
Peter
James W Sager Iii
Way to Get Rid .. by Ed Cog...@greene.xtn.ne
ROFLMAO
Thats a quote if I ever saw one.
James W Sager Iii
Way to Get Rid .. by Ed Cog...@greene.xtn.ne
> an auto destroy for monsters, since for Fleas, they are supposed to be
> a nuisance.
square' ability implemented some how. When things are so weak he just
attax and pushes past. Maybe if holding shift while moving, if you have
extra attacks left, you move onto the square. Or something.
Andrew Tait
Ed Cogburn <ecog...@greene.xtn.net> wrote in message
news:384E4986...@greene.xtn.net...
> Andrew Tait wrote:
> Actually, I've got an idea better than mine. How about extending the
> "easy sense" feature so junk you successfully ID'd is auto-destroyed?
> No advantage here, if you can't ID it when created, it doesn't get
> destroyed. "easy sense" works for both stuff at level creation and
> monster drops.
This sounds good. And I do like the 'better idea than mine' bit :) There's
nothing like superceding yourself. I'm trying to think about this in terms
of what will appear in the game....
Hmmm. I somehow get back to the problem of /defining/ what the player
perceives as junk. If we can decide on this, there could be a couple of
different options in "Disturbance" eg "Destroy all junk on creation"
"Destroy all junk under player" (My favourite because it only affects stuff
you come in contact with) or "Unmunchkinish destroy"
These go in order of least to most 'Angbandness', IMO.
"Destroy on creation" means the stuff never clogs up the level, in theory
meaning the arrays of objects on the floor are kept small and tidy with all
the good stuff. "Destroy under player" means that when you walk onto a
stack (or single object), all the grotty stuff is gone before you see it.
And the last option actually uses game turns to destroy, a bit like turning
your player into a destroyborg when he walks onto an object or stack.
Options. Make everyone happy :)
I still think the problem is /defining/ what's junkable. There's obviously
the 'PoWeakness, Broken Skull " etc. stuff, but once I'm a Level 50
Death/Chaos Dwarf Priest, I'm destroying Wands of Stinking Cloud and Daggers
of Slay Dragon.
> I guess we're not likely to agree about this. I would never consider
> an auto destroy for monsters, since for Fleas, they are supposed to be
> a nuisance. Its when you meet them with other monsters where their
> abilities create a situation far more dangerous then facing them, or
> those other monsters, alone. Reducing the amount of junk being
> generated is not, to me, either an advantage of some kind, or
> munchkinish. As you said though, YMMV.
Well, it has to be a small advantage. Think of yer average Orc Pit.
Usually you're hitting these with Level 12+ Warriors or 20+ Mages, so you're
not invinsible by any stretch of the imagination. And how much stuff from
an Orc Pit is good? About 10%? My strategy is usually to pick things up as
they drop (using auto-pickup with query), and then pseudo-ID kicks in when
I'm sneaking away for a quick 'R'est.
D'oh! (Twice today??) You said "successfully ID'd is auto-destroyed".
Yes! You're ideas fit was I was thinking. Adding the ability to flag
certain objects as undesirable (eg Scrolls of Light) would make the whole
trip more enjoyable. No munchkinism, as _in theory_, you could have left
the objects where they were without touching them (0 game turns)
I need a break....
Totty <8^)
Jonathan Ellis
Ed Cogburn wrote in message <384E46D6...@greene.xtn.net>...
>Jonathan Ellis wrote:
>>
>> Johan Kullstam wrote in message ...
>> >Ed Cogburn <ecog...@greene.xtn.net> writes:
>> >
>> >> Johan Kullstam wrote:
>> >
>> across
>> >> > the floor potentially blocking other items from falling.
however,
>> i
>> >> > want some way for crap to get wiped out.
>> >>
>> >> really a moot point now that we have stacking in squares. I've
>> >> *never* seen the "The <whatever> disappears" message, indicating
it
>> >> couldn't find a drop spot.
>> that's full of traps in a Greater Traps Vault. I have... And I've
also
>> mindwaved a multi-hued dragon pit (or, in fact, *ultrablasted* it
with
>> Psiband's power of that name, which does 100d(level) damage and is
>> easily capable of taking down an AMHD with one shot. It does,
however,
>> cost 255 mana and is for level 50 characters only.) and seen hundreds
of
>> "Too many objects! Compacting objects..." messages. Objects do still
>> disappear if there is no place for them to go - apart from artifacts
and
>> the set of stairs dropped by Sauron and Morgoth, which basically
>> eliminate whatever is underneath them instead. (I've seen the stairs
>> generated in the place where a wall used to be.)
>>
>> Jonathan.
>
>
> I've obviously not gone deep enough to see that kind of situation.
>Which *band were you playing? I'm going by my experience with
>orc/troll pits for example, where junk is piled up 6-7 deep in squares
>around the entrance to the pit, yet I never saw the disappearing item
>message. I've played [V] and [O] deep (not as deep as you though),
>and [GSN] shallow.
Vaults, of which there are only four in vanilla and millions of
different types in ZAngband. One of these Greater Vaults has two very
large areas filled *entirely* with traps, and no item can be generated
on top of a trap (or a door, open or closed.) Also, items generated when
a monster dies have to be with a certain distance of the monster in
question (I think it's radius 2.) So if you shoot down a monster right
in the middle of this massive triangular area filled entirely with
traps, its drop will disappear. (I think artifacts will appear, though,
and cancel the trap underneath, but I'm not sure: for whatever reason,
you certainly can't get any "The Long Sword (4d5) disappears" messages
here.)
You will never lose any of the junk from orc and troll pits as all
of the floor squares are normal and not trapped.
Jonathan.
John R. Chapman
On 6 Dec 1999, Andrew Tait wrote:
*SNICKER-SNACK*
> It would be nice, but IMO too Munchkinish. A bit like having a
> keypress to destroy all Giant Fleas once you get past the point where
> they go from dangerous to nuisance. The game has good stuff and bad
> stuff, but I think that issues like dropped junk have to be dealt with
> through the user interface, and not through game modification.
But I DO have a keypress to destroy all Giant Fleas. (Gotta love Genocide.)
-john
Ed Cogburn
>
> On Wed, 8 Dec 1999, Ed Cogburn wrote:
> >
> >Peter Ellis wrote:
> >>
> >> Concurrently with this, you institute a squelch system so that you can
> >> tell it to automatically destroy inscribed {cursed} items at generation
> >> time, or whatever.
> >
> > The pseudo-ID you are talking about already exists. Its the "easy
> >sense" option in GSN, and probably other variants.
> >
>
> I don't think this option is in Z, and certainly not in Vanilla.
Well, there is a whole *bunch* of stuff thats not in [V], and may
never be. :-)
Alas, I've looked, and 'easy sense' only works when the char (L)ooks
at an item, or picks it up. My assumption it was working at level
creation time or monster drop time is wrong.
>
> >
> >Why
> >not destroy stuff at creation, so the player doesn't have to deal with
> >any of this.
> >
>
> I point you to the quoted section of my post above...
Ok, sorry I missed that. Since 'easy sense' is out, some kind of
auto-sense-at-item-creation-time is needed. If the ID succeeds, bad
items are auto-destroyed, else the char must ID the item manually.
This is better than what I've tried to do, since this doesn't give the
player a 'free' ID, but requires much more work, and is likely to end
up being a third ID method (original, easy sense,
easy-sense-at-item-creation) .... [sigh]
>
> > You know, an extension of the "easy sense" code could do something
> >similar to what you want (and what I want).
>
> That's *exactly* what I was suggesting. Having an array to store the
> squelch setting for each equipment slot also makes sense, since that
> allows squelching of useless light items once you have the Phial, etc.
> etc.
My thing can already do that, since lite items, *besides artifacts*,
don't require an ID to see what they are, just look for TV_LITE and
TV_FLASK. Artifacts *including the bad ones* are never auto-ID'd or
auto-destroyed.
The only problem is how weapons and armor are handled. Once a potion
has been ID'd its true nature is revealed for the rest of the game,
but weapons/armor must be ID'd every time to see if its
cursed/normal/good/great/artifact. A version of easy sense could be
used here, where an attempt at ID is made at creation time, using easy
sense's method, which relies on the char's characteristics, and if
succeeds, the item is auto-destroyed if bad, or is selected by the
player to be suppressed.
Your reference to equipment slots sounds like you talking about
handling items by category, which I do also, except I haven't defined
nearly as many flags as you would have. Mainly because a) I'm lazy,
and b) it would need a brand new method of selecting this stuff in a
multi-window selection screen which currently doesn't exist. Right
now, the limit is about 20, or the number of rows on the display minus
the header at the top.
I may break down and take a shot at a multi-window selection display
with many dozens of flags to be set, but for now, I'm going to see
what can be done about an 'easy sense at creation time'.
>
> Add another couple of elements to the array for junk like pottery/spikes,
Done that too.
> one element for food items, one for missiles, and this eliminates 90% of
I have two options for food, conventional and good (waybread, cakes,
and good mushrooms).
I have three options for missiles, whether you want shot/arrow/bolts,
but currently no option to specify what type you are interested in,
eg. normal/good/great. With more flag space this could be done.
> the hassle. Wands, rods, potions etc. are too diverse to squelch, so leave
Certainly not. I'm tired of having to sift through wands of heal
monster and rods of haste monster, etc. Right now these types of
items are useless, but in the future that may change. For example, in
NetHack you can aim a wand at yourself, which would make a wand of
heal monster into a wand of healing. For all I know this may have
happened in some variants already. I need to have a look and see if
GSN has done anything like that. For now though, they remain
useless. Anyway, once 'bad' scrolls/potions/wands/staffs/rods are
ID'd for the first time, they are suppressed for the rest of the game.
> those as-is. And for those who disapprove of the option, they can
> leave it unset. Perhaps a notice on the main character screen "you are
> squelching unwanted items" to alert readers of character dumps.
>
> >When the code goes to ID
> >everything on the level, have it auto-destroy bad stuff which was
> >sensed. Remember, the "easy sense" thing is graduated, it gets better
> >as you improve your character, so there isn't really an advantage to a
> >low-level character. I like this idea.....
>
> Ditto. I don't know if the current "easy-sense" implementation is
> graduated, but it seems as if it would be simple to make it so, though I'm
> not a coder.
I agree with your sentiment that it should be graduated, and not an
arbitrary option like what I was doing when I said the procedure would
be available to a character of a given level. I never liked that, but
the alternative was ugly. Now though, with the easy sense code to
borrow or just use as a template, a more realistic and graduated
response is possible.
>
> Peter
Richard Henry
<koderscu@nospam_poczta.onet.pl> forced innocent electrons to deliver
this message:
>Hi all,
>
>I was lurking for a while, but seeing all those ideas flying around the NG,
>I tought it's about time to contribute.
>
>Why don't create something like 'rod/wand/staff of useless items destruction'?
>
>It would zap for a ball which would pseudo-id all those stuff lying around
>the dungeon, and destroyed the crap (let's say anything below {good}).
>You don't have to modify the savefile, or create any external files, you don't
>abuse the balance of the game, and you don't have to do any extra menus.
>
>What do you guys think?
I've wondered about something like that myself. Hadn't suggested it
because I'm still unsure if it's a good idea or how it could best be
implemented.
Actually, when playing with floor stacking on, I'd be happy with a
*destroy stack* option, so once I removed the good stuff, I could kill
a whole stack of junk in one go.
Then again, playing with mixed items stacking on the floor still feels
a bit like cheating. Especially when using an anti-summoning corridor.
--
R. Dan Henry (danh...@inreach.com)
http://home.inreach.com/danhenry/enter.html
Ed Cogburn
>
> Ed Cogburn, a beast of pure hatred with purpose malign, wrote:
> > Jason Willoughby wrote:
> >> counting INSTA_ARTs), but there's no need for a switch. Each item is
> >> specified uniquely by a tval,sval pair, and the "obvious" interface is a
> >> two level list of all tvals (potion, wand, armor...) and then all svals
> >> (water, apple juice...). The work is already done, see the wizard mode
> >> Create Item command.
>
> > The create item code isn't doing the same thing we're talking about,
> > so suggesting that all the work is already done is a little
> > premature. :-)
>
> Obviously it can't be used as is, but the hard part is giving the user a
> decent way of specifying a particular k_idx. It's also what people keep
> complaining about, "You mean I have to go through a list of 500 items to
> squelch something! What a waste of time!" You don't. You only have to
> go through two lists of <40.
I don't understand this. In [GSN] there are ~560 entries in
k_info.txt. I don't see how using k_info reduces the number of things
to be dealt with. If you insist on being able to control every single
item, then you'll inevitably need more than one flag for each item,
all of which adds up quickly. Working by categories, such as
mundane/good/great for weapons/armor and bad/good for potions as well
as based on TV_<name> instead of each individual item via SV_<name>,
seems the most practical.
>
> > You also forgot to mention that there is a third level below the
> > first two, because you need to be able to say "let me see good/great
> > objects of this type, but suppress the mundane versions".
>
> Ego items was the subject of the very next sentence after that bit you
> quoted, and by the time this is useful, the difference betwen
> cursed/average/good is negligible.
Yes, agreed. By the time my routine would kick in, the category of
mundane weapons/armor would already be pointless to the player. What
it looks like now, is the need to develop an 'easy sense' clone
routine which acts on items at creation time, and destroys items known
by the player via 'easy sense' commands, ie. sense_object(), to be
bad/worthless. Those items which fail the sense check would still
have to be dealt with manually, but the sense ability increases with
player class, experience, and INT & WIS, so the player's ability to
avoid junk increases over time. All this, without the need to give
the player a free ID on weapons/armor.
