Changes needed to Patrols
Davin
effectively as they were supposed to. However, Jon has pointed out
that they now work TOO well, and that Scouts don't have much of a
chance of getting away in a moderately-patrolled system. Being able
to Scout a defended system (with reasonable effort) is an important
part of the game design, so it looks like we need to find a way for
large numbers of patrolling ships to not be so overwhelmingly
effective against Scouts.
Jon's original plan for Patrols called for there to be eight "sectors"
around the system and that each Patrolling ship would be assigned to
patrol only one of those eight sectors. Likewise, each Scout would
only appear in one of those sectors, and thus only Patrols that had
been assigned to the Scout's sector would be eligible to catch them.
It appears that this design never made it into the code or the manual,
but was Jon's original intention.
However, this severely limits the ability of Patrols to catch Scouts.
At best, you have to put up eight times as many Patrols just to have
the same chance of catching any given Scout, and that could amount to
hundreds of ships per system. This might even be possible in Home
Worlds and Production Centers, but it sounds to me like it would
virtually eliminate all Patrols and Sentries in Colonies, because you
couldn't reasonably field enough Patrols to have much of a chance of
catching Scouts peeking into Colonies. So this wouldn't be my
favorite choice, at least for Colonies.
We could also reduce the percentage chance for each Patrol to catch a
Scout, down from 1% per Impulse engine to whatever seems appropriate.
Or we could use fewer "sectors" (4 or 6, for instance). But I don't
think that these alternate mechanisms give us the kind of patrolling
success ratio that we want.
I think Jon's main issue is with having very high levels of Patrol
coverage to keep out virtually all Scouts. My major complaint with
this is that reducing the top end of Patrol coverage also makes the
low end of coverage TOO weak (making Patrols nearly useless). So, how
might we allow for a reasonable amount of Patrol coverage with, say, a
Jeep w/4 fighters? We might decide that a Jeep should end up with
maybe a 25%-50% chance to catch a Scout. But we would still want to
limit Patrols' abilities to something that would still allow a Scout
to get in even if we had a dozen of these loaded Jeeps - maybe that
would only give us about an 80%-90% chance, so that a Scout would
still have a 10%+ chance to escape notice, even in such a heavily-
Patrolled system. I've also seen that some players like to put up
Platforms with nothing but dozens of Fighters in them. To satisfy a
need for Scouts to sometimes get away, these and other similar
scenarios would also need to be restricted.
At the moment, the programming for catching Scouts is very simple.
Each Patrol rolls percentile dice against his own Impulse engine count
for each Scout in the system. That's it. Simply having each ship-to-
ship combination rolling separate dice causes the composite
probability effect that more Patrols give better coverage, approaching
99%-100% for large numbers of Patrols. This is not a programmed-in
calculation but simply a resultant effect of applying a simple rule to
all the ships in the system. So it's not a simple matter to apply a
"cap" to the percentage chance, even if we decide that a fixed "cap"
is what we want.
So, how might we implement a patrolling restriction without being TOO
restrictive? The only thing I've come up with so far is to limit the
number of patrolling ships that can be effectively used (say, to a
dozen). But that limit seems awfully arbitrary to me and doesn't
allow someone to spend a ton of time and money putting up a better
"net", so it's not my favorite idea. Maybe we can come up with some
sort of sliding scale where individual Patrol effectiveness is reduced
by the number of Patrols in the system?
I had a thought that we might add the Scouts' Impulse engines back
into the equation... What if we used the Scouts' engine count to
reduce part of the effectiveness of the Patrols' engines? But that
sounds like it's just the equivalent of making Scouts cost slightly
more and reducing the percent chance of Patrols overall. It doesn't
sound like a balanced solution. It also doesn't sound like using more
engines to escape detection is quite in keeping of the scout's concept
of "looking like a rock". But that's the kind of thinking that I'm
looking for - some "other way of thinking" to make Scouts more
feasible while keeping some Patrol effectiveness.
So, who's got some ideas? I'd like some input from beginner players
as well as experienced ones, because I expect their viewpoints to be
radically different.
Come on, speak up!!
Robert White
I really don't have any mechanical input to put forward at the moment, but as an alternative 'flavor' to the scouts concept, I've always liked the image from some book I read (title currently eludes me). A ship's sensors detect primarily energy signatures, ie: engine power, weapons systems, shields, etc. If you turn all of these things off, a ship becomes effectively 'invisible'. Of course, this invisibility comes with the associated costs of having everything turned off... including life support. But with everything off, an object that was initially traveling at a very great speed in a given direction will, by the laws of physics, continue to travel at a similar speed in a similar direction. Not precisely a 'rock', and something closer to the 'stealth' mechanic that I've heard fellow 'newbies' question when talking about the Scout
command. Most of the mechanical interpretations I'm pulling from my own flavor seem to mostly just lead to needless complications to the system.. but I do feel that a 100% 'catch' concept is a bit unreasonable, and perhaps also unfair. Space is an awfully big place, even when you can jump from star to star. One little piece of 'debris' floating by is just as easy to overlook as any other piece of space-junk. If it happens to be manned, and keeping a runny tally of all your gunships.. You might never know. I think a 100% catch chance should be POSSIBLE... but should be very very cost-prohibitive. Something that, even at tech 6 and half the galaxy worth of income, you'd only consider viable at your homeworld and perhaps a single production center. But at the same time, I'd like to be able to have a feasable 25-40%-ish catch rate on my colonies, should I decide to throw a few patrols their way. |
Davin Church
a new perspective.
