New ID system for Angband
Andrew Sidwell
Eddie has other ideas on this... It may not be radical enough, so I
appreciate all feedback.
Problem: Current ID is tedious.
Proposed system:
There are two kinds of knowledge which apply globally to all items of a
given kind:
* Kind lore: this means you recognise the kind of item something is;
e.g. kind of ego-item, kind of wand. At birth, a character has kind
lore of randomly chosen items.
* Charge lore: this means you know how many charges a rechargeable has
without needing to ID it.
There are two kinds of knowledge which apply only to individual items:
* Bonuses: this means you know the (hit, dam, ac) bonuses of
un-ID'd gear.
* Identification: this means you know everything about the item.
Pseudo-ID
---------
There are three/four tiers of items:
* average
* masterwork -- current "good", they are simply well-made items
* magic -- current "everything else"
* (special)
Average items and items the player has lore of are automatically ID'd.
Other items are displayed with their status in brackets appended (e.g.
"(magic)"), until ID'd. Cursed items are unnoticeable until they are
worn or Identified -- curses are powerful magics, more than capable of
hiding themselves. (This would go hand-in-hand with removing
pointlessly cursed items.)
If you use a piece of masterwork combat gear for a while, you ID it. If
you use magic combat gear, then you learn its bonuses, but ID is still
required to figure out what artifact/ego it is.
If you have an artifact currently known as "magic", then after n turns
(or n/2 turns if it is being worn), its status changes to "special".
When using ID
-------------
If you identify an ego-item, there is a small chance you will gain kind
lore of it. Similarly, every time you ID a rechargeable, there is a
chance (proportional to magic devices skill?) that you will gain charge
lore of it.
When using wands and staffs
---------------------------
If you use un-identified wands and staffs, then the game tracks the
number times the item has been used by appending "(used x times)" to the
description in lieu of the actual number of charges.
--
Andrew Sidwell
http://rephial.org/ -- the home of Angband
My email address changes monthly, and is the first three letters of the
month (in English), followed by the last two digits of the current year,
@entai.co.uk.
Antoine
> Well, here's some thoughts I had on how to improve ID in V. I know
> Eddie has other ideas on this... It may not be radical enough, so I
> appreciate all feedback.
>
> Problem: Current ID is tedious.
> Proposed system:
>
> There are two kinds of knowledge which apply globally to all items of a
> given kind:
> * Kind lore: this means you recognise the kind of item something is;
> e.g. kind of ego-item, kind of wand. At birth, a character has kind
> lore of randomly chosen items.
I don't see why you'd start with random kind lore (either in terms of
argument from realism or improving gameplay).
I'd rather see it depend on race/class, ie halflings have kind lore of
all mushrooms, elves have kind lore of 'of Elvenkind' egos, etc.
> There are three/four tiers of items:
> * average
> * masterwork -- current "good", they are simply well-made items
I don't like the term 'masterwork', it's too AD&D.
> Average items and items the player has lore of are automatically ID'd.
> Other items are displayed with their status in brackets appended (e.g.
> "(magic)"), until ID'd. Cursed items are unnoticeable until they are
> worn or Identified -- curses are powerful magics, more than capable of
> hiding themselves. (This would go hand-in-hand with removing
> pointlessly cursed items.)
Works for me.
> If you use a piece of masterwork combat gear for a while, you ID it. If
> you use magic combat gear, then you learn its bonuses, but ID is still
> required to figure out what artifact/ego it is.
Again, works for me.
> If you have an artifact currently known as "magic", then after n turns
> (or n/2 turns if it is being worn), its status changes to "special".
Yup
> When using ID
> -------------
> If you identify an ego-item, there is a small chance you will gain kind
> lore of it.
Yup (higher chance for fighter classes, please?)
> Similarly, every time you ID a rechargeable, there is a
> chance (proportional to magic devices skill?) that you will gain charge
> lore of it.
Yup
> When using wands and staffs
> ---------------------------
> If you use un-identified wands and staffs, then the game tracks the
> number times the item has been used by appending "(used x times)" to the
> description in lieu of the actual number of charges.
Yup
A.
Nick
> On Jul 9, 8:44 am, Andrew Sidwell wrote:
> > There are two kinds of knowledge which apply globally to all items of a
> > given kind:
> > * Kind lore: this means you recognise the kind of item something is;
> > e.g. kind of ego-item, kind of wand. At birth, a character has kind
> > lore of randomly chosen items.
>
> I don't see why you'd start with random kind lore (either in terms of
> argument from realism or improving gameplay).
>
> I'd rather see it depend on race/class, ie halflings have kind lore of
> all mushrooms, elves have kind lore of 'of Elvenkind' egos, etc.
I agree with Antoine. In fact, I'd be inclined to micromanage this - priests
should have kind lore of healing potions, for example. So an Elf Priest would
know Armor of Elvenkind, healing potions, staffs of Holiness, etc.
> > There are three/four tiers of items:
> > * average
> > * masterwork -- current "good", they are simply well-made items
>
> I don't like the term 'masterwork', it's too AD&D.
Agreed. I would like to see pluses to hit, to dam and to AC called something
other than encjantments, too. Or that's what I think today, anyhow.
[Snippy snip]
Agreed.
Nick.
--
"There is no safety, and there is no end. The word must be heard in silence;
there must be darkness to see the stars. The dance is always danced above the
hollow place, above the terrible abyss."
- The Farthest Shore, Ursula Le Guin
Antoine
> Average items and items the player has lore of are automatically ID'd.
> Other items are displayed with their status in brackets appended (e.g.
> "(magic)"), until ID'd. Cursed items are unnoticeable until they are
> worn or Identified -- curses are powerful magics, more than capable of
> hiding themselves. (This would go hand-in-hand with removing
> pointlessly cursed items.)
Afterthought - I think cursed items should appear as "magic".
I also think the number of junk items ('potion of confusion' etc)
should be greatly reduced. As recently pointed out by someone, only
noobs fall for these. (And Ironman players, I suppose, but then you
shouldn't be designing for Ironman.)
A.
nebul...@gmail.com
In ironman, this is balanced by their use as poor-man's-ID; e.g.
potions of confusion provide a relatively safe and cheap way to test
if your current kit gives rConf.
And if we're going to treat items with just pluses as non-magical, why
not go whole hog?
* Change "enchant foo" to "improve foo", or replace with armorsmith
and weaponsmith store services
* Items with pluses pseudo as "good" and with minuses as "poor", or
are not distinguished from "average".
* Items with minuses are not automatically sticky-cursed.
* Weapons with a minus and a plus might be generated sometimes.
* There is a "cursed" ego type for weapons that just gives sticky-
cursing and minuses to hit and to dam.
* There is a "cursed" ego type for armours that just gives sticky-
cursing and a minus to AC.
* Many of the cursed ego types including the above two are uncursable
using plain Remove Curse. Only artifacts and some of the deeper/rarer/
worse ego curses require *Remove Curse*.
* Cursed ego items pseudo as "cursed" or "enchanted", with the odds of
"cursed" improving with perception and level and depending on race and
class (paladins do better, priests do best).
* Other ego items pseudo as "enchanted".
* Artifacts maybe pseudo as egos for a while but may be realized to be
"special" or "terrible".
* Pseudo maybe progresses with time possessing the item, especially if
actually using it:
may be good cursed and realized to be bad
{unknown}
{good} {bad}
{enchanted} {cursed}
{special} {terrible}
Each time you'd get a chance to notice cursing and if you passed the
"bad version" would now be displayed if the item was in fact cursed.
The chain stops short of special if it's not actually an artifact of
course. Non-cursed ordinary items with minuses show as {bad} without
having to pass a separate "curse detection" perception check. {broken}
or {iffy} might be used for mixed pluses and minuses on such items in
lieu of {bad}. Egos that you feel are cursed and that require *Remove
Curse* might show as {awful} once some check had been passed; perhaps
the same check that would show {terrible} if it were an artifact.
The odds of sensing a curse should also go up the more powerful the
item is; maybe with certain knowledge for {special} vs. {terrible}.
Christer Nyfalt
> On Jul 9, 8:44 am, Andrew Sidwell <j...@entai.co.uk> wrote:
>> Well, here's some thoughts I had on how to improve ID in V. I know
>> Eddie has other ideas on this... It may not be radical enough, so I
>> appreciate all feedback.
>>
>> Problem: Current ID is tedious.
>> Proposed system:
>>
>> There are two kinds of knowledge which apply globally to all items of a
>> given kind:
>> * Kind lore: this means you recognise the kind of item something is;
>> e.g. kind of ego-item, kind of wand. At birth, a character has kind
>> lore of randomly chosen items.
>
> I don't see why you'd start with random kind lore (either in terms of
> argument from realism or improving gameplay).
>
> I'd rather see it depend on race/class, ie halflings have kind lore of
> all mushrooms, elves have kind lore of 'of Elvenkind' egos, etc.
>
Agree.
--
Christer Nyfalt
andrewdoull
> Well, here's some thoughts I had on how to improve ID in V. I know
> Eddie has other ideas on this... It may not be radical enough, so I
> appreciate all feedback.
Please play with the Unangband id system for a while. Althought I do not
advocate adopting everything there, I have implemented most feasible
permutations of id systems as a part of it.
> Problem: Current ID is tedious.
> Proposed system:
>
> There are two kinds of knowledge which apply globally to all items of a
> given kind:
> * Kind lore: this means you recognise the kind of item something is;
> e.g. kind of ego-item, kind of wand. At birth, a character has kind
> lore of randomly chosen items.
See below, but I believe in general its worthwhile relearning all flavoured
types each game, so I disagree here. In particular, note most suggestions will
default to 'make the class learn about items that correspond to a particular
shop'. In that case, why not just buy the item from the shop. We don't want to
make money less useful in Angband. Its already loses utitily after a certain
point in time.
> There are two kinds of knowledge which apply only to individual items:
> * Bonuses: this means you know the (hit, dam, ac) bonuses of
> un-ID'd gear.
> * Identification: this means you know everything about the item.
I have no problems with the 'charge tracking' proposing. That makes perfect
sense and makes it feasible to use wands without knowing how many charges they
have.
Having said that, its the only proposal I agree with.
How about considering how you can learn about an item and build from there?
Firstly, there's the 'power usage' counter. An item is considered 'power used'
if a player quaffs/zaps/activates it (potions, scrolls, wands etc.) and can see
the result. The 'power usage' counter provides information about what power the
item has.
At the moment, for Angband:
Flavoured items: usage. 1 power usage reveals powers and learns flavour.
Ego items: no 'usage' possible
Artifacts: id required before power usage. 1 power usage reveals powers.
The only difference in Unangband, is you need 10 uses to learn detail
information about the power (range, damage dice etc).
Then there's a 'quality usage' counter. An item is considered 'quality used' if
a player swings the weapon, shoots the bow, uses an item as ammo or throws it.
An armour item is 'quality used' 1/6 of the time the player is hit (there are 6
armour slots) and a ring is 'quality used' 1/2 the time the player swings,
shoots or throws (there are two ring slots).
The quality usage counter provides information about the item bonuses and
corresponds to your master work information.
In Unangband, there are 3 levels of quality usage information:
1. Sensing whether an item is {good}, {very good} or {great}. Very good is
approximately +5 or better. Great is approximately +10 or better.
2. Learn full bonus information, and heavily sense the item.
3. Identify the item through use.
Unangband combines the 'power usage' and 'quality usage' counters as they rarely
overlap.
Then there is 'flag usage'. Flag usage is when you notice that you have an item
flag because you have an unidentified item equipped and affected by a particular
attack, but are unsure which item it is.
Unangband has a complex system of tracking which set of possible items it could
be, and only tells you which item has it when only one possible item could. It
does this by also tracking which item flags you don't have, and which you may
have, but its quite fragile and breaks easily if you don't have the code right.
Eddie Grove has suggested just immediately noticing which item has the power.
I'd go further and reward the player by immediately identifying the item, or
*id*ing it, if its the hidden power of an ego item. The reason being is that it
doesn't happen very often, and this feels like a good reward for the risk of
testing unknown items. e.g. you wield an unknown sword. You learn it increases
wisdom and immediatley identify it as a (Blessed) (+3). Note that in most
instances it is possible for the player to compute the pval of an item
immediately. So make this explicit (IDENT_PVAL).
Now, I disagree with you about what makes identification tedious. The real
problem is that you have lots of items generated which may or may not be better
than items you are currently equipped with. You really only want to use id
spells on a subset of these.
So you want to encourage a system which cuts down your examination to a subset
of all possible items.
I'd propose having all items sensed immediately (either at creation time, or
when you walk over the item, depending on what you prefer).
Have more detailled levels of sensing:
{good}, {very good}, {great} - as above
{excellent}, {superb} - superb is for ego items with hidden powers
If you are uncomfortable with making remove curse useless, replace {excellent}
with {ego item} and superb with {high ego item}, {superb} and {terrible} with
{artifact}.
If you want classed based id, change what is sensed and how. Look at Unangband
for implementation details here.
Have flavours for ego items: I think you are steering towards this, but come out
and explicitly say this. This will be the biggest improvement over the existing
system.
Have the flag usage above, and quality usage.
The process then becomes:
1. Find an unknown ring.
2. Wield it. Notice it's a ring of Speed (+5).
3. Learn the ring of speed flavour.
4. Find another ring of speed.
5. Wield it.
6. Notice its a cursed ring of speed.
7. Remove curse.
8. Destroy.
9. Find another ring of speed.
10. Wield it.
11. Notice its a ring of Speed (+6).
Note that this is all possible currently in Angband, however the game doesn't
record for you the pvals, even though you can deduce them straight away.
Notice you can substitute armour of Elvenkind {good} and armour of Elvenkind
{great} in the above. However, what you really want is amour of Elvenkind
{resist nether} and armour of Elvenkind {resist blindness}. Those require you
have a source of nether and blindness damage.
Wands are a separate story. I believe you need a wand collection like Hengband,
or wand bags like Unangband for these to be truely useful.
Andrew
--
The Roflwtfzomgbbq Quylthulg summons L33t Paladins -more-
ASCII Dreams: http://roguelikedeveloper.blogspot.com
Unangband: http://unangband.blogspot.com
Eddie Grove
> Problem: Current ID is tedious.
> Proposed system:
There you go jumping to solutions before exploring the problem.
I think it makes sense to examine the rationale behind ID and
rethink implicit assumptions before moving on to solutions.
Also, it is vital to differentiate between weapons/armor and flavored objects.
I'll assume we are talking about weapons/armor and maybe rings/amulets.
You've proposed a huge change elsewhere about curses. That might change the
whole environment. If there are objects that, when you test them, could end
your game, then you need one kind of ID. If you decide not to include such
objects, you can take an entirely different route.
The first issue you have to address is whether you wish to encourage testing
items or discourage testing items. I think you want to encourage, but I
haven't seen a definitive statement that e.g. things like weapons of morgul
and potions of death will be eliminated 100%.
To me, the history looks like the following. ID is vital. So lets design the
game around different amounts of ID for different classes, and make all sorts
of little tweaks. One could take a contrary viewpoint, that ID is only for
players unwilling to do testing, and then you come to very different
solutions.
I think you need to decide what is noticeable about a weapon
(1) from a distance
(2) when you pick it up
(3) when you wield it
(4) when you hit with it
Further, there are things you notice over time. Currently (1) is damage dice,
(3) is a small chance at immediate pseudo, and (2)/(3)/(4) over time is
pseudo.
I'd change all this. Again, it matters if you re-work curses. It is tedious
to wield each item individually checking for things, at least in the current
TMJ environment. With re-worked curses [that do not stop you from unwielding]
it is reasonable to automate so that the player is assumed to wield for a
quick test each item picked up, combining (2) and (3) above.
Consider a flaming sword. It might be reasonable to recognize that it is
magic in any of the 4 situations. The only unreasonable situation, IMO,
is the current one where you only notice randomly over time.
I have my opinions about what ought to be in each case, but there is no point
discussing specifics unless you agree that the framework needs discussion.
> There are two kinds of knowledge which apply globally to all items of a
> given kind:
> * Kind lore: this means you recognise the kind of item something is;
> e.g. kind of ego-item, kind of wand. At birth, a character has kind
> lore of randomly chosen items.
I don't like this. Let players learn everything.
> * Charge lore: this means you know how many charges a rechargeable has
> without needing to ID it.
I find this belief that ID is needed for charges very strange. The player can
tell the difference between 0 and 1 charges on a rod. Why not allow him to
tell 0 versus 1 on a wand? [Btw that is already implicit in wand stacking,
but maybe that is just a bug in the current stacking implementation method.]
Why should I need to use ID to make my wands stack? IMO this comes from the
perspective "ID is necessary, so lets make it do all sorts of different
things". I'd rather take the viewpoint "let's make ID as unimportant as
possible".
> There are two kinds of knowledge which apply only to individual items:
> * Bonuses: this means you know the (hit, dam, ac) bonuses of
> un-ID'd gear.
I think you should get this when you wield something, (3) above.
> * Identification: this means you know everything about the item.
> Pseudo-ID
> ---------
> There are three/four tiers of items:
> * average
> * masterwork -- current "good", they are simply well-made items
Surely a masterwork should be immediately apparent when you wield it, and
perhaps from a distance, not needing pseudo-id in the current sense. Or do
you mean something else?
> * magic -- current "everything else"
> * (special)
>
> Average items and items the player has lore of are automatically ID'd.
> Other items are displayed with their status in brackets appended (e.g.
> "(magic)"), until ID'd. Cursed items are unnoticeable until they are
> worn or Identified -- curses are powerful magics, more than capable of
> hiding themselves. (This would go hand-in-hand with removing
> pointlessly cursed items.)
This is the sort of thing I would love to see, but it means that essentially
you are eliminating pseudo entirely. I found when coding stuff related to
this that what stopped my confusion and endless bugs was to come up with a
physical interpretation. Brands and resists and slays are caused by magic
runes built into the objects, and you can see them when you inspect an object.
Thus you know immediately if something is "magic" as per above. If you don't
like that one, pick another, but pick something or else bugs await you.
> If you use a piece of masterwork combat gear for a while, you ID it. If
> you use magic combat gear, then you learn its bonuses, but ID is still
> required to figure out what artifact/ego it is.
I don't see that there should be a difference between masterwork and average.
You wield, you get the +hit/+dam/+AC, and what else is there? I suppose
it could be about my (1) above, seeing the difference at a distance.
Do you plan to eliminate the spells of enchant weapons? I haven't seen that
mentioned yet. Do you really want there to be a difference between a
masterwork(+9,+9) and an average item that is enchanted up to (+9,+9)?
> If you have an artifact currently known as "magic", then after n turns
> (or n/2 turns if it is being worn), its status changes to "special".
This seems bogus to me. Of course, I view "artifact" as just a method of item
generation. I've never understood this excellent vs special distinction.
Feel free to ignore me on this.
> When using ID
> -------------
> If you identify an ego-item, there is a small chance you will gain kind
> lore of it. Similarly, every time you ID a rechargeable, there is a
> chance (proportional to magic devices skill?) that you will gain charge
> lore of it.
The same recharge spell works on ALL wands. Liches drain charges from all
wands indiscriminately. IMO, charges should be the same and obvious on all
wands. I'll try to shut up about this now, will probably fail.
In any case, if you have to have charge lore, I think you should make it
depend on the number of times you use it, not the number of times you ID it.
The idea would be that you notice something changes in the appearance of the
wand as you use it. Why am I still talking about this? Must shut up ...
Eddie
Eddie Grove
> Then there is 'flag usage'. Flag usage is when you notice that you have an item
> flag because you have an unidentified item equipped and affected by a particular
> attack, but are unsure which item it is.
>
> Unangband has a complex system of tracking which set of possible items it could
> be, and only tells you which item has it when only one possible item could. It
> does this by also tracking which item flags you don't have, and which you may
> have, but its quite fragile and breaks easily if you don't have the code right.
>
> Eddie Grove has suggested just immediately noticing which item has the power.
I have coded and played something along these lines, with no ID at all.
I found for playability, and for coding, that I much prefer the model that
when you are hit by a fire attack you learn immediately for each item wielded
whether or not it provides fire resistance. I use the physical interpretation
that items vibrate or glow when they actively resist attacks.
> I'd go further and reward the player by immediately identifying the item, or
> *id*ing it, if its the hidden power of an ego item. The reason being is that it
> doesn't happen very often, and this feels like a good reward for the risk of
> testing unknown items. e.g. you wield an unknown sword. You learn it increases
> wisdom and immediatley identify it as a (Blessed) (+3). Note that in most
> instances it is possible for the player to compute the pval of an item
> immediately. So make this explicit (IDENT_PVAL).
This might have been where I had a problem with Un. It might not be Blessed,
it might also be Holy Avenger, e.g. IMO, you need to test for slayEvil
e.g. before you know which one it is. Of course, I tried out Un a long long
time ago and things have surely improved since then.
> The process then becomes:
>
> 1. Find an unknown ring.
> 2. Wield it. Notice it's a ring of Speed (+5).
> 3. Learn the ring of speed flavour.
> 4. Find another ring of speed.
> 5. Wield it.
> 6. Notice its a cursed ring of speed.
> 7. Remove curse.
> 8. Destroy.
> 9. Find another ring of speed.
> 10. Wield it.
> 11. Notice its a ring of Speed (+6).
>
> Note that this is all possible currently in Angband, however the game doesn't
> record for you the pvals, even though you can deduce them straight away.
>
> Notice you can substitute armour of Elvenkind {good} and armour of Elvenkind
> {great} in the above. However, what you really want is amour of Elvenkind
> {resist nether} and armour of Elvenkind {resist blindness}. Those require you
> have a source of nether and blindness damage.
In my rune-inspired implementation, I decided that the best method was to
assume different runes for each slot. So learning rBlind on a hat would mean
that you immediately recognize any further hat the gives rBlind, but would not
help with Elvenkind{rBlind} until you learn rBlind on the armor slot.
Eddie
andrewdoull
> andrewdoull writes:
> > I'd go further and reward the player by immediately identifying the item, or
> > *id*ing it, if its the hidden power of an ego item. The reason being is that it
> > doesn't happen very often, and this feels like a good reward for the risk of
> > testing unknown items. e.g. you wield an unknown sword. You learn it increases
> > wisdom and immediatley identify it as a (Blessed) (+3). Note that in most
> > instances it is possible for the player to compute the pval of an item
> > immediately. So make this explicit (IDENT_PVAL).
>
> This might have been where I had a problem with Un. It might not be Blessed,
> it might also be Holy Avenger, e.g. IMO, you need to test for slayEvil
> e.g. before you know which one it is. Of course, I tried out Un a long long
> time ago and things have surely improved since then.
Note that the suggestion I made above is not what Un uses. So it can't have been
the problem you had.
In Un, you still have to test for slay evil etc.
Give it another go if you want: lots has changed.
Eddie Grove
> On 2007-07-09 20:02:43, Eddie Grove <eddie...@hotmail.com> wrote:
>
> > andrewdoull writes:
> > > I'd go further and reward the player by immediately identifying the item, or
> > > *id*ing it, if its the hidden power of an ego item. The reason being is that it
> > > doesn't happen very often, and this feels like a good reward for the risk of
> > > testing unknown items. e.g. you wield an unknown sword. You learn it increases
> > > wisdom and immediatley identify it as a (Blessed) (+3). Note that in most
> > > instances it is possible for the player to compute the pval of an item
> > > immediately. So make this explicit (IDENT_PVAL).
> >
> > This might have been where I had a problem with Un. It might not be Blessed,
> > it might also be Holy Avenger, e.g. IMO, you need to test for slayEvil
> > e.g. before you know which one it is. Of course, I tried out Un a long long
> > time ago and things have surely improved since then.
>
> Note that the suggestion I made above is not what Un uses. So it can't have been
> the problem you had.
It may not have been the code, but poorly written doc? I remember something
about learning things about stuff you wore for a while, just because.
> In Un, you still have to test for slay evil etc.
>
> Give it another go if you want: lots has changed.
I really should. As soon as I finish playtesting new V and harranguing that
other Andrew about it, that is. I need to play a spellcaster and face the new
amnesia attack at a minimum.
Eddie
andrewdoull
> andrewdoull writes:
> > Note that the suggestion I made above is not what Un uses. So it can't have been
> > the problem you had.
>
> It may not have been the code, but poorly written doc? I remember something
> about learning things about stuff you wore for a while, just because.
Yes, you do learn some things you wore for a while. Regen hp, Regen mana, slow
digestion, the absence of teleport flag. Stuff that you'd expect to pick up
after wearing an item for a while.
In general, the docs are well out of date.
> > In Un, you still have to test for slay evil etc.
> >
> > Give it another go if you want: lots has changed.
>
> I really should. As soon as I finish playtesting new V and harranguing that
> other Andrew about it, that is. I need to play a spellcaster and face the new
> amnesia attack at a minimum.
I'm actually 'moving on' from where Angband is on the amnesia attack, but
haven't had the oppotunity to implement what I want to do due to having my
laptops stolen recently.
What I'm planning on doing is allowing the caster to be able to cast while blind
or in the darkness, but have a percentage chance of 'forgetting' the spell when
they do so. This 'forgetting' will only be temporary, until the player can
browse or study the book that has the spell in the light.
Amnesia will then block this ability to cast while blind/in darkness. So you'll
need to be both blinded and amnesiac before losing the ability to cast.
Note: confusion at the moment still allows you to cast spells in Unangband.
Being berserk (from berserk strength), or being in a magically silent room stops
you.
Andrew Sidwell
<snip>
> You've proposed a huge change elsewhere about curses. That might change the
> whole environment. If there are objects that, when you test them, could end
> your game, then you need one kind of ID. If you decide not to include such
> objects, you can take an entirely different route.
Quite. I've realised this myself, so I don't think I'll be changing
anything significant about ID until the TMJ problem is substantially
fixed. As a result, I'm not replying to everything you've said here,
but rather filing most of it away for later reconsideration when it
makes sense to think things through again.
> The first issue you have to address is whether you wish to encourage testing
> items or discourage testing items. I think you want to encourage, but I
> haven't seen a definitive statement that e.g. things like weapons of morgul
> and potions of death will be eliminated 100%.
I think testing items is the way to go. The next release will
definitely encourage it more than the current one, since a lot of
pointlessly bad items will be disappearing.
> I think you need to decide what is noticeable about a weapon
> (1) from a distance
> (2) when you pick it up
> (3) when you wield it
> (4) when you hit with it
This makes a lot of sense, and is a much better approach than tweaking
the current system. Expect another post in six months after I've hacked
on object generation and played Un for a bit, that responds to
everything else.
R. Dan Henry
<andre...@gmail.com> wrote:
>Now, I disagree with you about what makes identification tedious. The real
>problem is that you have lots of items generated which may or may not be better
>than items you are currently equipped with. You really only want to use id
>spells on a subset of these.
Well, I certainly agree with this. ID-tedium is closely connected with
TMJ, but it is the very part of TMJ not helped by squelching. And this
is part of TMJ I think is the worst. The arms and armor drops need to be
reduced and (somewhat) improved at the deeper levels.
>So you want to encourage a system which cuts down your examination to a subset
>of all possible items.
>
>I'd propose having all items sensed immediately (either at creation time, or
>when you walk over the item, depending on what you prefer).
I think you should have to at least walk over it to get pseudo-ID. I'd
rather see you have to pick it up (and then you could have a few curse
effects that activate on pick up instead of wielding. "The dagger
screams!" In any case, you ought to have to defeat any monsters in the
way before you get a clue.
>Have more detailled levels of sensing:
>
>{good}, {very good}, {great} - as above
I don't think this degree of distinction is necessary.
>{excellent}, {superb} - superb is for ego items with hidden powers
>
>If you are uncomfortable with making remove curse useless, replace {excellent}
>with {ego item} and superb with {high ego item}, {superb} and {terrible} with
>{artifact}.
{ordinary} no special properties
{enchanted} only +/- and within "normal limits"[*]
limit might be total no more than 20 for weapon, 12 for armor
{powerful} extraordinary bonus/malus or single power ego
or extra damage dice (which I think should be hidden until IDed)
{wondrous} egos with multiple powers
{unique} artifacts
I think this provide enough information to be useful, while vague enough
to retain some sense of mystery.
This would not be radical change, but would probably be more than
adequate if combined with a serious effort to reduce total number of
items. Right now, waiting for pseudo-ID is major source of tedium for
characters without the spell (resting for mana is relatively minor
tedium for those with spell).
I think this should at least be tried before launching any kind of total
revamp of the ID system. With TMJ and pseudo-ID modified, I think
existing ID should work fine.
[*] - The idea here is that only items with a bonus more than you'd
expect to self-enchant the item up to should be {powerful}.
--
R. Dan Henry
danh...@inreach.com
Paul is dead!
R. Dan Henry
wrote:
>The first issue you have to address is whether you wish to encourage testing
>items or discourage testing items. I think you want to encourage, but I
>haven't seen a definitive statement that e.g. things like weapons of morgul
>and potions of death will be eliminated 100%.
Potions of Death could become Potions of Really Close to Death (but with
a better name) and leave you at 0 hit points. If you're desperation
quaffing while under attack, that'll be the end of you, but you should
virtually never find a !Death so early that this would be anything but
the result of your mistakes, so that's okay. Quaffed in momentary
safety, it's a bit scary, but quite survivable.
No hit points! Wait, I have to go negative to die. That was close. Oh,
wait, forgot I was still poisoned. Oops. YASD.
>Why should I need to use ID to make my wands stack?
You shouldn't. Either wands of the same types should stack or they
should not, which is why the whole "label my wands with how many times
I've used them" idea really isn't a good one for Angband. I shouldn't be
choosing to use my wand of fire I've already used twice, instead of one
of the two I haven't used, because they should all be in a nice bundle
being handled for me just as if they were IDed. I should need to worry
about knowledge related to charges when they stop working. And if I
haven't bothered with an ID scroll by then, tough.
"But what if I IDed some wands of
waving-around-like-a-Harry-Potter-clone and find a new, un-IDed wand of
walaHPc? What then? How do they stack?" I hear the masses crying out. Or
not. But in any case, in this case, there are two ways to go. Either the
whole stack becomes unIDed and you don't know how many charges you have
(because 22 + x = YOU DON'T KNOW) or you rationalize a reason to have
the whole stack IDed ("Since identifying the wand stack puts your aura
in a harmonic sympathy with the wand stack, when you add to it, you
instinctively sense the magnitude shift in the power of the wand
gestalt, thereby learning the new number of charges available.")
Or if you want a third, extremely clumsy option, you could display
things as "g)5 Wands of Tickling (19 charges + 2 unknown)". I strongly
recommend not doing this.
>The same recharge spell works on ALL wands. Liches drain charges from all
>wands indiscriminately. IMO, charges should be the same and obvious on all
>wands. I'll try to shut up about this now, will probably fail.
Sure. You just check the LED display on the side of the wand.
>In any case, if you have to have charge lore, I think you should make it
>depend on the number of times you use it, not the number of times you ID it.
By charging gives advantage to mages (recharge spell) without special
casing and it means no advantage to many-charged items versus items with
few charges. Actually, I'd think it'd be easiest to get a feel for
charges by running the item down to empty, recharging, then running down
to empty again to learn how big that charge really was.
But now we are getting into "realism arguments about magic" territory
and that way lies madness.
R. Dan Henry
<antoine....@gmail.com> wrote:
>On Jul 9, 8:44 am, Andrew Sidwell <j...@entai.co.uk> wrote:
>> There are two kinds of knowledge which apply globally to all items of a
>> given kind:
>> * Kind lore: this means you recognise the kind of item something is;
>> e.g. kind of ego-item, kind of wand. At birth, a character has kind
>> lore of randomly chosen items.
>
>I don't see why you'd start with random kind lore (either in terms of
>argument from realism or improving gameplay).
>
>I'd rather see it depend on race/class, ie halflings have kind lore of
>all mushrooms, elves have kind lore of 'of Elvenkind' egos, etc.
I strongly agree. If you hand out free ID, it should be used to increase
the variation between races (I wouldn't bother with classes; it's the
races that need some distinction).
>> There are three/four tiers of items:
>> * average
>> * masterwork -- current "good", they are simply well-made items
>
>I don't like the term 'masterwork', it's too AD&D.
I agree, although much less strongly. It isn't like D&D invented the
term, but (a) it is now associated with D&D and (b) something like
Narsil or the Ruling Ring is a *real* masterwork, the term is wasted on
a +3,+2 dagger.
Timo Pietilä
> Well, here's some thoughts I had on how to improve ID in V. I know
> Eddie has other ideas on this... It may not be radical enough, so I
> appreciate all feedback.
>
> Problem: Current ID is tedious.
> Proposed system:
>
> There are two kinds of knowledge which apply globally to all items of a
> given kind:
> * Kind lore: this means you recognise the kind of item something is;
> e.g. kind of ego-item, kind of wand. At birth, a character has kind
> lore of randomly chosen items.
Make this class specific. Warriors have knowledge of weapons and armor,
Priest would recognize blessed/cursed, Ranger missile-weapons etc.
> * Charge lore: this means you know how many charges a rechargeable has
> without needing to ID it.
This would be mage only. Mage could also recognize details about
activations more easily than other classes.
> There are two kinds of knowledge which apply only to individual items:
> * Bonuses: this means you know the (hit, dam, ac) bonuses of
> un-ID'd gear.
> * Identification: this means you know everything about the item.
>
>
> Pseudo-ID
> ---------
> There are three/four tiers of items:
> * average
> * masterwork -- current "good", they are simply well-made items
I don't like word "masterwork". "Enchanted" would be better.
> * magic -- current "everything else"
> * (special)
>
> Average items and items the player has lore of are automatically ID'd.
> Other items are displayed with their status in brackets appended (e.g.
> "(magic)"), until ID'd. Cursed items are unnoticeable until they are
> worn or Identified -- curses are powerful magics, more than capable of
> hiding themselves. (This would go hand-in-hand with removing
> pointlessly cursed items.)
eg. cursed ring of strength = ring of weakness. I like that.
Curses should be noticeable if you have it in your inventory long
enough, but in same time cursed items should not be something you just
have to refuse to use. Mixed blessing items.
> If you use a piece of masterwork combat gear for a while, you ID it. If
> you use magic combat gear, then you learn its bonuses, but ID is still
> required to figure out what artifact/ego it is.
I would like to have possibility to get lore-info about resists, slays
and brands if you use item, so that while being non-identified you gain
knowledge of those without ID. Ringil would show cold brand if you have
lore-info about cold brand without you knowing that it is Ringil.
Resists could have 100% chance to get identified divided by items if you
have several items that might possible have that resist. So 100% ID if
you have only one. Once you know about one item then rest is still same
chance so that five {magic} -items + one with full ID would be 100/6 %
~17% not 100/5=20% (unless you know that this full ID item _doesn't_
have that resist). And getting kind lore is some % when item gets
recognized.