>
> > So we are talking about at least a 16 bit flag for every item, which
> > would require a different interface than the one provided by the
> > wizard create item code.
>
> It needs one bool on each object_kind and ego_item_type. That's nothing
> more than a checkbox.
So, ~560 bools on k_info entries? How many screen pages would it
take to present all these flags to the player for setting? And if you
want to differentiate between mundane/good/great for weapons/armor,
you'll need at least 2 more flags *per weapon/armor item*.
No need for ego_item_type, just use ego_item_p(o_ptr) to separate ego
items from everything else.
>
> > Which sounds good, until you ask yourself, if everyone is going to use
> > these pre-built sets because nobody really wants to set all those
> > flags manually, then why in the world would you go to all the trouble
> > to allow for flags on each individual item in the first place?
>
> Because you can never give the user too much control.
And the flipside of this statement is drowning the player in
minutiae. There's nothing wrong with giving the player control, just
don't beat him over the head with it. :-) How many players would
like having to manage 560 flags? Heck, how many of them would want to
manage just 75 flags? We need to keep this as simple as possible,
while remaining as flexible as possible to the player. If we make
this thing complex, we run the real risk of discovering that no one
will code it. Better something than nothing at all.
>
> > Breaking things down by their usefulness still sounds better than a
> > flag for every item. Suppressing items by categories like 'mundane',
> > 'good', and 'great' is more useful then trying to say the same thing
> > by setting 200 flags, and only requires one flag.
>
> But those catagories are huge! "Mundane" covers everything from Rings
> of Searching to Rings of Damage, "good" goes from Robes [0,+1] to Power
> Dragon Scale Mail, and "great" covers Boots of Feathering Falling and
> Boots of Speed.
I'm using the "language" of Angband when I say this. Well, mundane
is my word for average/normal. First, Worthless/Mundane/Good/Great
refers *only* to weapons and armor:
Worthless == cursed/broken.
Mundane == an item with no plusses, ie. Dagger (+0,+0).
Good == an item with plusses, ie. Dagger (+5,+10).
Great == ego/artifact.
No ring other than adornment is "non-magical". They are all
"magical", thus the word mundane isn't used here. So rings are
divided into good/bad.
In my work, I put DSMs in their own category.
Your right about the boots thing and 'great'. If I forced myself to
add to the number of flags involved, I could separate the most
important items of each category into their own special class, like
speed, resistance, or westernesse weapons/armor, and speed, or damage
for rings, etc. Hmmmm, we could also use the price of the item to
determine an 'exception' to the auto-destroy rule: If xyz is worth >
250,000, don't suppress it....?
>
> A squelch mechanism like that is practically useless. I, for one,
Considering how much it already suppresses on a normal level, I'd say
its far from useless. I've seen it suppress 15+ items during creation
of a deep level, and it works for monster drops the same way
> always want to see the high-end of each catagory.
I agree about this. I'm still trying to get a simpler version
working first before deciding how to improve it. So far, I've limited
myself to flags that I can get on one screen, single column, ie.
"label: flag", because that was the simplest thing to do, but I'm
starting to think I should have gone ahead and developed a
multi-screen flag selection display. This is something Angband lacks,
which means I couldn't 'borrow' the code from somewhere else, it would
have to be original. And I'm a very lazy person. :-)
Ed Cogburn
>
> Way to Get Rid .. by Ed Cog...@greene.xtn.ne
>
>
> Thats a quote if I ever saw one.
Now that you mention it, it is rather, how shall we say, an
'interesting' statement to make. :-)
Jason Willoughby
> used here, where an attempt at ID is made at creation time, using easy
> sense's method, which relies on the char's characteristics, and if
> succeeds, the item is auto-destroyed if bad, or is selected by the
> player to be suppressed.
My preference here is a Kam-style Identify ball spell/rod/staff. I'd
rather extend the current system than see entirely new mechanisms like
easy_sense.
But these are gameplay decisions, and better left to the variant
maintainers, rather than patch authors.
Ed Cogburn
>
> Ed Cogburn <ecog...@greene.xtn.net> wrote in message
> news:384E4986...@greene.xtn.net...
> > Andrew Tait wrote:
>
> > Actually, I've got an idea better than mine. How about extending the
> > "easy sense" feature so junk you successfully ID'd is auto-destroyed?
> > No advantage here, if you can't ID it when created, it doesn't get
> > destroyed. "easy sense" works for both stuff at level creation and
> > monster drops.
>
> This sounds good. And I do like the 'better idea than mine' bit :) There's
The evidence is overwhelming, I must admit: I actually did say
that. Bizarre isn't it? :-)
> nothing like superceding yourself. I'm trying to think about this in terms
> of what will appear in the game....
>
> Hmmm. I somehow get back to the problem of /defining/ what the player
> perceives as junk. If we can decide on this, there could be a couple of
> different options in "Disturbance" eg "Destroy all junk on creation"
> "Destroy all junk under player" (My favourite because it only affects stuff
> you come in contact with) or "Unmunchkinish destroy"
>
> These go in order of least to most 'Angbandness', IMO.
>
> "Destroy on creation" means the stuff never clogs up the level, in theory
> meaning the arrays of objects on the floor are kept small and tidy with all
> the good stuff. "Destroy under player" means that when you walk onto a
> stack (or single object), all the grotty stuff is gone before you see it.
> And the last option actually uses game turns to destroy, a bit like turning
> your player into a destroyborg when he walks onto an object or stack.
>
> Options. Make everyone happy :)
First, I was wrong in my assumption about 'easy sense', it doesn't
work at creation time, it works when you (L)ook at an item or walk
over it, but it shouldn't be too hard to borrow the easy sense
functions and use them at creation time.
"Destroy under player" is certainly possible, since 'easy sense'
essentially does that now. You could simply extend 'easy sense' to
become 'easy sense and destroy worthless'.
The only problem is will people ever get used to the idea of items
disappearing when they walk over it? Or how about items *visually*
disappearing when the (L)ook at it? I don't think this would fly too
well; it would be a clear reminder that they are playing a game with a
cheesy way of suppressing junk. And no cute message will help here.
It would ruin the atmosphere, it seems to me, but this is probably an
individual thing. Destroying junk at creation time is silent; the
player never knows its happening (unless he turns on debug mode).
>
> I still think the problem is /defining/ what's junkable. There's obviously
> the 'PoWeakness, Broken Skull " etc. stuff, but once I'm a Level 50
> Death/Chaos Dwarf Priest, I'm destroying Wands of Stinking Cloud and Daggers
> of Slay Dragon.
>
> > I guess we're not likely to agree about this. I would never consider
> > an auto destroy for monsters, since for Fleas, they are supposed to be
> > a nuisance. Its when you meet them with other monsters where their
> > abilities create a situation far more dangerous then facing them, or
> > those other monsters, alone. Reducing the amount of junk being
> > generated is not, to me, either an advantage of some kind, or
> > munchkinish. As you said though, YMMV.
>
> Well, it has to be a small advantage. Think of yer average Orc Pit.
> Usually you're hitting these with Level 12+ Warriors or 20+ Mages, so you're
> not invinsible by any stretch of the imagination. And how much stuff from
> an Orc Pit is good? About 10%? My strategy is usually to pick things up as
> they drop (using auto-pickup with query), and then pseudo-ID kicks in when
> I'm sneaking away for a quick 'R'est.
>
> D'oh! (Twice today??) You said "successfully ID'd is auto-destroyed".
This is only an issue with weapons/armor which always requires an ID
to see if they are good, bad, or ugly (any Clint E. fans? :-) ). With
a potion for example you only ID it once, you know it's clones from
then on. Or wands, where you only need a second ID to see how many
charges it has, which is basically irrelevant to what we're talking
about. If you took out the part of my thing which ID's weapons/armor
for the (high-level) player, the rest is not only easy to code for,
but perfectly reasonable and non-munchkinish.
>
> Yes! You're ideas fit was I was thinking. Adding the ability to flag
> certain objects as undesirable (eg Scrolls of Light) would make the whole
> trip more enjoyable. No munchkinism, as _in theory_, you could have left
> the objects where they were without touching them (0 game turns)
Right, definitely non-munchkinish here. The problem, as I've been
talking to Jason about, is how many flags should we setup for the
player to deal with all the possibilities people might want. Do we
make a flag for *every* single item type in the game? Or do we go for
general categories (like "mundane" weapons), with a number of
exceptions to deal with things a lot of people would like to have
control over (like treating a RoS separate from all the other lesser
rings).
My thing, as it stands now, doesn't give you the option of
suppressing a SoL specifically, or suppressing all rings except RoS
and possibly RoD, so .... should it?
My idea about this is the general category thing will work, with a
number of automatic exceptions. Rings of Speed, Westernesse weapons
(the best ego item and arguably better than many artifact weapons),
Rings of Damage, Boots of Speed, armor of Resistance, would never be
suppressed, unless its cursed, in the case of a RoS or RoD.
>
> I need a break....
So do I! I've spent more time in the last 2-3 weeks tinkering with
Angband than playing it.
>
> Totty <8^)
Ed Cogburn
Oh, come on now, is floor stacking really so bad? I always wondered
why a scroll took up just as much space as a suit of armor or a chest
on the floor. The stacking limitation was a programmatic one, not one
which should be defended as somehow making sense. It was wrong, and
eventually someone with the skill to fix it, did.
William Tanksley
>Richard Henry wrote:
[a spell/scroll to zap items and only destroy the dumb ones...]
>> I've wondered about something like that myself. Hadn't suggested it
>> because I'm still unsure if it's a good idea or how it could best be
>> implemented.
Hmm... It is an unusual function, at that. I can easily see how to
implement it, but it's not clear how to justify it. It's a pretty raw
exposure of internal mechanics.
>> Actually, when playing with floor stacking on, I'd be happy with a
>> *destroy stack* option, so once I removed the good stuff, I could kill
>> a whole stack of junk in one go.
That sure makes a lot of sense.
>> Then again, playing with mixed items stacking on the floor still feels
>> a bit like cheating. Especially when using an anti-summoning corridor.
> Oh, come on now, is floor stacking really so bad?
BTW, Ed, do you realize how confrontational you sound? I can tell here
that you don't mean anything bad, but you consistently sound like you
expect the person you're discussing an issue with to throw everything down
and start swinging at you ;-).
He didn't say it was bad; he just said that he didn't feel it fit in quite
yet. An appropriate concern when discussing a newly proposed spell which
fits in so poorly.
>I always wondered
>why a scroll took up just as much space as a suit of armor or a chest
>on the floor. The stacking limitation was a programmatic one, not one
>which should be defended as somehow making sense.
You're totally right.
>It was wrong, and eventually someone with the skill to fix it, did.
No, it wasn't ever wrong, any more than the rule that a chess knight can
only follow an L is wrong. Nor has the person with the skill fixed it
yet; all we have is a quick hack which allows more reasonable behavior.
The problem, as Richard notes, is that we have a partial fix: the feature
we wanted (stacking) is added, and it has proved to be fun and useful.
However, the feature doesn't fit yet with the rest of the game; he feels
the misfit especially badly in anti-summoning corridors, where the former
cost of the tactic is no longer applicable.
I'm not sure what the solution is, though. Perhaps the only solution is
to make the AI clever enough to avoid anti-summoning corridors in general
(although if you're clever enough you still ought to be able to trap
summoners). That would be hard, because it would also require rebalancing
all summoning so that AS corridors weren't needed.
Oh well :-). I need to get back to my project.
>Ed C.
--
-William "Billy" Tanksley, in hoc signo hack
James W Sager Iii
> and cancel the trap underneath, but I'm not sure: for whatever reason,
Artifact teleports to someplace else in the dungeon.
James W Sager Iii
Way to Get Rid .. by William Tanksley@hawking
>
> BTW, Ed, do you realize how confrontational you sound? I can tell here
> that you don't mean anything bad, but you consistently sound like you
> expect the person you're discussing an issue with to throw everything down
> and start swinging at you ;-).
Maybe if you were a good enough Angband player, you'd realize where he
is coming from. ^^
Johan Kullstam
> Richard Henry wrote:
> >
> > On Wed, 08 Dec 1999 06:08:23 GMT, the notorious Mariusz Chwalba
> > <koderscu@nospam_poczta.onet.pl> forced innocent electrons to deliver
> > this message:
> >
> > >Hi all,
> > >
> > >I was lurking for a while, but seeing all those ideas flying around the NG,
> > >I tought it's about time to contribute.
> > >
> > >Why don't create something like 'rod/wand/staff of useless items destruction'?
> > >
> > >It would zap for a ball which would pseudo-id all those stuff lying around
> > >the dungeon, and destroyed the crap (let's say anything below {good}).
> > >You don't have to modify the savefile, or create any external files, you don't
> > >abuse the balance of the game, and you don't have to do any extra menus.
> > >
> > >What do you guys think?
> >
> > I've wondered about something like that myself. Hadn't suggested it
> > because I'm still unsure if it's a good idea or how it could best be
> > implemented.
> >
> > Actually, when playing with floor stacking on, I'd be happy with a
> > *destroy stack* option, so once I removed the good stuff, I could kill
> > a whole stack of junk in one go.
> >
> > Then again, playing with mixed items stacking on the floor still feels
> > a bit like cheating. Especially when using an anti-summoning corridor.
>
>
> Oh, come on now, is floor stacking really so bad?
imho yes.
> I always wondered
> why a scroll took up just as much space as a suit of armor or a chest
> on the floor.
it's not a realism issue; it's a *balance* issue.
angband is already non-realistic wear it makes game-play and balance
sense. it makes as much sense as the two rings or dozen items in your
inventory. how does an adventurer with eight arms (to
wield/wear/carry all that stuff) end up with only two fingers (for
rings)?
> The stacking limitation was a programmatic one, not one
> which should be defended as somehow making sense.
it originally was pragmatic, but it was noticeed to serve a balance
issue. fighting in an anti-summon corridor is a huge advantage to the
player. even against non-summoners, it means fighting only one
monster at a time (barring ghosts and the like). will no stacking,
you had to hop out into a large open space with a great wyrm in order
to reap its full drop. being in the middle of a room with even a
severly wounded wyrm of chaos is very dangerous. now you can sit back
in your anti-summon and just mash the movement key whilst holding down
control.
> It was wrong, and
> eventually someone with the skill to fix it, did.
and broke balance in the process.
Johan Kullstam
> Excerpts from netnews.rec.games.roguelike.angband: 9-Dec-99 Re: [All] A
> Way to Get Rid .. by William Tanksley@hawking
> > > Oh, come on now, is floor stacking really so bad?
> >
> > BTW, Ed, do you realize how confrontational you sound? I can tell here
> > that you don't mean anything bad, but you consistently sound like you
> > expect the person you're discussing an issue with to throw everything down
> > and start swinging at you ;-).
>
> Maybe if you were a good enough Angband player, you'd realize where he
> is coming from. ^^
and so now *you* want everyone to throw everything down and start
swinging at you *too*?
Andrew Tait
> Andrew Tait wrote:
> >
> > Ed Cogburn <ecog...@greene.xtn.net> wrote in message
> > news:384E4986...@greene.xtn.net...
> > This sounds good. And I do like the 'better idea than mine' bit :)
There's
>
>
> The evidence is overwhelming, I must admit: I actually did say
> that. Bizarre isn't it? :-)
's wonderful. Best to play it down, and make it appear that you meant to
say it in the first place. After all, then you're not making it look as
though any idea you had was better than _mine_ ;) Couldn't have that.
> First, I was wrong in my assumption about 'easy sense', it doesn't
> work at creation time, it works when you (L)ook at an item or walk
> over it, but it shouldn't be too hard to borrow the easy sense
> functions and use them at creation time.
Ahhh. You were 'wronger than you'? ;) I think the _option_ of being able
to destroy at creation time should be there...
> "Destroy under player" is certainly possible, since 'easy sense'
> essentially does that now. You could simply extend 'easy sense' to
> become 'easy sense and destroy worthless'.
Yip. Sounds practicable.
> The only problem is will people ever get used to the idea of items
> disappearing when they walk over it? Or how about items *visually*
> disappearing when the (L)ook at it?
"You look at the Potion of Weakness. You destroy the Potion of Weakness."?
A bit Mindcraftery for me. As an extension of "Query Pickup", I can see
things looking and feeling more appropriate. "Pickup the Wand of
Light? -more- You destroy the Ring of Weakness -more- Pickup the Ring of
Speed -more- You destroy the Broken Stick." etc. This would all be done
in your "walk" or "stand still" turn, so wouldn't take up extra game turns
the way that "k-" does.
> I don't think this would fly too
> well; it would be a clear reminder that they are playing a game with a
> cheesy way of suppressing junk. And no cute message will help here.
> It would ruin the atmosphere, it seems to me, but this is probably an
> individual thing. Destroying junk at creation time is silent; the
> player never knows its happening (unless he turns on debug mode).
Some dieharders are going to want to (a) play as it is, with their "destroy
top item" macro, using OoD etc., or (b) at least be able to see the process
happening. A message saying "The dross of the level has been suppressed"
isn't exactly encouraging, IMO.
> > D'oh! (Twice today??) You said "successfully ID'd is auto-destroyed".
>
>
> This is only an issue with weapons/armor which always requires an ID
> to see if they are good, bad, or ugly (any Clint E. fans? :-) ). With
> a potion for example you only ID it once, you know it's clones from
> then on. Or wands, where you only need a second ID to see how many
> charges it has, which is basically irrelevant to what we're talking
> about. If you took out the part of my thing which ID's weapons/armor
> for the (high-level) player, the rest is not only easy to code for,
> but perfectly reasonable and non-munchkinish.
Yeah, you don't want to get into a situation where you can select
autodestroy for all Wands of Light with less than 20 charges. "You aim the
Wand of Light -more- The Snaga dies -more- Your WoL has 19 charges. You
destroy..." Silly!
> > Yes! You're ideas fit was I was thinking. Adding the ability to flag
> > certain objects as undesirable (eg Scrolls of Light) would make the
whole
> > trip more enjoyable. No munchkinism, as _in theory_, you could have
left
> > the objects where they were without touching them (0 game turns)
>
>
> Right, definitely non-munchkinish here. The problem, as I've been
> talking to Jason about, is how many flags should we setup for the
> player to deal with all the possibilities people might want. Do we
> make a flag for *every* single item type in the game? Or do we go for
> general categories (like "mundane" weapons), with a number of
> exceptions to deal with things a lot of people would like to have
> control over (like treating a RoS separate from all the other lesser
> rings).
> My thing, as it stands now, doesn't give you the option of
> suppressing a SoL specifically, or suppressing all rings except RoS
> and possibly RoD, so .... should it?
> My idea about this is the general category thing will work, with a
> number of automatic exceptions. Rings of Speed, Westernesse weapons
> (the best ego item and arguably better than many artifact weapons),
> Rings of Damage, Boots of Speed, armor of Resistance, would never be
> suppressed, unless its cursed, in the case of a RoS or RoD.
We may beg to differ. I think it's easier to have each object flaggable.
After all, a Level 50 Vampire Warrior might want to destroy all Rings of
Slaying, but hang onto Flasks of Oil. My suggestion that when you 'k' an
object you get an option to flag it for future destruction seems fairly
logical.
> > I need a break....
>
>
> So do I! I've spent more time in the last 2-3 weeks tinkering with
> Angband than playing it.
I'm struggling with GCC and 'Make'ing. Spending 15 minutes at a time on it
is not a good way to learn, but I've suspending playing until I can compile.
Totty <8^)
James W Sager Iii
> sense. it makes as much sense as the two rings or dozen items in your
> inventory. how does an adventurer with eight arms (to
> wield/wear/carry all that stuff) end up with only two fingers (for
> rings)?
LucFrench
>> > > Oh, come on now, is floor stacking really so bad?
>> >
>> > BTW, Ed, do you realize how confrontational you sound? I can tell here
>> > that you don't mean anything bad, but you consistently sound like you
>> > expect the person you're discussing an issue with to throw everything
>down
>> > and start swinging at you ;-).
>>
>> Maybe if you were a good enough Angband player, you'd realize where he
>> is coming from. ^^
>
>and so now *you* want everyone to throw everything down and start
>swinging at you *too*?
-10 points for falling for a troll.
+05 points for not flaming same.
-10 points for not getting the joke.
-03 points for not using capital letters.
-03 points for repeating an earlier joke made by someone else.
-02 points for using a cliche in your sig.
+01 points for being flamed in such an unoriginal way.
---
-22 total points lost.
Care to play again?
Sorry
Luc "Life is Unbalanced" French
Peter Ellis
>>
>> But those catagories are huge! "Mundane" covers everything from Rings
>> of Searching to Rings of Damage, "good" goes from Robes [0,+1] to Power
>> Dragon Scale Mail, and "great" covers Boots of Feathering Falling and
>> Boots of Speed.
>
Deliberately so. To have as much control as you suggest would be
unworkably finicky; *and* take all the thrill out of finding (say) a ring
of Speed. You want to suppress boots of Feather Falling -- tough cookie,
you suppress speed boots too. Sifting through ego-boots for speed ones is
*still* a lot easier than sifting through *all* boots, (worthless, cursed,
mundane, good *and* ego).
If you want to, it shouldn't be too tricky to seperate out DSM from other
armour; but I'm actually against breaking things down on principle. It
should be at least *something* of a tricky decision as to what you choose
to squelch. I think squelching on category (ordinary, good, ego) is
plenty specific enough.
>
> I'm using the "language" of Angband when I say this. Well, mundane
>is my word for average/normal. First, Worthless/Mundane/Good/Great
>refers *only* to weapons and armor:
> Worthless == cursed/broken.
> Mundane == an item with no plusses, ie. Dagger (+0,+0).
> Good == an item with plusses, ie. Dagger (+5,+10).
> Great == ego/artifact.
>
> No ring other than adornment is "non-magical". They are all
>"magical", thus the word mundane isn't used here. So rings are
>divided into good/bad.
>
> In my work, I put DSMs in their own category.
>
> Your right about the boots thing and 'great'. If I forced myself to
>add to the number of flags involved, I could separate the most
>important items of each category into their own special class, like
>speed, resistance, or westernesse weapons/armor, and speed, or damage
>for rings, etc. Hmmmm, we could also use the price of the item to
>determine an 'exception' to the auto-destroy rule: If xyz is worth >
>250,000, don't suppress it....?
A bad idea IMHO, since this *would* begin to verge on the munchkinish.
You have the JoJ for Enlightenment... you suppress everything except Boots
of Speed... you head to the level they're most likely to appear, and bob
up and down stairs till you detect a pair of boots somewhere on the level.
>
>> A squelch mechanism like that is practically useless. I, for one,
>
> Considering how much it already suppresses on a normal level, I'd say
>its far from useless. I've seen it suppress 15+ items during creation
>of a deep level, and it works for monster drops the same way
Once I'm level 40 or so, I'm only interested in ego items and artifacts.
Sorting through the piles of normal and (+8, +10) weapons is a *royal*
pain in the ass, so squelching those would be a Good Thing. But it's wise
to keep the categories reasonably broad, specifically to prevent abuse of
the system.
>> always want to see the high-end of each catagory.
Pooh! Once I have a decent artifact weapon, I only want to see ego
weapons or artifacts. I don't give a toss if a "good" weapon is a Dagger
(0,+1) or a Blade of Chaos (+12,+12). Similarly for cloaks and
armor (excepting only the case of DSM, which is equivalent to an
ego item, and should thus be treated as such)
Once I have the Phial, I want to squelch all light items (unless I'm a
vampire). I always want to squelch cursed things if possible.
>
>
> I agree about this. I'm still trying to get a simpler version
>working first before deciding how to improve it. So far, I've limited
>myself to flags that I can get on one screen, single column, ie.
>"label: flag",
That's fairly easy. You have to have sort categories for which rings fall
into (cursed), (ordinary), (good) and (ego) categories... include this as
a separate text file in the Help. Ditto for Food, light sources, amulets,
wands and rods.
The following could then be an example screen for someone
1) with Soulsword but keeping an eye out for other artifacts or a Holy
Avenger (+4 attacks),
2) They've got Brand, so only want to see artifact bows
3) Squelch all except *really* good rings (eg speed, damage, extra
attacks)
4) Objects to squelching wands and rods on principle, so only weed out
cursed stuff.
5) Wants decent amulets (Anti-Magic, Reflection, or artifact)
6) Has the Phial, so wants onlt artifact light
7) Has an artifact armor, but will accept a good DSM -- counts as ego, so
squelch (g)ood or worse. Robe of Permanence would be nice too.
8) Has Thorin, but is keeping eyes open for a shield of Reflection
9) Looking for BoS
10) Hasn't found *any* good gloves, so is wearing self-enchanted (+9, +9)
gloves, but wants to look for anything (g)ood or better (some decent magic
cesti would be better, see?)
11) Has a shit-hot randart crown, so only wants to see artifacts.
12) Looking for ego or better cloaks (has Colluin, but a decent Elven
cloak of Stealth might be better...)
13) Only wants to see *really* good mushrooms (restoring etc.)
14) Squashing all the crud like spikes, pottery, skeletons.
_________________________________________________________________________
Squelch List: (c)ursed, (o)rdinary, (g)ood, (e)go
Weapon: [g]
Missile Weapon: [e]
Rings: [g]
Wands: [c]
Rods: [c]
Amulet: [g]
Light Source: [e]
Body Armour: [g]
Shield: [g]
Boots: [g]
Gloves: [o]
Headgear: [e]
Cloak: [g]
Food: [g]
Random Junk: [e]
__________________________________________________________________________
This would do all I want -- but there's room for more lines if you want to
subdivide the wands, rings and rods further. Subdividing too far leaves
you open to abuse, though. Reasonable-sized categories preserves that
sense of wonder when you hit your first ever Boots of Speed (never ever
found any in my life, actually).
What do you reckon? A pseudo-ID on creation, followed by this sort of
squelch system. Graduated with character level, so at lower levels the
pseudo-ID won't even knock out skeletons, but by level 40 you can trim out
all but ego items.
Peter
Andrew Tait
> On Thu, 9 Dec 1999, Ed Cogburn wrote:
> >>
> >> But those catagories are huge! "Mundane" covers everything from Rings
> >> of Searching to Rings of Damage, "good" goes from Robes [0,+1] to Power
> >> Dragon Scale Mail, and "great" covers Boots of Feathering Falling and
> >> Boots of Speed.
> >
<SNIP>
> If you want to, it shouldn't be too tricky to seperate out DSM from other
> armour; but I'm actually against breaking things down on principle. It
> should be at least *something* of a tricky decision as to what you choose
> to squelch. I think squelching on category (ordinary, good, ego) is
> plenty specific enough.
But, as I said earlier, this is jugs (not my exact words, IIRC). You
can't define the usefulness of an item based upon its value. There
_will_ be characters (my Vampire example seems useful) of characters
who need Flasks of Oil and Scrolls of Light, but who don't want Gloves
of Slaying or Potions of Wisdom.
Some difference? One man's junk is another man's saviour, and if you
want to pick up Broken Sticks to get the fatal blow on Draebor, then
you should be able to.
> > Your right about the boots thing and 'great'. If I forced myself to
> >add to the number of flags involved, I could separate the most
> >important items of each category into their own special class, like
> >speed, resistance, or westernesse weapons/armor, and speed, or damage
> >for rings, etc. Hmmmm, we could also use the price of the item to
> >determine an 'exception' to the auto-destroy rule: If xyz is worth >
> >250,000, don't suppress it....?
250,000? Geez.
> A bad idea IMHO, since this *would* begin to verge on the munchkinish.
> You have the JoJ for Enlightenment... you suppress everything except Boots
> of Speed... you head to the level they're most likely to appear, and bob
> up and down stairs till you detect a pair of boots somewhere on the level.
The squelching has to work on ID's items only, or when the player
comes into contact with the items.
> Once I'm level 40 or so, I'm only interested in ego items and artifacts.
> Sorting through the piles of normal and (+8, +10) weapons is a *royal*
> pain in the ass, so squelching those would be a Good Thing. But it's wise
> to keep the categories reasonably broad, specifically to prevent abuse of
> the system.
Sounds reasonable. But it sounds to me like you're justifying
creating a potentiatally munchkinish system by crippling it. I want
to see a system which isn't abusable, therefore keeps its usefulness
complete. Categorising may indeed be the answer, but I still suspect
that too many people would find holes in this to actually use the
system for anything apart from the stuff the existing "destroy without
asking" system allows them to destroy anyway.
> >> always want to see the high-end of each catagory.
>
> Pooh! Once I have a decent artifact weapon, I only want to see ego
> weapons or artifacts. I don't give a toss if a "good" weapon is a Dagger
> (0,+1) or a Blade of Chaos (+12,+12). Similarly for cloaks and
> armor (excepting only the case of DSM, which is equivalent to an
> ego item, and should thus be treated as such)
> Once I have the Phial, I want to squelch all light items (unless I'm a
> vampire). I always want to squelch cursed things if possible.
But you still have to ID the item. This involves either (a) picking
it up and waiting for P-ID to kick in, (b) picking it up and ID it
using ID, *ID*, SoP or a spell. At this point, the autodestroy kicks
in and says "No use", and pops the (0,+1) Dagger in the 'k' heap. Not
before!
<BIG snip!>
> What do you reckon? A pseudo-ID on creation, followed by this sort of
> squelch system. Graduated with character level, so at lower levels the
> pseudo-ID won't even knock out skeletons, but by level 40 you can trim out
> all but ego items.
Well, IMVHO, this changes the feel of the current game too much. My
instict would be to go with a system working within the boundaries of
the Angband and co. object and ID system. Making such a generalised
change to pseudo-ID seems chronically misplaced. Would a Level 40
Priest have the same ability to squelch as a Level 40 Warrior??
Within Ang and Zang the pseudo-ID is drastically different, and
provides a genuine depowering of a Priest. Changing this too much in
the other direction knackers the work of years in getting the
character classes to the point they're at now.
Bed time!
Totty <8^)
Johan Kullstam
Ed Cogburn
>
> On Thu, 9 Dec 1999, Ed Cogburn wrote:
> >>
> >> But those catagories are huge! "Mundane" covers everything from Rings
> >> of Searching to Rings of Damage, "good" goes from Robes [0,+1] to Power
> >> Dragon Scale Mail, and "great" covers Boots of Feathering Falling and
> >> Boots of Speed.
> >
>
> Deliberately so. To have as much control as you suggest would be
> unworkably finicky; *and* take all the thrill out of finding (say) a ring
> of Speed. You want to suppress boots of Feather Falling -- tough cookie,
> you suppress speed boots too. Sifting through ego-boots for speed ones is
> *still* a lot easier than sifting through *all* boots, (worthless, cursed,
> mundane, good *and* ego).
Yes, but this feature of suppressing mundane/good/great/artifact is
now dependent on your character's class (in the thing I'm doing). I'm
using the code from the 'easy sense' routines. This is basically a
way of doing pseudo-ID in a way appealing to lazy people. :-)
Anyway, instead of waiting for the player to (L)ook at, or pick up an
item before auto-sensing it, as 'easy sense' does now, I have the
routine run at the time the item is *created*. No cheating, no free
IDs, if your sensing failed, the items remain unknown, and requires
you to ID them as before. As your character gets better though, so
will his ability to successfully sense items, which in turn means more
junk items will be quietly suppressed without the player even knowing
about it (outside of debug mode). This only applies to weapons/armor
however, as you identify 'bad' magical items, eg. ring of
teleportation, wand of heal monster, amulet of DOOM, mushroom of
confusion, etc., they are auto-suppressed from that point on in the
game without having you to fiddle with any switches. The ability to
suppress bad magical items works equally for all classes. There are
other switches too, for suppressing conventional food, 'good' food
like waybread and the good mushrooms, torches/lamps/oil flasks (once
you have the phial), and a few more. Projectile weapons are in two
groups, the latter being the crossbows, which are very popular, so if
you are looking only for a good crossbow, you can suppress
slings/bows/shots/arrows. Spellbooks from magic realms other than
your own are auto-suppressed unless you set a flag the prevents that,
if for example, you want to get the books to sell for gold. Spell
books are split into lesser and greater groups (the latter being the
real profitable ones).
The "problem" will be for those classes which have only a 'light'
sensing ability, like mages and priests. Because light sensing
doesn't distinguish between a 'mundane', 'good', and 'great'
weapons/armor, the only thing that can be done here is the suppression
of cursed/broken weapons/armor, assuming the sensing was successful.
For a high level warrior with great sensing ability on the other
hand, cleaning out a troll pit without ever seeing bad magical items,
mundane or good weapons/armor (assuming he's set the suppression on
for both 'mundane' and 'good'), life will be *much* easier.
There is a much bigger problem, unfortunately. [Z] doesn't have
'easy sense'. [GSN] does, but not [Z], [V], or [O]. Which means this
thing, if I ever release it (I just might keep on tinkering with it
for years (-: ), is not likely to make it to other *bands, even
assuming people are interested in it, which itself isn't really known
at this point. I *really don't* want to duplicate code, here.
>
> If you want to, it shouldn't be too tricky to seperate out DSM from other
> armour; but I'm actually against breaking things down on principle. It
> should be at least *something* of a tricky decision as to what you choose
> to squelch. I think squelching on category (ordinary, good, ego) is
> plenty specific enough.
There's a good discussion, including this thread, about how
fine-grained it should go. I don't want to go to flags for every
entry in k_info (which in [GSN] is about 567 items!), but I can also
see the rational for separating out some of the really important
things, like DSM (which *is* handled separately) from the armor
category, crossbows from the bows category, food split into a
conventional and good group, spellbooks in two groups, and maybe
others in the future. However, I'm now up to *31* flags, which I
think is pushing the limits of practicality, right now. And testing
this monstrosity is going to be tedious, and quite painful.
> >
> > I'm using the "language" of Angband when I say this. Well, mundane
> >is my word for average/normal. First, Worthless/Mundane/Good/Great
> >refers *only* to weapons and armor:
> > Worthless == cursed/broken.
> > Mundane == an item with no plusses, ie. Dagger (+0,+0).
> > Good == an item with plusses, ie. Dagger (+5,+10).
> > Great == ego/artifact.
> >
> > No ring other than adornment is "non-magical". They are all
> >"magical", thus the word mundane isn't used here. So rings are
> >divided into good/bad.
> >
> > In my work, I put DSMs in their own category.
> >
> > Your right about the boots thing and 'great'. If I forced myself to
> >add to the number of flags involved, I could separate the most
> >important items of each category into their own special class, like
> >speed, resistance, or westernesse weapons/armor, and speed, or damage
> >for rings, etc. Hmmmm, we could also use the price of the item to
> >determine an 'exception' to the auto-destroy rule: If xyz is worth >
> >250,000, don't suppress it....?
>
> A bad idea IMHO, since this *would* begin to verge on the munchkinish.
> You have the JoJ for Enlightenment... you suppress everything except Boots
> of Speed... you head to the level they're most likely to appear, and bob
> up and down stairs till you detect a pair of boots somewhere on the level.
What I had before was accused of being munchkinish! There doesn't
seem to be any neutral ground here. :-) I'm a little dense: JoJ?
>
> >
> >> A squelch mechanism like that is practically useless. I, for one,
> >
> > Considering how much it already suppresses on a normal level, I'd say
> >its far from useless. I've seen it suppress 15+ items during creation
> >of a deep level, and it works for monster drops the same way
>
> Once I'm level 40 or so, I'm only interested in ego items and artifacts.
> Sorting through the piles of normal and (+8, +10) weapons is a *royal*
> pain in the ass, so squelching those would be a Good Thing. But it's wise
> to keep the categories reasonably broad, specifically to prevent abuse of
> the system.
>
> >> always want to see the high-end of each catagory.
>
> Pooh! Once I have a decent artifact weapon, I only want to see ego
> weapons or artifacts. I don't give a toss if a "good" weapon is a Dagger
> (0,+1) or a Blade of Chaos (+12,+12). Similarly for cloaks and
> armor (excepting only the case of DSM, which is equivalent to an
> ego item, and should thus be treated as such)
> Once I have the Phial, I want to squelch all light items (unless I'm a
> vampire). I always want to squelch cursed things if possible.
>
> >
> >
> > I agree about this. I'm still trying to get a simpler version
> >working first before deciding how to improve it. So far, I've limited
> >myself to flags that I can get on one screen, single column, ie.
> >"label: flag",
I've already blown this clear out of the water. I've got 31 flags
which will require a multi-page options display, that doesn't
currently exist (yet) :-).
>
> That's fairly easy. You have to have sort categories for which rings fall
> into (cursed), (ordinary), (good) and (ego) categories... include this as
> a separate text file in the Help. Ditto for Food, light sources, amulets,
> wands and rods.
>
> The following could then be an example screen for someone
>
> 1) with Soulsword but keeping an eye out for other artifacts or a Holy
> Avenger (+4 attacks),
Ok. Keep looking for 'great' and suppress mundane and good. P.S.
artifacts, *even the bad ones*, are never suppressed.
> 2) They've got Brand, so only want to see artifact bows
Ok. Suppress mundane/good/great for both bow categories (but don't
suppress ammo).
> 3) Squelch all except *really* good rings (eg speed, damage, extra
> attacks)
Don't have this. What do I put in the 'favored' group? Speed,
Damage, and Extra Attacks sounds good, but others will want, say,
Poison Resistance, Anti-Teleportation, etc. There are too many good
rings to simply pick three and say everyone will be happy with these
things and suppressing everything else. I'm between a rock and a hard
place on this one.
> 4) Objects to squelching wands and rods on principle, so only weed out
> cursed stuff.
Don't understand this.
> 5) Wants decent amulets (Anti-Magic, Reflection, or artifact)
Amulets right now are just good/bad.
> 6) Has the Phial, so wants onlt artifact light
Yep, after getting the Phial, suppress 'lite' items.
> 7) Has an artifact armor, but will accept a good DSM -- counts as ego, so
> squelch (g)ood or worse. Robe of Permanence would be nice too.
Yep, suppress nothing on DSM, and mundane/good for regular armor.
> 8) Has Thorin, but is keeping eyes open for a shield of Reflection
Shields are lumped into the armor category. If I break this out,
I'll have to do the same for cloaks, gauntlets, helms, boots.....
[shudder]
> 9) Looking for BoS
I may make an exception about this, because its an obvious example of
a category with one outstanding member that everyone wants.
> 10) Hasn't found *any* good gloves, so is wearing self-enchanted (+9, +9)
> gloves, but wants to look for anything (g)ood or better (some decent magic
> cesti would be better, see?)
Same problem with the shields from 8, above.
> 11) Has a shit-hot randart crown, so only wants to see artifacts.
Ditto.
> 12) Looking for ego or better cloaks (has Colluin, but a decent Elven
> cloak of Stealth might be better...)
Ditto.
> 13) Only wants to see *really* good mushrooms (restoring etc.)
Yep, suppress conventional food, but not 'good food'.
> 14) Squashing all the crud like spikes, pottery, skeletons.
Done automatically. P.S. does anyone *ever* use spikes?
> _________________________________________________________________________
>
> Squelch List: (c)ursed, (o)rdinary, (g)ood, (e)go
>
> Weapon: [g]
> Missile Weapon: [e]
> Rings: [g]
> Wands: [c]
> Rods: [c]
> Amulet: [g]
> Light Source: [e]
> Body Armour: [g]
> Shield: [g]
> Boots: [g]
> Gloves: [o]
> Headgear: [e]
> Cloak: [g]
> Food: [g]
> Random Junk: [e]
You've gone a little further here than I have. If this things gets
enough interest, it may evolve into this.
> This would do all I want -- but there's room for more lines if you want to
> subdivide the wands, rings and rods further. Subdividing too far leaves
> you open to abuse, though. Reasonable-sized categories preserves that
> sense of wonder when you hit your first ever Boots of Speed (never ever
> found any in my life, actually).
>
> What do you reckon? A pseudo-ID on creation, followed by this sort of
> squelch system. Graduated with character level, so at lower levels the
> pseudo-ID won't even knock out skeletons, but by level 40 you can trim out
> all but ego items.
'easy sense' uses the same internal routines that normal pseudo-ID
uses, so it will be graduated, the only difference will be that
sensing happens at creation time instead of occurring in the backpack
(conventional pseudo-ID), or from the look command or by walking over
it (easy sense).
>
> Peter
Ed Cogburn
>
> Peter Ellis wrote the following about Re: [All] A Way to Get Rid of Garbage Items?:
>
> > On Thu, 9 Dec 1999, Ed Cogburn wrote:
> > >>
> > >> But those catagories are huge! "Mundane" covers everything from Rings
> > >> of Searching to Rings of Damage, "good" goes from Robes [0,+1] to Power
> > >> Dragon Scale Mail, and "great" covers Boots of Feathering Falling and
> > >> Boots of Speed.
> > >
>
>
> > If you want to, it shouldn't be too tricky to seperate out DSM from other
> > armour; but I'm actually against breaking things down on principle. It
> > should be at least *something* of a tricky decision as to what you choose
> > to squelch. I think squelching on category (ordinary, good, ego) is
> > plenty specific enough.
>
> can't define the usefulness of an item based upon its value. There
> _will_ be characters (my Vampire example seems useful) of characters
> who need Flasks of Oil and Scrolls of Light, but who don't want Gloves
> of Slaying or Potions of Wisdom.
>
> Some difference? One man's junk is another man's saviour, and if you
> want to pick up Broken Sticks to get the fatal blow on Draebor, then
> you should be able to.
Broken Sticks? BROKEN STICKS? Come on Andrew, be reasonable. NO
ONE is interested in broken sticks, except, apparently, you.
The ONLY way to get what you want is to setup a multi-value flag for
every possible item. [GSN] has 567 items in k_info. Even throwing
out the artifacts, you're still talking about *hundreds* of switches.
NO ONE is going to go for that. I've got 31 boolean switches now, and
I'm already wondering if that is too much.
>
> > > Your right about the boots thing and 'great'. If I forced myself to
> > >add to the number of flags involved, I could separate the most
> > >important items of each category into their own special class, like
> > >speed, resistance, or westernesse weapons/armor, and speed, or damage
> > >for rings, etc. Hmmmm, we could also use the price of the item to
> > >determine an 'exception' to the auto-destroy rule: If xyz is worth >
> > >250,000, don't suppress it....?
>
> 250,000? Geez.
Your point being?
>
> > A bad idea IMHO, since this *would* begin to verge on the munchkinish.
> > You have the JoJ for Enlightenment... you suppress everything except Boots
> > of Speed... you head to the level they're most likely to appear, and bob
> > up and down stairs till you detect a pair of boots somewhere on the level.
>
> The squelching has to work on ID's items only, or when the player
> comes into contact with the items.
Yes, they have to be ID'd before being squelched, but it happens at
creation time, not when the player reaches it. The test is still the
same though, so there is no difference between testing at creation
time vs. when the player picks it up. The latter thing is what 'easy
sense' does. I don't think people will like things going *poof* right
in front of them. It'll ruin the game for some people because of that
kind of really munchkinish behavior.
>
> > Once I'm level 40 or so, I'm only interested in ego items and artifacts.
> > Sorting through the piles of normal and (+8, +10) weapons is a *royal*
> > pain in the ass, so squelching those would be a Good Thing. But it's wise
> > to keep the categories reasonably broad, specifically to prevent abuse of
> > the system.
>
> Sounds reasonable. But it sounds to me like you're justifying
> creating a potentiatally munchkinish system by crippling it. I want
> to see a system which isn't abusable, therefore keeps its usefulness
> complete. Categorising may indeed be the answer, but I still suspect
> that too many people would find holes in this to actually use the
> system for anything apart from the stuff the existing "destroy without
> asking" system allows them to destroy anyway.
What holes? What abuse? What are you talking about? The whole
process is under computer control, except for setting the flags to
begin with. How do you 'abuse' that?
>
> > >> always want to see the high-end of each catagory.
> >
> > Pooh! Once I have a decent artifact weapon, I only want to see ego
> > weapons or artifacts. I don't give a toss if a "good" weapon is a Dagger
> > (0,+1) or a Blade of Chaos (+12,+12). Similarly for cloaks and
> > armor (excepting only the case of DSM, which is equivalent to an
> > ego item, and should thus be treated as such)
> > Once I have the Phial, I want to squelch all light items (unless I'm a
> > vampire). I always want to squelch cursed things if possible.
>
> But you still have to ID the item. This involves either (a) picking
> it up and waiting for P-ID to kick in, (b) picking it up and ID it
> using ID, *ID*, SoP or a spell. At this point, the autodestroy kicks
> in and says "No use", and pops the (0,+1) Dagger in the 'k' heap. Not
> before!
The ID method is from the 'easy sense' routine. It uses the same
internal routines used by conventional pseudo ID, it just happens at a
different time than what everyone is used to (in backpack). All I've
done is to apply it at yet another different time.
Your not making sense here. Your suggestion does *nothing* useful;
it still requires the player to pick everything up before dealing with
junk items. The whole *point* of the junk suppression idea is to
*reduce* the tedious work of ID'ing and assaying everything dropped
after clearing out a troll pit.
>
> <BIG snip!>
>
> > What do you reckon? A pseudo-ID on creation, followed by this sort of
> > squelch system. Graduated with character level, so at lower levels the
> > pseudo-ID won't even knock out skeletons, but by level 40 you can trim out
> > all but ego items.
>
> Well, IMVHO, this changes the feel of the current game too much. My
> instict would be to go with a system working within the boundaries of
> the Angband and co. object and ID system. Making such a generalised
How does it change the feel of the game? It happens silently, the
player is never aware of what's going on, unless he turns debug mode
on. What boundaries? The ID system used is the same as we use now.
> change to pseudo-ID seems chronically misplaced. Would a Level 40
> Priest have the same ability to squelch as a Level 40 Warrior??
> Within Ang and Zang the pseudo-ID is drastically different, and
> provides a genuine depowering of a Priest. Changing this too much in
> the other direction knackers the work of years in getting the
> character classes to the point they're at now.
Good grief Andrew, its THE SAME PSEUDO-ID ROUTINE YOU'VE BEEN USING.
The ID mechanism is the same, it just occurs at a different time.
There will be no change to what you already know. Warriors have a
better ability at pseudo-id (heavy sensing) than a Priest (light
sensing), so your Level 40 Warrior will be better at ID'ing stuff than
a Level 40 Priest, *JUST* *AS* *IT* *IS* *NOW*, so therefore he'll be
better at suppressing junk. Logical, ain't it.
>
> Bed time!
>
> Totty <8^)
Peter Ellis
>>
>> Deliberately so. To have as much control as you suggest would be
>> unworkably finicky; *and* take all the thrill out of finding (say) a ring
>> of Speed. You want to suppress boots of Feather Falling -- tough cookie,
>> you suppress speed boots too. Sifting through ego-boots for speed ones is
>> *still* a lot easier than sifting through *all* boots, (worthless, cursed,
>> mundane, good *and* ego).
>
>
> Yes, but this feature of suppressing mundane/good/great/artifact is
>now dependent on your character's class (in the thing I'm doing). I'm
>using the code from the 'easy sense' routines. This is basically a
>way of doing pseudo-ID in a way appealing to lazy people. :-)
>
Umph. Not sure I like this, since it means that (say) warriors end up
with really good sensing ability right from first level, while a level 50
mage is still crap (at least in terms of being able to distinguish
categories of item).
I'd really much rather see a system where low-level characters *can't* get
any advantage from it, so it's just there to simplify things for
high-level characters.
If you want to retain the difference between warriors and priets, say, you
could have the warrior's easy-sensing kick in (in a graduated manner) at
levels 15/25/40 -- and for priests it would be 25/35/50, or some similar
system.
>Anyway, instead of waiting for the player to (L)ook at, or pick up an
>item before auto-sensing it, as 'easy sense' does now, I have the
>routine run at the time the item is *created*. No cheating, no free
>IDs, if your sensing failed, the items remain unknown, and requires
>you to ID them as before.
Ah, right -- and this provides a level-graduated effect of the sort I'd be
looking for.
As your character gets better though, so
>will his ability to successfully sense items, which in turn means more
>junk items will be quietly suppressed without the player even knowing
>about it (outside of debug mode). This only applies to weapons/armor
>however, as you identify 'bad' magical items, eg. ring of
>teleportation, wand of heal monster, amulet of DOOM, mushroom of
>confusion, etc., they are auto-suppressed from that point on in the
>game without having you to fiddle with any switches. The ability to
>suppress bad magical items works equally for all classes.
>There are
>other switches too, for suppressing conventional food, 'good' food
>like waybread and the good mushrooms,
With the sort of system I had in mind, that would combine into one switch.
Cursed mushrooms count as "bad", food rations count as "normal", waybread
and some mushrooms count as "good", and (perhaps) the very best mushrooms
count as "ego". Then you set an appropriate squelch level so you only see
what you want to; but it only takes up one line on the choices screen.
>Projectile weapons are in two
>groups, the latter being the crossbows, which are very popular, so if
>you are looking only for a good crossbow, you can suppress
>slings/bows/shots/arrows. Spellbooks from magic realms other than
>your own are auto-suppressed unless you set a flag the prevents that,
>if for example, you want to get the books to sell for gold. Spell
>books are split into lesser and greater groups (the latter being the
>real profitable ones).
All of these are good things, but again I think should only be available
at higher levels.
> The "problem" will be for those classes which have only a 'light'
>sensing ability, like mages and priests. Because light sensing
>doesn't distinguish between a 'mundane', 'good', and 'great'
>weapons/armor, the only thing that can be done here is the suppression
>of cursed/broken weapons/armor, assuming the sensing was successful.
I don't like this. Upgrade to a warrior's pseudo-ID at high character
level?
> There is a much bigger problem, unfortunately. [Z] doesn't have
>'easy sense'. [GSN] does, but not [Z], [V], or [O]. Which means this
>thing, if I ever release it (I just might keep on tinkering with it
>for years (-: ), is not likely to make it to other *bands, even
>assuming people are interested in it, which itself isn't really known
>at this point. I *really don't* want to duplicate code, here.
I'm kinda hoping the relevant DevTeams are watching the debate and having
thoughts of their own...
>
> There's a good discussion, including this thread, about how
>fine-grained it should go. I don't want to go to flags for every
>entry in k_info (which in [GSN] is about 567 items!), but I can also
>see the rational for separating out some of the really important
>things, like DSM (which *is* handled separately) from the armor
>category, crossbows from the bows category, food split into a
>conventional and good group, spellbooks in two groups, and maybe
>others in the future. However, I'm now up to *31* flags, which I
>think is pushing the limits of practicality, right now. And testing
>this monstrosity is going to be tedious, and quite painful.
Ah yes, but for food you're using several flags where one would suffice.
Lessee...
********************************
Hand weapon
Missile Weapon (x2 if you seperate out crossbows)
Amulet
Light
Body Armour (have DSM count as "ego" Body Armour, so no need for another
flag)
Cloak
Shield
Helm
Gloves
Boots
Random Junk
Food
Books (x2 -- one flag for suppressing unusable realms, one for value)
Cursed magical items (Bad mushrooms, amulets of Doom etc. etc.)
***********************
I count 16 flags only. How you get to 31 I have no idea, unless you're
literally having every value category seperate. That's daft. Am I ever
going to want to squelch ego weapons but still see "good" weapons? If you
allow each flag to have several states (rather than just yes/no), you gain
a lot of clarity and screen real estate without making it conceptually any
more difficult for the player.
If you *wanted* to draw up some "value categories" for rings, potions,
wands, rods etc., you could; but you'll never manage to please people with
that, so I wouldn't bother.
As regards any other type of item I haven't mentioned above (chests etc.),
well they're rare enough that it's not enough of a problem to bother
squelching them.
>> A bad idea IMHO, since this *would* begin to verge on the munchkinish.
>> You have the JoJ for Enlightenment... you suppress everything except Boots
>> of Speed... you head to the level they're most likely to appear, and bob
>> up and down stairs till you detect a pair of boots somewhere on the level.
>
> What I had before was accused of being munchkinish! There doesn't
>seem to be any neutral ground here. :-) I'm a little dense: JoJ?
>
Jewel of Judgement [Z] -- equivalent of Vanilla's Arkenstone. Activates
to light and map the entire level, and detect all objects. In conjunction
with a fine-grained squelch system and connected stairs, it could easily
be used to scum for the items you want without any risk whatsoever.
> I've already blown this clear out of the water. I've got 31 flags
>which will require a multi-page options display, that doesn't
>currently exist (yet) :-).
Stop using two-state flags, then. Just because it's a computer doesn't
mean you have to think binary.
>
> You've gone a little further here than I have. If this things gets
>enough interest, it may evolve into this.
It would save you having to write a multi-screen display...
> 'easy sense' uses the same internal routines that normal pseudo-ID
>uses, so it will be graduated, the only difference will be that
>sensing happens at creation time instead of occurring in the backpack
>(conventional pseudo-ID), or from the look command or by walking over
>it (easy sense).
Fair enough. I still think mages and priests should upgrade to the better
pseudo-ID at some stage, though. The original rationale for the reduced
pseudo-ID was that they have their own ID spell, so don't need to spend
money on scrolls -- warrios got better pseudo-ID to compensate...
But at higher levels money is no object, so this falls down; and all it is
is the tedium of identifying reams and reams of crap. That should be
dispensable!
Peter
Peter Ellis
>
>Within Ang and Zang the pseudo-ID is drastically different, and
>provides a genuine depowering of a Priest. Changing this too much in
>the other direction knackers the work of years in getting the
>character classes to the point they're at now.
>
Yeah, right. So your level 30 priest really hangs around waiting for
Pseudo-ID to kick in before casting his Identify spell, and your level 30
Warrior really waits to pseudo-ID things before reading a scroll...
Balderdash. Once money is no object, or you have an Identify spell, then
pseudo-ID as it currently stands becomes utterly irrelevant. Sure, at low
levels it has something going for it, and yes, genuinely does depower
Priests.
I'm not talking about that. I'm taking about a squelch system that kicks
in at higher levels (once pseudo-ID has been rendered redundant by huge
amounts of cash), and makes like less boring for high-level characters.
Peter
Ed Cogburn
>
> Way to Get Rid .. by William Tanksley@hawking
> > > Oh, come on now, is floor stacking really so bad?
> >
> > BTW, Ed, do you realize how confrontational you sound? I can tell here
> > that you don't mean anything bad, but you consistently sound like you
> > expect the person you're discussing an issue with to throw everything down
> > and start swinging at you ;-).
>
> Maybe if you were a good enough Angband player, you'd realize where he
> is coming from. ^^
Me? I wasn't trying to be confrontational, I thought it was a little
sardonic humor, no offense intended.
What does my lack of playing skill have to do with this? What, from
now on only Angband WINNERS are allowed to post?
--
Ed C.
[That's funny William, *you* accusing *me* of being confrontational.]
Ed Cogburn
>
> Ed Cogburn <ecog...@greene.xtn.net> writes:
>
> > Richard Henry wrote:
> > >
> > > Then again, playing with mixed items stacking on the floor still feels
> > > a bit like cheating. Especially when using an anti-summoning corridor.
> >
> >
>
>
> > I always wondered
> > why a scroll took up just as much space as a suit of armor or a chest
> > on the floor.
>
> it's not a realism issue; it's a *balance* issue.
Floor stacking has little effect on balance. Look at all the
variants out there with radically different balance issues. The
escalation of dangerous monsters with obscene amounts of hit points is
far more important when it comes to balance than floor stacking.
The situation got so bad, that one person, Leon Marrick, produced
[Oangband] with a radical redesign of monster capabilities and hit
points, a vastly improved combat system that put a little sanity back
into the game, but note that not only did he keep floor stacking, but
in [O] its no longer an option that can be turned off. I also note
that in 2.8.5, Ben Harrison has done the same thing. I'm not your
problem, ask Ben about floor stacking, I'd like to hear his response
too.
>
> angband is already non-realistic wear it makes game-play and balance
> sense. it makes as much sense as the two rings or dozen items in your
> inventory. how does an adventurer with eight arms (to
> wield/wear/carry all that stuff) end up with only two fingers (for
> rings)?
Not this again. Angband has far worse 'realism' problems than floor
stacking. I didn't start the ring war last time, and I'm not doing it
again, here.
>
> > The stacking limitation was a programmatic one, not one
> > which should be defended as somehow making sense.
>
> it originally was pragmatic, but it was noticeed to serve a balance
> issue. fighting in an anti-summon corridor is a huge advantage to the
> player. even against non-summoners, it means fighting only one
> monster at a time (barring ghosts and the like). will no stacking,
> you had to hop out into a large open space with a great wyrm in order
> to reap its full drop. being in the middle of a room with even a
> severly wounded wyrm of chaos is very dangerous. now you can sit back
> in your anti-summon and just mash the movement key whilst holding down
> control.
In other words, you're not using floor stacking because a player
technique has unbalanced the game, so the one-item-per-square should
be kept to make that technique less exploitable. It sounds to me that
floor-stacking isn't the problem, its the anti-summoning corridor
technique thats the *real* problem. Lets try to deal with it
directly, instead of avoiding it by use of another game limitation to
mask the real problem. Look folks, if we had kept restricting
ourselves because every new idea caused a potential/real, large/small
balance problem, we'd all still be playing Moria.
>
> > It was wrong, and
> > eventually someone with the skill to fix it, did.
>
> and broke balance in the process.
So ask Ben to take it out of [V]. Lets see what he says.
--
Ed C.
Ed Cogburn
>
> Ed Cogburn <ecog...@greene.xtn.net> wrote in message
> news:384F3E2C...@greene.xtn.net...
> > Andrew Tait wrote:
> > >
> > > Ed Cogburn <ecog...@greene.xtn.net> wrote in message
> > > news:384E4986...@greene.xtn.net...
>
> > > This sounds good. And I do like the 'better idea than mine' bit :)
> There's
> >
> >
> > The evidence is overwhelming, I must admit: I actually did say
> > that. Bizarre isn't it? :-)
>
> 's wonderful. Best to play it down, and make it appear that you meant to
> say it in the first place. After all, then you're not making it look as
> though any idea you had was better than _mine_ ;) Couldn't have that.
>
> > First, I was wrong in my assumption about 'easy sense', it doesn't
> > work at creation time, it works when you (L)ook at an item or walk
> > over it, but it shouldn't be too hard to borrow the easy sense
> > functions and use them at creation time.
>
> Ahhh. You were 'wronger than you'? ;) I think the _option_ of being able
> to destroy at creation time should be there...
Actually, I've found out that the easy-sense routines can be used at
creation time, which makes it the ideal solution to the problem of how
to handle weapons/armor, without giving the player an advantage with a
free ID.
>
> > "Destroy under player" is certainly possible, since 'easy sense'
> > essentially does that now. You could simply extend 'easy sense' to
> > become 'easy sense and destroy worthless'.
>
> Yip. Sounds practicable.
>
> > The only problem is will people ever get used to the idea of items
> > disappearing when they walk over it? Or how about items *visually*
> > disappearing when the (L)ook at it?
>
> "You look at the Potion of Weakness. You destroy the Potion of Weakness."?
> A bit Mindcraftery for me. As an extension of "Query Pickup", I can see
> things looking and feeling more appropriate. "Pickup the Wand of
> Light? -more- You destroy the Ring of Weakness -more- Pickup the Ring of
> Speed -more- You destroy the Broken Stick." etc. This would all be done
> in your "walk" or "stand still" turn, so wouldn't take up extra game turns
> the way that "k-" does.
>
> > I don't think this would fly too
> > well; it would be a clear reminder that they are playing a game with a
> > cheesy way of suppressing junk. And no cute message will help here.
> > It would ruin the atmosphere, it seems to me, but this is probably an
> > individual thing. Destroying junk at creation time is silent; the
> > player never knows its happening (unless he turns on debug mode).
>
> Some dieharders are going to want to (a) play as it is, with their "destroy
> top item" macro, using OoD etc., or (b) at least be able to see the process
> happening. A message saying "The dross of the level has been suppressed"
> isn't exactly encouraging, IMO.
Do you really want to be constantly interrupted by messages telling
you whats been suppressed while hacking away in a <insert favorite
monster> pit? It works at both dungeon creation time, and at creation
time during monster drops.
>
> > > D'oh! (Twice today??) You said "successfully ID'd is auto-destroyed".
> >
> >
> > This is only an issue with weapons/armor which always requires an ID
> > to see if they are good, bad, or ugly (any Clint E. fans? :-) ). With
> > a potion for example you only ID it once, you know it's clones from
> > then on. Or wands, where you only need a second ID to see how many
> > charges it has, which is basically irrelevant to what we're talking
> > about. If you took out the part of my thing which ID's weapons/armor
> > for the (high-level) player, the rest is not only easy to code for,
> > but perfectly reasonable and non-munchkinish.
>
> Yeah, you don't want to get into a situation where you can select
> autodestroy for all Wands of Light with less than 20 charges. "You aim the
> Wand of Light -more- The Snaga dies -more- Your WoL has 19 charges. You
> destroy..." Silly!
I'm also getting suggestions that it should have much more
fine-grained control. I don't see much middle ground nearby. :-)
Setting up a multi-value switch for every item type in the game is
downright scary to me. I wouldn't know what kind of interface to use
for allowing a player to interact with a hundred or more switches.
SCARY.
>
> > > I need a break....
> >
> >
> > So do I! I've spent more time in the last 2-3 weeks tinkering with
> > Angband than playing it.
>
> I'm struggling with GCC and 'Make'ing. Spending 15 minutes at a time on it
> is not a good way to learn, but I've suspending playing until I can compile.
>
> Totty <8^)
--
Andrew Tait
> Andrew Tait wrote:
> >
> > But, as I said earlier, this is jugs (not my exact words, IIRC). You
> > can't define the usefulness of an item based upon its value. There
> > _will_ be characters (my Vampire example seems useful) of characters
> > who need Flasks of Oil and Scrolls of Light, but who don't want Gloves
> > of Slaying or Potions of Wisdom.
> >
> > Some difference? One man's junk is another man's saviour, and if you
> > want to pick up Broken Sticks to get the fatal blow on Draebor, then
> > you should be able to.
>
>
> Broken Sticks? BROKEN STICKS? Come on Andrew, be reasonable. NO
> ONE is interested in broken sticks, except, apparently, you.
> The ONLY way to get what you want is to setup a multi-value flag for
> every possible item. [GSN] has 567 items in k_info. Even throwing
> out the artifacts, you're still talking about *hundreds* of switches.
> NO ONE is going to go for that. I've got 31 boolean switches now, and
> I'm already wondering if that is too much.
And I quote:
'BTW, Ed, do you realize how confrontational you sound? I can tell here
that you don't mean anything bad, but you consistently sound like you
expect the person you're discussing an issue with to throw everything down
and start swinging at you ;-).'
A couple of days ago, someone (I only keep 24hrs of rgra at work) said that
they kept Broken Sticks to kill Novice Rogues as they ran away. This is
incidental though. I simply meant that scaling objects from 'bad' to 'good'
on any system of value or function doesn't always work.
<sigh> Hundreds of switches. Each switch is a single bit, and I suggested
3 or 4 per item, meaning that even with 600 items you could get all the
switches into 600 bytes. Easily.
Two quick things then: (1) With my system, there is no object management
page. Object autodestroy is added progressively. When you see an object
you don't want to see any more, you set the flag when you destroy it. (2)
I agree overall that having even a small number of switches can be a pain,
but if the /mechanism/ for toggling the switches is inherent in the 'k' and
'l' mechanism, I don't see there being a huge problem with the
implementation and usage of such a system.
>
> >
> > > > Your right about the boots thing and 'great'. If I forced
myself to
> > > >add to the number of flags involved, I could separate the most
> > > >important items of each category into their own special class, like
> > > >speed, resistance, or westernesse weapons/armor, and speed, or damage
> > > >for rings, etc. Hmmmm, we could also use the price of the item to
> > > >determine an 'exception' to the auto-destroy rule: If xyz is worth >
> > > >250,000, don't suppress it....?
> >
> > 250,000? Geez.
>
> Your point being?
How many items are worth 250,000 or more? Apart from artifacts which can't
be destroyed anyway, of course. Value is a dodgy thing to base suppression
on was my point.
> > The squelching has to work on ID's items only, or when the player
> > comes into contact with the items.
>
>
> Yes, they have to be ID'd before being squelched, but it happens at
> creation time, not when the player reaches it. The test is still the
> same though, so there is no difference between testing at creation
> time vs. when the player picks it up. The latter thing is what 'easy
> sense' does. I don't think people will like things going *poof* right
> in front of them. It'll ruin the game for some people because of that
> kind of really munchkinish behavior.
100% correct. Unless the interface 'feels' Angbandy, no-one will go for it,
especially with in Z and V.
>
> >
> > > Once I'm level 40 or so, I'm only interested in ego items and
artifacts.
> > > Sorting through the piles of normal and (+8, +10) weapons is a *royal*
> > > pain in the ass, so squelching those would be a Good Thing. But it's
wise
> > > to keep the categories reasonably broad, specifically to prevent abuse
of
> > > the system.
> >
> > Sounds reasonable. But it sounds to me like you're justifying
> > creating a potentiatally munchkinish system by crippling it. I want
> > to see a system which isn't abusable, therefore keeps its usefulness
> > complete. Categorising may indeed be the answer, but I still suspect
> > that too many people would find holes in this to actually use the
> > system for anything apart from the stuff the existing "destroy without
> > asking" system allows them to destroy anyway.
>
>
> What holes? What abuse? What are you talking about? The whole
> process is under computer control, except for setting the flags to
> begin with. How do you 'abuse' that?
Here we go again: 'BTW, Ed, do you realize how confrontational you sound?'
[Do you realise I learned the basics of Netware 5 on a new server this
morning for the first time with less "What are you talking about?" from my
engineer? Hmmm....]
I think the original statement was that you could suppress all Boots apart
from Boots of Speed with a sufficiently specialised system. This is clearly
Bad. Not being able to suppress {good} Boots is also Bad once you reach a
certain point. My point was that having the system too finely tuned could
be perceived as munchkinish in the extreme, but broadening the categories to
simple "Mundane", "Good" and "Excellent" may leave things unusable.
I'm not coding this beast, just making suggestions. But I would appreciate
a more dignified response than "What are you talking about?" <sighs> Ah
well, I'm still getting paid, I s'pose.
> Good grief Andrew, its THE SAME PSEUDO-ID ROUTINE YOU'VE BEEN USING.
> The ID mechanism is the same, it just occurs at a different time.
> There will be no change to what you already know. Warriors have a
> better ability at pseudo-id (heavy sensing) than a Priest (light
> sensing), so your Level 40 Warrior will be better at ID'ing stuff than
> a Level 40 Priest, *JUST* *AS* *IT* *IS* *NOW*, so therefore he'll be
> better at suppressing junk. Logical, ain't it.
I give in. I hadn't read this far before...
Later.
Totty <8^)
Andrew Tait
> On 9 Dec 1999, Andrew Tait wrote:
> >
> >Within Ang and Zang the pseudo-ID is drastically different, and
> >provides a genuine depowering of a Priest. Changing this too much in
> >the other direction knackers the work of years in getting the
> >character classes to the point they're at now.
> >
>
> Yeah, right.
> Balderdash.
EEK! Everyone's being nasty to me today.
Totty <8^(
James Wilbur Sager
Way to Get Rid .. by Ed Cog...@greene.xtn.ne
> sardonic humor, no offense intended.
>
> What does my lack of playing skill have to do with this? What, from
> now on only Angband WINNERS are allowed to post?
Nah, just losers shouldn't post ^^
James Wilbur Sager
Way to Get Rid .. by Ed Cog...@greene.xtn.ne
> > A bit Mindcraftery for me. As an extension of "Query Pickup", I can see
> > things looking and feeling more appropriate. "Pickup the Wand of
> > Light? -more- You destroy the Ring of Weakness -more- Pickup the Ring of
> > Speed -more- You destroy the Broken Stick." etc. This would all be done
> > in your "walk" or "stand still" turn, so wouldn't take up extra game turns
> > the way that "k-" does.
Wow, just thinking along these lines...
We could walk around with autopickup on!
If everything bad is squelched, then everything we pick up will be
interesting. Hense, no need to sort through piles of stuff, but also
no need to use the 'not so cumbersome' 'g'et command. On top of that,
if you're on a treasure run, you don't have to evaluate if you want to
take stuff, just run on through killing and looting :)
Sounds pretty fun.
Ed Cogburn
>
> Ed Cogburn <ecog...@greene.xtn.net> wrote in message
> news:3850E72C...@greene.xtn.net...
> > Andrew Tait wrote:
> > >
> > > But, as I said earlier, this is jugs (not my exact words, IIRC). You
> > > can't define the usefulness of an item based upon its value. There
> > > _will_ be characters (my Vampire example seems useful) of characters
> > > who need Flasks of Oil and Scrolls of Light, but who don't want Gloves
> > > of Slaying or Potions of Wisdom.
> > >
> > > Some difference? One man's junk is another man's saviour, and if you
> > > want to pick up Broken Sticks to get the fatal blow on Draebor, then
> > > you should be able to.
> >
> >
> > Broken Sticks? BROKEN STICKS? Come on Andrew, be reasonable. NO
> > ONE is interested in broken sticks, except, apparently, you.
> > The ONLY way to get what you want is to setup a multi-value flag for
> > every possible item. [GSN] has 567 items in k_info. Even throwing
> > out the artifacts, you're still talking about *hundreds* of switches.
> > NO ONE is going to go for that. I've got 31 boolean switches now, and
> > I'm already wondering if that is too much.
>
> And I quote:
>
> 'BTW, Ed, do you realize how confrontational you sound? I can tell here
> that you don't mean anything bad, but you consistently sound like you
> expect the person you're discussing an issue with to throw everything down
> and start swinging at you ;-).'
I sorry about being harsh in my last response, I admit that. All I
can say in my own defense, is that if I had written my response to you
*before* reading the other posts, my last reply would not have been
harsh.
Quoting William doesn't help. I've had a run-in with William
before. Anyone who disagrees with him, he considers confrontational.
The broken sticks thing caught me offguard, several people had talked
about fine-grained control of suppressing objects, but no one has
suggested, until now, that we should set aside some switches for
broken sticks. Ok, we can do that, we'll have 32nd bit for
controlling the suppression of the 'natural' junk items, like broken
sticks, broken pottery, and skeletons. Now who is really going to use
this switch to keep the natural junk?
>
> A couple of days ago, someone (I only keep 24hrs of rgra at work) said that
> they kept Broken Sticks to kill Novice Rogues as they ran away. This is
> incidental though. I simply meant that scaling objects from 'bad' to 'good'
> on any system of value or function doesn't always work.
Agreed, no system, short of allowing control over suppression of each
individual object kind, is going to please everyone. I'm trying to be
reasonable and hold the middle ground here, but the truth is I'm
lazy. Trying to provide fine-grained control over everything gives me
a headache just thinking about it. :-)
>
> <sigh> Hundreds of switches. Each switch is a single bit, and I suggested
> 3 or 4 per item, meaning that even with 600 items you could get all the
> switches into 600 bytes. Easily.
That's what I'm doing now. Each switch is a bit. Three switches are
used to give the player the option (when talking about weapons/armor)
of suppressing mundane/good/great. Your right, ignoring the other
stuff that doesn't add up to much, we're talking about less than three
hundred bytes. Its not the memory used by the switches that concerns
me, its the interface to these switches that has me running scared.
Lets see, assuming 20 lines on each screen display, we'll need *90*
option screens to present the user with all of the switches. Even if
we assume only 300 objects to worry about, 45 screens is still way too
much.
>
> Two quick things then: (1) With my system, there is no object management
> page. Object autodestroy is added progressively. When you see an object
> you don't want to see any more, you set the flag when you destroy it. (2)
> I agree overall that having even a small number of switches can be a pain,
> but if the /mechanism/ for toggling the switches is inherent in the 'k' and
> 'l' mechanism, I don't see there being a huge problem with the
> implementation and usage of such a system.
Now this does sound interesting... By 'my system' do you mean you're
trying to code this? I like this, although the internal interface to
300 switches will still be a nightmare. We would, at the very least,
end up the record for the most gigantic C switch statement in the
entire program! :-)
There is one other thing though, what if the player makes a mistake?
He gets a keystroke ahead of himself and suppresses something we
didn't want to suppress. The only way to solve that is for an
alternative interface to the all the switches like an options display,
but then we're back to square one. Hmmm...
>
> >
> > >
> > > > > Your right about the boots thing and 'great'. If I forced
> myself to
> > > > >add to the number of flags involved, I could separate the most
> > > > >important items of each category into their own special class, like
> > > > >speed, resistance, or westernesse weapons/armor, and speed, or damage
> > > > >for rings, etc. Hmmmm, we could also use the price of the item to
> > > > >determine an 'exception' to the auto-destroy rule: If xyz is worth >
> > > > >250,000, don't suppress it....?
> > >
> > > 250,000? Geez.
> >
> > Your point being?
>
> How many items are worth 250,000 or more? Apart from artifacts which can't
> be destroyed anyway, of course. Value is a dodgy thing to base suppression
> on was my point.
Ok, 250,000 is too high. I was just suggesting a number that would
catch the best items from each category, but in hindsight that would
only capture items from some categories. In the misc. armor group
(shield/helm/boots/gloves), nothing goes over 10,000, yet several
items from the body armor group go over 50,000, and RoS and the
greater spellbooks are 100,000 (not counting artifacts). Ok, we would
have to have a value for each group to make this work, but at this
point why not directly pick the best items and give them their own
category?
Please realize that I've been taking it from both sides. One other
person has complained about not being able to treat a BoS or RoS
separately from their normal category. It looks like no matter what I
do, someone is going to hate it. [sigh]
> I'm not coding this beast, just making suggestions. But I would appreciate
> a more dignified response than "What are you talking about?" <sighs> Ah
> well, I'm still getting paid, I s'pose.
I'm sorry, but I still consider it a valid question. What are the
holes to my system? How does allowing a person to treat a BoS
separately from other boots (as someone suggested) 'abuse' the
system? I honestly don't understand this.
>
> > Good grief Andrew, its THE SAME PSEUDO-ID ROUTINE YOU'VE BEEN USING.
> > The ID mechanism is the same, it just occurs at a different time.
> > There will be no change to what you already know. Warriors have a
> > better ability at pseudo-id (heavy sensing) than a Priest (light
> > sensing), so your Level 40 Warrior will be better at ID'ing stuff than
> > a Level 40 Priest, *JUST* *AS* *IT* *IS* *NOW*, so therefore he'll be
> > better at suppressing junk. Logical, ain't it.
>
> I give in. I hadn't read this far before...
The uppercase stuff and liberal use of "*" is the harshness that I
said I regretted. I'm sorry.
>
> Later.
Ed Cogburn
>
> Excerpts from netnews.rec.games.roguelike.angband: 10-Dec-99 Re: [All] A
> Way to Get Rid .. by Ed Cog...@greene.xtn.ne
> > sardonic humor, no offense intended.
> >
> > What does my lack of playing skill have to do with this? What, from
> > now on only Angband WINNERS are allowed to post?
>
> Nah, just losers shouldn't post ^^
Considering everyone has lost perhaps as many as hundreds of times
before winning this game, you'd better define 'loser'.
Jason Willoughby
; Jason Willoughby wrote:
; >
; > Ed Cogburn, a beast of pure hatred with purpose malign, wrote:
; > > Jason Willoughby wrote:
; > Obviously it can't be used as is, but the hard part is giving the user a
; > decent way of specifying a particular k_idx. It's also what people keep
; > complaining about, "You mean I have to go through a list of 500 items to
; > squelch something! What a waste of time!" You don't. You only have to
; > go through two lists of <40.
; I don't understand this. In [GSN] there are ~560 entries in
; k_info.txt. I don't see how using k_info reduces the number of things
; to be dealt with. If you insist on being able to control every single
; item, then you'll inevitably need more than one flag for each item,
; all of which adds up quickly.
Why? Either you squelch an item, or you don't. This seems quite suited
to a boolean value. As I said, the difference between cursed, average,
and good is negligible. Kill 'em all and let God sort it out.
Way, way back at the beginning of this thread, I proposed giving a similar
bool to each ego type, and then only killing great items if both the base
type and the ego type had been killed.
; Working by categories, such as mundane/good/great for weapons/armor and
; bad/good for potions as well as based on TV_<name> instead of each
; individual item via SV_<name>, seems the most practical.
I came up with my system on the basis that is was simplest for the coder,
not overly burdensome on the user, and provides as much power and
flexibility as you could ever possibly want. A lot of the questions
you're trying to solve simply don't come up with a one-flag-per-item
system, and I don't know that the simplified ("dumbed down") interface is
worth the trouble.
Of course, you're writing it and I ain't...
; So, ~560 bools on k_info entries? How many screen pages would it
; take to present all these flags to the player for setting?
One per tval. You seem to think that the player will regularly have to
trawl through all ~700 types, but when would that happen? You put on
Anarion, and squelch all stat and restore potions. Put on Feanor and kill
all the boots. Normally you want to get rid of some chunk of only one
tval, and that only takes about four keystrokes, plus you hold down 'N'
(or whatever) for a while. New items or forgotten old ones can be
squelched with a 'k' modification as you find them, and more diverse
groups (e.g. "junk"), what few there are, could be loaded from and
external file.
Jason Willoughby
; It should be at least *something* of a tricky decision as to what you
; choose to squelch.
We have different ideas of the job of the UI. I believe it should be as
helpful and accessible as possible, including highlighting or hiding
information based on its worth. The *game* should be a complete bastard,
full of trade-offs and nastiness, not the interface.
; A bad idea IMHO, since this *would* begin to verge on the munchkinish.
; You have the JoJ for Enlightenment... you suppress everything except Boots
; of Speed... you head to the level they're most likely to appear, and bob
; up and down stairs till you detect a pair of boots somewhere on the level.
a) This is only a concern if you give free creation-time id.
b) In Z, at least, you'll have a ton of useless randart boots for every
pair of Speed boots.
c) I have no respect and less concern for this sort of abusive scummer.
If you really care, execute the character on 10 consecutive "Looks
like any other level" feelings.
; 1) with Soulsword but keeping an eye out for other artifacts or a Holy
; Avenger (+4 attacks),
; 2) They've got Brand, so only want to see artifact bows
; 3) Squelch all except *really* good rings (eg speed, damage, extra
; attacks)
; 4) Objects to squelching wands and rods on principle, so only weed out
; cursed stuff.
; 5) Wants decent amulets (Anti-Magic, Reflection, or artifact)
[etc]
Problem: This requires complete, spoiler-level knowledge of all normal
and ego items. It's quite possible to make it very deep without seeing a
Mace of Disruption or a Robe of Permanence or Gloves of Great Niftiness.
Without going through the k_ or e_info a player runs the risk of deleting
an unseen, but possibly useful, item.
Ed Cogburn
>
> Ed Cogburn <ecog...@greene.xtn.net> wrote:
> ; Jason Willoughby wrote:
> ; >
> ; > Ed Cogburn, a beast of pure hatred with purpose malign, wrote:
> ; > > Jason Willoughby wrote:
>
> ; > Obviously it can't be used as is, but the hard part is giving the user a
> ; > decent way of specifying a particular k_idx. It's also what people keep
> ; > complaining about, "You mean I have to go through a list of 500 items to
> ; > squelch something! What a waste of time!" You don't. You only have to
> ; > go through two lists of <40.
>
> ; I don't understand this. In [GSN] there are ~560 entries in
> ; k_info.txt. I don't see how using k_info reduces the number of things
> ; to be dealt with. If you insist on being able to control every single
> ; item, then you'll inevitably need more than one flag for each item,
> ; all of which adds up quickly.
>
> Why? Either you squelch an item, or you don't. This seems quite suited
> to a boolean value. As I said, the difference between cursed, average,
> and good is negligible. Kill 'em all and let God sort it out.
Ok, I'm still assuming we want to distinguish between
mundane/good/great in weapons/armor, which would take three flags per
item (suppress mundane, suppress good, suppress ego items).
>
> Way, way back at the beginning of this thread, I proposed giving a similar
> bool to each ego type, and then only killing great items if both the base
> type and the ego type had been killed.
You mean breaking out the 'great' group, into each individual ego
type? Internally, that can be done, there are 'only' about 116 (some
of them are 'bad' ego types so the real number may be under a hundred)
ego types ([GSN]) It would take about 6 screens with 20 flags per
screen to present the flags to the player for setting, and this is
just for ego items. Do you also want to still discriminate between
mundane and good (for classes with heavy sensing)? This all starts
adding up. Does anyone else want to select ego items individually?
To everyone: how many option screens are too many? If we did the
above we'd have 6 screens for ego, and say 3 more for everything
else. Is maintaining 9 screens of flags too many?
>
> ; Working by categories, such as mundane/good/great for weapons/armor and
> ; bad/good for potions as well as based on TV_<name> instead of each
> ; individual item via SV_<name>, seems the most practical.
>
> I came up with my system on the basis that is was simplest for the coder,
> not overly burdensome on the user, and provides as much power and
> flexibility as you could ever possibly want. A lot of the questions
> you're trying to solve simply don't come up with a one-flag-per-item
> system, and I don't know that the simplified ("dumbed down") interface is
> worth the trouble.
>
> Of course, you're writing it and I ain't...
You've got a point: if we handle everything individually it would
only take a bool for each one. To do it by category requires multiple
flags (one for mundane, one for good, etc.)
>
> ; So, ~560 bools on k_info entries? How many screen pages would it
> ; take to present all these flags to the player for setting?
>
> One per tval. You seem to think that the player will regularly have to
The tvals basically are the categories I'm trying to handle, but are
too broad for some. So I either make a lot of exceptions and handle a
lot of things specifically, or I end up with broad categories that no
one likes. People want specific control over boots so they can look
for BoS without having to deal with any other types of boots, they
want to distinguish between a heavy crossbow and all other types, they
want to look for RoS and RoD while suppressing every other type of
ring. On top of this, some believe that this level of fine-grained
control is munchkinish. My only objection to so many switches was the
painful realization of how tedious it will be to code this. I don't
know what to think.
> trawl through all ~700 types, but when would that happen? You put on
> Anarion, and squelch all stat and restore potions. Put on Feanor and kill
> all the boots. Normally you want to get rid of some chunk of only one
> tval, and that only takes about four keystrokes, plus you hold down 'N'
Ideally this would be fine, but the 4 keystroke thing assumes that
everyone will know where the specific flag they want to set/clear is
in, what, maybe 12 screens of flags. Wouldn't trawling through 12
screens of switches to find the one you want be tedious for you?
> (or whatever) for a while. New items or forgotten old ones can be
> squelched with a 'k' modification as you find them, and more diverse
> groups (e.g. "junk"), what few there are, could be loaded from and
> external file.
An extension of the 'k' would be a great idea, for some to avoid
having to use those option screens, but it would take a bit of work.
Also, if the person makes a mistake, he'll still need something like
the option screens anyway to fix the mistake. I'll leave 'k' alone
for the time being; we can do it later, after we got the rest of it
working.
Mariusz Chwalba
> On Thu, 09 Dec 1999 06:32:32 +0000, Ed Cogburn wrote:
>> Richard Henry wrote:
>
> [a spell/scroll to zap items and only destroy the dumb ones...]
>
>>> I've wondered about something like that myself. Hadn't suggested it
>>> because I'm still unsure if it's a good idea or how it could best be
>>> implemented.
>
> implement it, but it's not clear how to justify it. It's a pretty raw
> exposure of internal mechanics.
>
It doesn't have to be. Just say that this is some minor ball spell (Minor
Mana Storm?) which isn't powerfull enough to destroy any magical item, but
easily melts all this junk lying around. It could even do some minor damage,
so it would be a great flea-removal as well.
I don't know about you, but when my ~40 lv character clears a pit, I just
'expand a power of the 'l'ook command' to browse trough items in search of
*healing* potions, RoS, etc, picks them up, and clears the rest with rod
of acid balls/mana storm/any-big-ball-spell. Ever since I did it the
first time, I wanted some kind of 'utility' spell/rod which would help me
handle all these stuff lying there. 'Junk Destruction' is one possibility.
The other I can think of is a 'Mass Telekinesis' - gather all items in one
pile, ID or pseudo-ID it and display them in a shop-like list so I can pick
the stuff I want and destroy ones I don't.
[ snip ]
>> I always wondered
>> why a scroll took up just as much space as a suit of armor or a chest
>> on the floor. The stacking limitation was a programmatic one, not one
>> which should be defended as somehow making sense.
You could always go for some realism and make a maximum weight limit for
floor tile. Five scrolls? Sure. Five armours? Power Dragon Scale Mail
dissapears -more-
This would even encourage people to stop using those anti-summoning coridors.
[ snip the rest ]
Later
--
Mariusz 'Koder' Chwalba
Success is like a fart - only your own smells nice.
-- James P. Hogan
James Wilbur Sager
He meant WINNER.
I meant loser like the insult you shoot around at people.
Just playing the role of a troll for comedic effect(hope its that obvious)
Excerpts from netnews.rec.games.roguelike.angband: 11-Dec-99 Re: [All] A
James Wilbur Sager
> helpful and accessible as possible, including highlighting or hiding
> information based on its worth. The *game* should be a complete bastard,
> full of trade-offs and nastiness, not the interface.
>
> ; You have the JoJ for Enlightenment... you suppress everything except Boots
> ; of Speed... you head to the level they're most likely to appear, and bob
> ; up and down stairs till you detect a pair of boots somewhere on the level.
No, you missed the point completely.
You can only suppress boots in my method.
People are suggesting this extra method:
Once identified, auto suppress based on extra things.
Ie: broken cursed dagger, upon pseudo id
-you determine that dagger was currsed, so you throw it out of your backpack-
Jason Willoughby
; Ideally this would be fine, but the 4 keystroke thing assumes that
; everyone will know where the specific flag they want to set/clear is
; in, what, maybe 12 screens of flags. Wouldn't trawling through 12
; screens of switches to find the one you want be tedious for you?
Have I mentioned the debug mode Create Item command, lately? It takes 2
(two, not twelve) screens to get to every item in the game. Admittedly,
that doesn't include ego items, which may take you all the way to three
screens, but that's still not that bad.
Andrew Tait
> Andrew Tait wrote:
> I sorry about being harsh in my last response, I admit that. All I
> can say in my own defense, is that if I had written my response to you
> *before* reading the other posts, my last reply would not have been
> harsh.
Oh, I was having a bad and very busy day, so flew off the handle a bit
too. Your gracious apology is accepted.
The essential reason we're at crossed purposes, the "I'm right and
you're wrong" argument, as it were, is that your system is an addition
to the existing #?band system, which changes the game mechanism to add
an 'auto-sense', and my system is merely an automation of a process in
the existing game.
> Quoting William doesn't help.
My fault. I can only apologise. Next time I'll think of an original
attack ;)
> The broken sticks thing caught me offguard, several people had talked
> about fine-grained control of suppressing objects, but no one has
> suggested, until now, that we should set aside some switches for
> broken sticks. Ok, we can do that, we'll have 32nd bit for
> controlling the suppression of the 'natural' junk items, like broken
> sticks, broken pottery, and skeletons. Now who is really going to use
> this switch to keep the natural junk?
Oh, conceded absolutely. In this context there's no way that Broken
Sticks could be considered keepable. I never really meant they would
be. Perhaps a better example would be a high level, maxed out
character wanting to throw away Potions of Augmentation, but keep
Potions of CCW.
> > A couple of days ago, someone (I only keep 24hrs of rgra at work) said that
> > they kept Broken Sticks to kill Novice Rogues as they ran away. This is
> > incidental though. I simply meant that scaling objects from 'bad' to 'good'
> > on any system of value or function doesn't always work.
>
>
> Agreed, no system, short of allowing control over suppression of each
> individual object kind, is going to please everyone. I'm trying to be
> reasonable and hold the middle ground here, but the truth is I'm
> lazy. Trying to provide fine-grained control over everything gives me
> a headache just thinking about it. :-)
:) I agree 100% with this, but I don't get a headache over the
fine-grained control. 12 years of programming in every language from
Apple IIe Basic through to x86 assembler to control £m's worth of
machinery have made me very thick skinned to writing gargantuan
table-based switchy logicy thingies like this.
Most mechanisms in Angband are 'hidden', in that you don't perceive
/how/ a monster picks an object up, but fails to pick an artifact up
(Terrible example), yet you know what it means when it happens. As
you rightly said, if every time you step close to a Broken Skull, it
blinks off the map, the game looks very cheesy.
>
> >
> > <sigh> Hundreds of switches. Each switch is a single bit, and I suggested
> > 3 or 4 per item, meaning that even with 600 items you could get all the
> > switches into 600 bytes. Easily.
>
>
> That's what I'm doing now. Each switch is a bit. Three switches are
> used to give the player the option (when talking about weapons/armor)
> of suppressing mundane/good/great. Your right, ignoring the other
> stuff that doesn't add up to much, we're talking about less than three
> hundred bytes. Its not the memory used by the switches that concerns
> me, its the interface to these switches that has me running scared.
> Lets see, assuming 20 lines on each screen display, we'll need *90*
> option screens to present the user with all of the switches. Even if
> we assume only 300 objects to worry about, 45 screens is still way too
> much.
And this /is/ an [All] thread, so you have to assume the worst case, I
s'pose. And currently (apart from in Kang beta?) there is /nothing/
in Anband which pages things like this apart from shop inventories.
Setting and unsetting switches manually during the game on huge lookup
tables was never my suggestion, though. That /does/ give me a
headache! Even with a clever 'search and switch' function, where you
could enter '/"Scroll of Light"+' to toggle, it would be too hairy to
maintain. Every character would have to go through this process from
birth. <shudders>
>
> >
> > Two quick things then: (1) With my system, there is no object management
> > page. Object autodestroy is added progressively. When you see an object
> > you don't want to see any more, you set the flag when you destroy it. (2)
> > I agree overall that having even a small number of switches can be a pain,
> > but if the /mechanism/ for toggling the switches is inherent in the 'k' and
> > 'l' mechanism, I don't see there being a huge problem with the
> > implementation and usage of such a system.
>
>
> Now this does sound interesting... By 'my system' do you mean you're
> trying to code this? I like this, although the internal interface to
> 300 switches will still be a nightmare. We would, at the very least,
> end up the record for the most gigantic C switch statement in the
> entire program! :-)
If it was anything apart from C, I wouldn't be 'trying to code'. It
would be written. Looking at k_info, I don't find a problem with
using arrays for the job. That way you wouldn't have to recode for
each variant too.
> There is one other thing though, what if the player makes a mistake?
> He gets a keystroke ahead of himself and suppresses something we
> didn't want to suppress. The only way to solve that is for an
> alternative interface to the all the switches like an options display,
> but then we're back to square one. Hmmm...
In my very, very original post I suggested an overall interface. To
summarise: Each object can be: (1) Unflagged. (2) Flagged so that
PLAYER gets queried to destroy when he 'touches' it (eg Scroll of Sat.
Hunger) (3) destroyed automatically (eg Broken Stick and Potion of
Augmentation)
Flags are added/taken away using the 'k' command. You may say 'But
how do I stop destroying Broken Sticks if they are flagged?' My
original idea, which may not be the best one, was that a single switch
would be used to temporarily disable /all/ (3) flags and make them
behave like (2) flags.
Example: You're a Priest(V) or Life Priest(Z), wandering happily
'round Mr Dungeon, when your last book with the "Satisfy Hunger" spell
gets burnt. "Drat," you say, "I should never have set Rations of Food
as Autodestroy items!" You flip the "Autodestroy enabled?" switch to
"N" and wander around. Next time you find a Ration of Food, rather
than "You see the Ration of Food. -more- You pound the Ration of Food
to dust with your Mace of Disruption", you see "You see the Ration of
Food -more- Destroy Ration of Food? (y/n/.....)"
'A-ha!" you scream merrily, striking the 'n' key with your little
finger. "Pick up Ration of Food?" 'y' 'E'
Happy with your lot, you can retoggle Autodestroy to its full effect.
It's not ideal, but it does mean that the overall mechanism is hidden
completely from the user.
<big SNIP!>
> Please realize that I've been taking it from both sides. One other
> person has complained about not being able to treat a BoS or RoS
> separately from their normal category. It looks like no matter what
I
> do, someone is going to hate it. [sigh]
I don't hate your idea. I don't /think/ I suggested that, but I
realise that things can get 'discussed' to the point of irritation
with certain people. The 'auto-sense' doesn't tally with my feeling
of the overall Roguelike thing. I can see that High Elves might have
a great 'auto-sense', but that Humans wouldn't. It feels wrong to me
for every class and race to get this ability, and if a poor Perception
character doesn't have 'auto-sense', this would be correct from a game
balance point of view, but still leave that character walking around
picking up Broken Sticks.
In my mind, and I would welcome your opinion, the inherent ability of
each character to ID/pseudo-ID should remain unchanged. If Boots of
Noise are suppressed from the word go by a system which doesn't let
the character know they ever existed, even at a high level, this
changes the feel of the game too much for my liking. Vaults can be
cleared more efficiently, or avoided with impunity if there are no
armour items in them at all, pit drops can be picked up happily
without really thinking about inventory slots, and you'd never see
"You feel the Boots in your backpack are Worthless" ever again.
> I'm sorry, but I still consider it a valid question. What are the
> holes to my system? How does allowing a person to treat a BoS
> separately from other boots (as someone suggested) 'abuse' the
> system? I honestly don't understand this.
Oh, I answered this above. Stet :)
But I am sorry I sounded so cheesed off at you yesterday. Just
getting through a long Friday. At least I won at the casino last
night :)
Take care,
Totty <8^)
Johan Kullstam
> James Wilbur Sager wrote:
> >
> > Excerpts from netnews.rec.games.roguelike.angband: 10-Dec-99 Re: [All] A
> > Way to Get Rid .. by Ed Cog...@greene.xtn.ne
> > > Me? I wasn't trying to be confrontational, I thought it was a little
> > > sardonic humor, no offense intended.
> > >
> > > What does my lack of playing skill have to do with this? What, from
> > > now on only Angband WINNERS are allowed to post?
> >
> > Nah, just losers shouldn't post ^^
>
>
> Considering everyone has lost perhaps as many as hundreds of times
> before winning this game, you'd better define 'loser'.
yo soy un peligor. i'm a loser baby so why don't you shoot me.
-- beck.