I think your book's idea makes a fair bit of sense from a realism
point of view, but we've abstracted Galac-Tac away from too much of
that sort of thing. If we try to bring it down to that much detail,
I'm afraid we'll lost much of the rest of the game rules due to
"reasonableness", not to mention making Scouts unstoppable. Don't
forget, though, that Patrols may see a signature of you "warping in"
to the system and know that something has happened and in what
general area -- that should conceptually improve the odds of finding
someone, even with their systems "off".
I agree with your assessment of "catch rates" in general, but I don't
know how to make them "effective, but not too effective". We need
both Patrols and Scouts to work reasonably well for the game to be playable.
Davin Church
work for overall game play, but let's discuss this a bit and see if
we can come up with pros and cons and how it might affect game play
with different players' styles.
The basic concept is to add ECM and ECCM to the game system. Since
only those of you with military backgrounds are likely to know what
that is, let me 'splain a little. ECM is short for Electronic
Counter Measures and would be used by scouts to help avoid
detection. This would be something like a classic sci-fi "cloaking
device". ECCM is short for Electronic Counter-Counter Measures and
would be used by patrols to help find scouts, especially those using
ECM to try to hide. This would be like special radars that are tuned
to look for minimal signatures (things that are trying to look like
rocks) and also designed to penetrate active "cloaks".
We wouldn't change the way that current scouts and patrols work, so
existing ships would continue to behave as they have been. But when
your scouts need to be extra-sneaky, or if you're trying to catch
those extra-sneaky scouts, then you can design new ships just for
that purpose. Make sense?
Of course, this raises an ugly question ... what do we do about ship
design rating codes to accommodate these new features? I don't
really want to change the rating system - it's complicated enough
already. So let me suggest this... since shields are only used
during combat, what if we used those same shield generators during
scouting/patrolling to create the electronic fields and power the
systems that provide for ECM/ECCM? For every shield generator you
have, you get an ECM bonus when you're scouting or an ECCM bonus when
you're patrolling. Just how much of a bonus you get for each shield
generator still needs to be discussed, but let's start by considering
a 1% to 3% bonus per shield generator.
Assuming we go with the above plan, what does that do for (and to) us
in the game system? First of all, you can still build your cheap
//29-1 flies/gnats to scout people's colonies that don't have up any
patrols and get away scot free. And building several fighters or
even little 1P//-20 patrol boats to patrol your production centers
will continue to work just fine against those same flies/gnats. But
if you want to add ECM/ECCM to the equation, adding shield generators
to those ships are going to make them much heavier (SSD-wise), and
that means they're going to cost a lot more. So you're not going to
want to do that to ALL of your scouts/patrols -- just the ones where
it's absolutely necessary. Of course, you can get a BIG bonus by
really loading up on shields (even making them battleship-sized
scouts and patrols), but that makes large numbers of them
prohibitively expensive.
The way I have the math figured is that this ECM/ECCM scheme strongly
favors the scouts. Each patrol ship gets only its own bonus against
a lone scout, but that scout's bonus works against ALL the patrol
ships simultaneously. Therefore, if you have one heavily beefed-up
scout then it takes a whole flock of heavily beefed-up patrols to
stand much of a chance of catching him, and that's so expensive that
as a practical matter nobody's going to be able to afford
it. Therefore, if ECM gives you a decent-sized bonus then you can
build just one giant, expensive scout (/30;;/12-1) and go anywhere
with it with practical impunity - nobody's going to be able to catch
it. This really worries me, as that could tip the scales too far in
the other direction and make patrols practically worthless! So we
need to come up with some kind of balance where scouts AND patrols
are both effective and affordable.
So who's got comments? What do you think of the idea in
general? And how might you suggest tweaking the numbers to keep
scouts and patrols balanced?
Davin Church
should) go in that direction. We still have the option of leaving
scouting just like it is and adjusting the percentage chances as
desired (effectively dividing things into quadrants, or the
like). Maybe we can even get rid of the individual ship-to-ship
calculations (that are limiting how we calculate things) and just
artificially lump everything into one big batch and come up with some
imaginary math formula to apply to the group as a whole.