This is basically just automatized testing-method that you can use right
now to ID things if you don't have ID.
Apply same for abilities.
You get to see what you know about item by "I"dentifying or from
character info second screen.
Eventually you would get full ID without ID.
This would allow "random egos", because that lowers need of *ID* a lot,
and removes need of ID almost completely. If not completely.
Timo Pietilä
andrewdoull
> On Mon, 9 Jul 2007 09:12:16 +0000 (UTC), andrewdoull
> wrote:
> >Have more detailled levels of sensing:
> >
> >{good}, {very good}, {great} - as above
>
> I don't think this degree of distinction is necessary.
My thinking here was to help warriors and other characters with not so good id
be able to distinguish several levels of item quality early in the game.
e.g. upgrading item slots from {average} to {good} to {very good} is a useful
strategy, even if you don't know whether {good} is +1 or +5, and {very good} is
+6 or +10.
R. Dan Henry
<andre...@gmail.com> wrote:
>On 2007-07-13 09:26:58, R. Dan Henry <danh...@inreach.com> wrote:
>
>> On Mon, 9 Jul 2007 09:12:16 +0000 (UTC), andrewdoull
>> wrote:
>> >Have more detailled levels of sensing:
>> >
>> >{good}, {very good}, {great} - as above
>>
>> I don't think this degree of distinction is necessary.
>
>My thinking here was to help warriors and other characters with not so good id
>be able to distinguish several levels of item quality early in the game.
>
>e.g. upgrading item slots from {average} to {good} to {very good} is a useful
>strategy, even if you don't know whether {good} is +1 or +5, and {very good} is
>+6 or +10.
In the early game, it's not troublesome to use an ID scroll on a {good}
weapon. And warriors find out that it is a {good} weapon reasonably
quickly even now. It's more of a hassle to rangers, who have to wait a
rather long time, so are more likely to just carry huge amounts of ID so
they can check weapons and armor. But if pseudo-ID doesn't involve a
bunch of waiting, you can just ID stuff at an interesting level. And I
don't think there's anything wrong with ID as such. I think the tedium
problems lie with TMJ and pseudo-ID being based on a
rest-a-long-time-to-find-things-out model.
--
R. Dan Henry
danh...@inreach.com
Holy Avenger should be a Paladin title,
not an ego item.
Christophe Cavalaria
> In the early game, it's not troublesome to use an ID scroll on a {good}
> weapon. And warriors find out that it is a {good} weapon reasonably
> quickly even now. It's more of a hassle to rangers, who have to wait a
> rather long time, so are more likely to just carry huge amounts of ID so
> they can check weapons and armor. But if pseudo-ID doesn't involve a
> bunch of waiting, you can just ID stuff at an interesting level. And I
> don't think there's anything wrong with ID as such. I think the tedium
> problems lie with TMJ and pseudo-ID being based on a
> rest-a-long-time-to-find-things-out model.
I'll have to agree with that. Judging from another commercial roguelikish
game (Titan Quest) I've played recently, let me tell you that there isn't
that much junk in Angband compared to that one. Why it feels like a problem
in Angband and not in TQ is simple: in the later, all items are instantly
fully IDed even on the ground, and you can check the name of all items on
the ground with a simple press of a button.
Except for some uncommon green items, all others are either a named item (an
artifact except you can get as many of them as you want) or a magic item
with magic properties tied to the name (Ring of endowement => around +50 to
offensive ability => useless to a mage or archer) The amount of junk
generated in the game is incredible. A simple chest like one can find
everywhere will often drop 3-4 magical items at once but at best there's
one item in 100 you might consider using. All that is made "relatively"
painless because it's dead easy to sort through them.
Also, the game makes sure not to drop the most useless or obsolete items if
possible. Get far enouth in the game and you'll never again drop the lesser
potions of healing for example. And all weapons and armor dropped are at
least of a power level corresponding to the player (the enchantment itself
can be anything though)
Phil Cartwright
> "But what if I IDed some wands of
> waving-around-like-a-Harry-Potter-clone and find a new, un-IDed wand of
> walaHPc? What then? How do they stack?" I hear the masses crying out. Or
> not. But in any case, in this case, there are two ways to go. Either the
> whole stack becomes unIDed and you don't know how many charges you have
> (because 22 + x = YOU DON'T KNOW)
Sangband does this.
--
There's only four things you can be certain of: taxes, change, spam, and
death.
Phil Cartwright
> eg. cursed ring of strength = ring of weakness. I like that.
Sangband does this, too.
magnate
> R. Dan Henry wrote:
> > "But what if I IDed some wands of
> > waving-around-like-a-Harry-Potter-clone and find a new, un-IDed wand of
> > walaHPc? What then? How do they stack?" I hear the masses crying out. Or
> > not. But in any case, in this case, there are two ways to go. Either the
> > whole stack becomes unIDed and you don't know how many charges you have
> > (because 22 + x = YOU DON'T KNOW)
>
> Sangband does this.
... and it's extremely annoying. IMO you should be able to tell the
number of charges in any wand with whose type you are familiar.
CC