However, for myself, I don't see much of a need to adjust things at
all. Jon is strenuously opposed to being able to set up 95%+ patrol
coverage in your production centers. For myself, I've always thought
it was that way all along - that everyone (with experience) always
set up heavy patrol protections in their PCs. In all the years that
I played, I seldom was able to get off a scout in a protected enemy
PC. So I just EXPECTED to not be able to scout the big systems to
find out how much protection was there. I just had to guess at how
much they had and bring enough to do what I hoped was a sufficient
job. But to me, that's part of the game, too. You should never be
able to tell exactly what the other guy's got there ... that
partially represents the "the fog of war" concept, too, not to
mention "no battle plan in history has ever survived the first
encounter with the enemy." So if you're going to start a big battle,
I don't feel like "knowing everything" is necessarily an option, much
less a requirement. Jon disagrees, and thinks he always ought to be
able to glean all that information in order to mount an effective
attack. Granted, always being able to tell what's waiting for you
means that you're never going to start a fight that you're not going
to win hands-down. But surely it's not fun to never being at risk of
losing at battle, is it? And why bother defending anything at all if
it's always going to lose? That seems counter-productive to me from
both sides of combat, and removes whole aspects of play from the
game. (And don't forget that everyone else would get to scout YOUR
home world unmolested, too.)
So, what do y'all think about the problem in general? Please ignore
the idea of ever getting complete 100% coverage, but consider the
concept of building enough patrolling ships (however [reasonably]
many we decide that should be) to provide 90%-95% coverage against
scouts in your home world and/or production centers. Is that very
idea in itself a game-breaking scenario to you, or is "not knowing
everything" part of the fun and challenge of the game? SHOULD we
even be trying to change that aspect of the game in the first place??
Davin
Robert White
Just to throw yet another variable in the mix.. I think an interesting twist on this issue might be to have the number of patrol ships have an impact on the accuracy of scouting. This would be in addition to a (perhaps drastically reduced) chance to catch the scout out-right. If they're spending all their time hiding from patrols, they'll only really get a good count of said patrols, and not get to look too closely in-system. I don't know what kinda math should be thrown around with that kinda idea, but it seems to balance out a little in my mind. Scouts are much less likely to be caught outright, thus making scouting a system worthwhile.. but their reports come back with a margin of error (maybe as large as 150%.. or 50%...) that prevents certainty of victory without bringing truly overwhelming odds to the table, ensuring patrolling is worthwhile. This could also be modified by your ECCMs in-system, but again, the precise math would probably be better left to other minds. I'm just babbling psycho-nonsense and seeing if something sticks. |
pbe...@yahoo.com
Other than that, patrols and sentrys do not seem cost effective or even effective.
Also, I think scouts should have an easy time escaping and a hard time scouting something that is occupied or has hostile ships there. Space is big - a scout could jump in very far from the sun and try to be a rock while creeping in to get more information, Sensors on both sides would constantly be playing a dueling ECM kinda game - think like perhaps those WW2 submarine movies, where the sub is trying to get in close before being detected, deliver the payload, and get out before being destroyed.
Dunno if it is practical, but scouting might be considered into three phases -
1. Initial contact - as the scout comes into system and tries to get started scouting
2. Information gathering - as the scout tries to avoid detection and gather information
3. Escape - as a scout either tries to leave of its own accord or escape after being detected.
Example - Captain Beuller of the "Rock of Ages" gently pushes his ship into a long range eleptical orbit around Alpha-Pandonia 6. Being this far out serves him well to protect his ship from any unwanted contact with local patrols, but now he faces the task of manuevering close enough to the system proper to gather useable intelligence. Captain Bueller waits three days until it is safe to do so, then makes a course correction to fade into the system. At the same time, the scout deploys it's famous "I am a rock" disguise and makes sure that all emissions are quieted - nobody even uses a space oven while they make their scouting pass. Bueler's plan is that they will pose as a small, break-away asteroid and travel slowly through the system, passively collecting data until they pass onto the far side of the sun out of range of patrols and are able to slip away, unnoticed...
Davin
So, at least on this topic, I think it's already doing what you're suggesting, unless I'm misunderstanding what you're getting at. Worse, when you abort an attack, not only do you not get a scouting report on them, THEY get an automatic scouting report on YOU, so they can tell how big a fleet you just tried to bring in.
For intelligence purposes, the only reasonable thing I see to do is to send in a few dozen scout ships, all with Attack orders of different sizes, and see which ones die and which ones run away. Of course, that only gets you very rough information and also informs the enemy of exactly what you're up to, so he'll either get mad and kill you or start changing his defenses every turn to keep you from seeing anything about what's really there.
Davin
Davin
They don't need to be expensive, either. You may put large warships on Patrol duty, but I rarely do. I build tiny little Patrol Boats (1P//-20) and put a half-dozen to a dozen of them on Patrol in each PC. They only cost 6 PI each, and six will give a 74% chance of catching each Scout (they'll only have a 26% chance of getting their information). A dozen will increase those odds to 93% (7% chance of scouting) and only costs a total of 72 PI. That's a pretty good return on investment to keep them from knowing just what to bring in against you that will win the battle.
Once a Scout is caught, the entire defensive forces of the system are brought to bear to destroy them, so the Patrolling ships don't need to have much going for them at all -- Inertia drives are all that's important.
When this issue of S&D comes out, check out the article I just submitted on the combat commands for more information.
Davin
On Thursday, August 21, 2014 12:03:11 AM UTC-5, Eric the Panda wrote:
Davin
On Thursday, August 21, 2014 12:03:11 AM UTC-5, Eric the Panda wrote: