36 replies to this topic

[ThetaOrion]

I know that I’m the new kid on the block, sort of an outsider, and thus what I have to say doesn’t really matter.

But, there are a couple of things that are bothering me, so I will say them, even if they are unpopular, unkosher, or not politically correct.

From what I can tell, the overall design philosophy of the Dawn of Skirmish is to ‘do nothing to change the original balance’ because ‘we are going to wait for eternity for Relic to rebalance the thing.’ As a design philosophy, it is fine, if you were to stick with it consistently.

The problem is that this design philosophy or mission statement is routinely abandoned whenever the designers and coders want to abandon it. It’s totally against the rules to tweak the Orks and make them better temporarily until the next patch comes out, but it is perfectly fine to delete Ubers from the Standard setting and create dead-ends in the research tree, for example. (And, I have other examples as well.)

1) The original 1.41 balance had the AI producing Ubers, so if you remove Ubers, it creates an imbalance or changes the original balance and creates a noticeable exception to the original mission statement of not changing the original balance.

2) The original 1.41 balance had the AI producing Rhinos, Chimeras, WarTruks, Vypers, Chaos Horror Demons, and SM landspeeder, yet these seem to come and go in the AI Mod. And whenever they come and go, it constitutes a change in the original balance of the game. The SM Rhino doesn’t shoot, but a Chimera has a weapon and is hard to kill, but the original balance of the gameplay was built on the idea of the AI producing at least one of these items while climbing up the research tree. Some humans have even tried a SM Landspeeder spam or Chimera spam as a strategy.

3) If you start removing these things, then you are changing the original balance of the game, whether you believe it or not. Why is it okay to start removing stuff from the AI mod? Why isn’t that considered a change of balance, but tweaking the Orks in the final release is considered changing the balance and against the rules? It doesn’t make sense to an outsider.

4) The mission statement is inconsistently applied. You can’t double the size of a Tank Busta Squad or double the speed of the Ork Killa Kan, yet you can remove stuff from the original balance or the original production tree with no qualms. It doesn’t make sense to an outsider watching the process.

5) Finally, people will eventually notice that the AI in 1.41 was laying mines, and that the AI in 1.60 of the Skirmish Mod was laying mines, but that the possibility of an infinite defense for the AI has been removed in 1.70 of the AI Skirmish Mod. Removing mines as an option is a blatant change in the original balance of the game, especially to a faction like the IG that depends upon the mines as part of its original balance or original defense strategy. Mines should be permitted when all else has been done, especially if the mission statement is to maintain Relic’s original balance at all costs. Removing mines as an option for the AI creates a big imbalance in the game, particularly if the whole goal is to maintain the original balance of the game. Oh my gosh! He mentioned mines again! That’s totally against the rules of sanity and acceptability. But trust me some people will eventually notice that the AI no longer produces mines. Such a removal changes the original balance of the game, especially for the IG, and thus goes against the original mission statement of the mod, as I see it.

--

They mess around with and destroy the original balance of the game all the time, whenever they want to do so, they just don’t label it as destruction of the original balance. They call it improving the AI, so that it’s kosher, but in actuality, it changes the ultimate balance or the final balance of the gameplay. And, when the potential users start wanting tweaks to the original balance then they are called broken records and given the implication that they should shut up and go away. However in truth, removing mines and creating dead-ends in the research tree and taking tanks and squads off the table IS the same thing as changing the original balance, to an outside observer at least. It affects the final balance of the gameplay, sometimes noticeably.

If the Space Marine AI isn’t permitted to waste resources on LandSpeeders and a single Rhino, no wonders they come across as so indestructible in the final Betas. All the stuff that balances them out and makes them weaker or normal has been removed from 1.70 of the AI mod. Removal of this stuff throws the Skirmish Mod further out of balance. The original balance has been destroyed or taken off the table. And it makes some factions a lot stronger than others, this selective pruning or selective creation of dead-ends in the research and production trees. Consequently, the overriding rule of not changing the original balance is thus inconsistently applied.

I mention it because I don’t think I’m the only one who will notice that the mines are missing or that certain tanks are not being produced. And, just the other day, Thudo was trying to justify the complete removal of the BaneBlade and the Squiggoth. That’s a radical change in the original balance of the game, and to an outsider felt like a dumb move. It had all of us fighting against the idea, his saying that the IG BaneBlade is superfluous and not needed. The original balance of the game was designed in such a way as to make room for the Uber, and trust me, the end user of 1.70 of the AI Mod will notice if the mod is not producing the Titan Uber Avatars.

The development rules or the original mission statement is being inconsistently applied. They have no qualms about removing stuff from the production and research trees, even though such removals radically change the original balance. But, they can’t knowingly improve the balance of the game by giving a single faction a few tweaks in their stats before the final release. It doesn’t make sense to an outsider trying to figure out why things are done or not done.

They can do anything they want to do, but I just thought that I should mention how it looks to some of us on the outside of the decision making process. What you do with the information doesn’t matter. It’s not my mod, I only play in it. There, now I have said it, making me very unpopular. But, hopefully it will open somebody’s eyes so they can see the inconsistencies in the application of the rules, or why the 1.70 AI Mod occasionally feels imbalanced to some of us.

[LarkinVB]

So you want the AI to build useless/weak units for strong races as to balance them ?

The mod tries to maximize performance to make the mod as hard as possible at hard skill level. It doesn't care about balance at all ! Likewise a good player automatching won't build rhinos so his ork opponent gets a chance.

I don't have a clue what "original balance" you are talking about which is violated by the mod. There is no such thing. The game is unbalanced, the mod tries to play each race as good as possible.

There are now some balance/fluff mods which will eventually use the latest skirmish AI tweaked for their needs. I guess you are looking for them.

I accept your oppinion but I completly disagree.

[Zenoth]

I see what Theta means, but I do not agree with everything, although I do agree with some of his points, especially regarding the uber units, and the mines.

Just a few days ago I gave a try to Dawn Of War Pro 1.52, and I must say that I was pleased by the changes. It gave a refresh from 1.41 and its imbalances. But I miss the Imperial Guards, and, without them, DoW Pro, to me, feels incomplete. So I will wait for version 2.0, which, if I am not mistaken, will include the IG.

I think even Arkhan suggested to "move on" to the DoW Pro eventually, with Dawn Of Skirmish.

Well, what I can say about your position on the modification Theta is that even if the whole coders team behind DoS A.I tries to "rebalance" the mod, it wouldn't give the same effect than a real official patch that (let's hope so) has been worked on seriously in terms of balances.

I'm trying to say that, in my humble opinion, Arkhan, Larkin, Thudo and all others involved behind the modification are doing a job that perhaps no others in the community could have done.

I myself am still interested in the video-game aspect of the Dawn of War universe actually, and only because of Dawn of Skirmish, since I do not really enjoy on-line games (with some rare exceptions). To me, the off-line part of it is extremely important. So after finishing the campaigns more than I can count, what matters the most is a great A.I, for a huge amount of replay value via Skirmishes, and trying out new custom maps.

The modification holds my interest of DoW, and, without it, I would have "quit" the community long ago because of boredom.

Hopefully the next expansion set will give a new start to the community and will welcome new members once it is released. And while we all wait for it, I believe that the Dawn Of War Pro mod could well fit a lot of your ideas regarding "how DoW should be". Give it a try Theta, you'll like it I'm pretty sure.

And, well, as a beta tester I simply try to do my job (if I can call it like that), and point out the technical issues I notice. I know some of the other testers including you Theta see your tests with another angle than I do. Looking at the imbalances of the very game-play itself. I myself simply focus on the mechanics of the A.I, stuff that could crash a game, or strange behavior leading to boring game-play (gathering issue anyone ?). But as far as "balance" and game-play elements are concerned, I'll let Larkin/Arkhan/Thudo, Relic, and the guys responsible for DoW Pro deal with that.

[Excedrin]

Discussions about balance are somewhat off topic when discussing the AI "mod". The Skirmish AI isn't really a mod. It doesn't adjust any of the entity statistics or introduce new content or change gameplay. Those are all things that mods do.

The AI in general isn't a mod since two people can play a game with AI allies/opponents even though one person has one AI (perhaps the default Relic AI) and the other has another AI (something custom, like the Skirmish AI). The only reason this isn't readily apparent is because the Skirmish AI is packaged as a mod, with a module file and data directory that contains no changes to any of the typical areas that mods change.

It seems like you need a better idea of what "balance" is all about. Please take the time to at least skim some of these articles:

http://www.gamedev.n...atures/balance/

Then read the articles on Rule Design and Balance here:
http://www.sirlin.net/

http://www.sirlin.net/Features/feature_GameBalancePart1.htm:
Ok, so balancing a game is damned hard. But how to we attempt
it? The first step is designing checks and balances into a game
from the start. Every unit in Starcraft has a strength and a
weakness. Even basic strategies have strengths and weaknesses.
Attacking early (at the expense of a strong economy) beats
expanding early, but loses to defending early (if the defender
can maintain a good economy and survive the attack). Defending
early loses to expanding early. Checks and balances. You could
even say, paper, rock, and scissors.

So what is the AI "mod"? It's a scripted player that attempts to emulate a human player. Whether it plays well or poorly has nothing to do with balance.

[Malkor]

There's units my Starcraft AI won't ever build because it can't use them right, such as Dark Archons, regular Archons... and it's stronger because of that.

You already know I play this mod to face an opponent that is adaptive, fast, and cunning without having to deal with the common idiot on the internet. If you did something to "nerf" the AI as it is, I'd go in and reverse your changes, just as I intend to introduce much, much tougher insane opponents.

Just because some races can't beat others means nothing. My Zerg AI will ALWAYS win in starcraft, because the AI has a natural economical advantage with 8 hatcheries than the terran will until late game with 10 factories, because the units are easier to fight with, ect... that's perfectly fine. No problem, there. I'm not going to suddenly knife the zerg AI's efficiency just because the others can't keep up. In fact, I make the zerg better and better with every change.

Edited by Malkor, 11 February 2006 - 12:49 AM.

[Finaldeath]

Excedrin, as for all classifications I'm aware of, the skirmish AI is most cirtaintly a mod. If you changed the default AI files and went online with a computer player, it'd go out of sync faster then a fast thing going really really fast very quickly.

I agree with reincluding, at least for lesser difficulties (where high cost units are never built) the low-tech utility vehicles.

Landspeeds and Vipers are useful early-tier 2 vehcles, and more then once a few landspeeders as backup (jumping over terrain quickly) help out my space marines , no doubt similar for Vipers and Eldar. They are cheap, much faster to produce then any walker, and can jump.

Horror deamons are left out? awfully odd, they pack a real punch for their cost, and deepstrike. Since the AI usually does have a lot of resources at teir 3, deepstriking and losing units like this won't harm its economy as much as producing heavy vehicles, which can cost a lot more power. Meh, I guess if you use them as chaos, you might actually not need a million defilers to defeat enemy armor

The others: Chimera, Wartrukk (but I saw wartrukks being built in beta 7?) should be built at least on easy and normal difficulties, to compensate for them rarely getting a baneblade or squiggoth.

And Malkor, I'd stop comparing this game to starcraft, which has flying units and different resource systems as major things which changes how it works It's hard to compare apples to oranges, especially since starcraft AI is hardcoded, only the buildorders are not But you can compare I guess, if you really wanted, meh...

Just my take on the situation on unit usage. Mines are, in my opinion, quite nice and cheap, and annoying for stealthed units too. I'll add this into singleplayer AI for the Tau campaign if the skirmish AI doesn't, heh

Oh, and I don't think this adversly affects *balance*, only *competance*. The AI can be much more competant taking advantage of all possibilties, if some are rarely used (say, only in one of the possible build orders/plans).

BTW: If you move the skirmish AI to DOW pro, you'll basically be on an island surrounded by sharks. Someone would take up a new AI project for Winter Assault, and of course all the race mods which run on it.

Edited by Finaldeath, 11 February 2006 - 04:08 AM.

[Malkor]

Starcraft can be compared to DoW, easily. If I want to restrict the battleground to groups of units with a time-based resource system, that's easy. That's kind of how I manage my AI, actually. I consider how their groups of units will face opposing groups of units. I generally try to make them as multi-purpose as possible but you can always find ways to abuse their lack of micro, or they do something completely retarded like try to assemble their force behind a bunch of enemy defenses.

In my eyes this mod is almost complete. Only a few build orders need a bit of playing around with for higher difficulties.

Edited by Malkor, 11 February 2006 - 04:59 AM.

[ThetaOrion]

Maybe the word 'balance' is the wrong word to use, because everybody seems to have a different opinion or a different definition of the word. LarkinVB clearly got the intention of the message wrong, whereas Zenoth seemed to understand my message as I intended.

Maybe the word 'feel' is the word to use. I was talking about the overall feel of the gameplay. In Plain Vanilla 1.41, the Space Marines feel more powerful and more competent than the Orks. In some of the recent Betas, the AI Mod feels a lot more like Plain Vanilla 1.41 than Beta 4 did. Beta 4 was a major departure from the original feel of Plain Vanilla 1.41. In Beta 4, the Chaos Horror Demons weren't showing up, for FinalDeath who asked.

I know this is not a democracy. Thudo does what he wants, even if all of us vote against it. I understand that. Nothing I can do about that, and I don't even want to try. But, the Thudo God lives in his own tower and gets fixated on his own explanations of things, and I wanted to try to tell him that some of us are perceiving things a bit differently.

So, I wanted to make a few points.

1) The removal of units from the production list and dead ends in the research, especially removal of the Uber units, makes the AI mod feel vastly different than the original intentions of the Plain Vanilla AI. The gameplay feels different, a feel that to some of us constitutes a change in the overall balance or the overall feel of the gameplay.

2) Removals and dead ends create confusion in the end user. He finds himself looking for the missing Ubers or the missing mines or the missing LandSpeeder or the missing Chimera and asking, "Why is the AI being hobbled or broken? Why can't the AI do these things? Something must be broken!" Those outside the decision making process start asking why things are missing. And, the questions won't end here in this quiet little forum. Soon, others will be asking too, the general public, when they get their hands on the final product, if these items are missing. They'll want to know why. Some of them will even feel as if part of the game they paid for has been taken away from them. Not everyone is going to like it.

3) The original Plain Vanilla has a feel that leads you to believe that the game is imbalanced, that the Space Marines are superior to the Orks, for example. And, by completely removing LandSpeeders, mines, and Rhinos from the Space Marines in the AI Mod, the original gap between the Space Marines and the Orks becomes even wider and more noticeable in the AI mod. The general feeling that an outsider gets is that Thudo purposefully went out and made the Space Marines even more uber or more imbalanced or more powerful, while at the same time purposefully neglecting the Orks. Hence, the question I originally asked, "Why is it fine or acceptable to remove the weak stuff from the Space Marines and make them even more powerful, whereas at the same time it is taboo or sin or forbidden to tweak the Orks and double the size of some of their squads or double the speed of the Killa Kans for this same release." To an outsider like me, it represents an inconsistency in the policy. It's okay to make the Space Marine AI feel better or feel more powerful in order to make the AI better, but it's not okay to tweak the Orks and make their AI or gameplay feel more powerful or better. It's an inconsistency.

4) I agree, the Rhino is a stupid unit, especially for the AI. But, allowing the SM AI to build a single Rhino helps the AI to keep its original feel, and will keep the general public from asking why the AI doesn't build Rhinos. And, an extra build program where the SM AI goes heavily into the LandSpeeder Route seems perfectly acceptable to me. General public paid to have LandSpeeders in their game, and a complete removal of them in all scenarios or all build programs just creates confusion, inconsistencies, and a greater feel of imbalance in the game.

5) I just wanted Thudo to know that some of us are perceiving these selective treatments as being arbitrary or inconsistent. He needs to know that selective removal of items from the different factions changes the overall feel of the game and is actually perceived as or feels like widening the imbalances that already exist in the original 1.41.

6) To an outsider with no control over the decision process, it just doesn't seem logical that it's fine to remove items from the build programs, but it's not fine to tweak the Orks stats. It seems arbitrary and inconsistent. Now, I'm not an Ork lover either. But, if it is fine to change the feel of the Space Marines and make them feel and be even more uber by removing the weak stuff from their build program, then why isn't it fine to change the Orks in some fashion? The rules for what is allowed and isn't allowed doesn't feel consistent, to some body who is actually playing the game, to this human here who is actually playing the final result. The definition of what is an acceptable change and what is an unacceptable change is totally unclear to some of us.

--

I was going to let all of this go and let Beta 8 and the 1.70 AI Mod pass into history. But, then I saw them all tweaking Beta 8 some more, so I decided to speak up.

Chances are that Thudo won't even take the time to read any of this. That's the way it goes.

But, a lot of this was nagging in the back of my head, and I decided to put up the flag of warning before 1.70 gets written into stone and released to the public. If things are missing from 1.70, Mr. General Public will start noticing and asking why. Yes, some things will go unnoticed. But, if Thudo purposefully kills off the BaneBlade or the Squiggoth in some misguided attempt to make the AI as strong as it can be, then I can pretty much promise that people will notice and some of them will start complaining. And I feel sure that some people will notice that the 1.70 AI doesn't use mines, and they will ask why. In the end, it's that person at the end of the line, the person who downloads and plays the thing that we are trying to please, and not necessarily the person who is trying to get the most perfect AI vs AI samples that we have ever seen. As I see it, we also want to try to mitigate confusion for the end user.

Sometimes I think Thudo's priorities and rules get a bit skewed or that they are applied inconsistently. To an outside observer like me, it can get really confusing at times. I find myself asking what is acceptable and what isn't. What is a kosher request and what is not. What will Thudo listen to and what will he reject outright or won't even bother to read. It's very confusing to some of us, especially those of us who are judging the whole project by the ultimate feel we get while actually playing the game. I don't know the details, I only know the ultimate feel that I get. And if the ultimate feel that the 1.70 AI Mod generates is confusion or questions or a feeling of greater imbalance, I don't know if that is necessarily a good thing.

I just wanted to mention it.

Edited by ThetaOrion, 11 February 2006 - 06:33 AM.

[Malkor]

I did suggest a land speeder rush but it was decided not to add it by the coders. If I could write my Insane scripts, I would definately re-introduce the idea. I do feel that "uber" units DEFINATELY need to be built as I already stated why in the other thread. But for some units that are just not going to be used by the AI in any remote matter, most notably the rhino, it shouldn't really waste resources on it, unless something comes up where the coders can find a way to have the comps use transports.

[Excedrin]

Excedrin, as for all classifications I'm aware of, the skirmish AI is most cirtaintly a mod. If you changed the default AI files and went online with a computer player, it'd go out of sync faster then a fast thing going really really fast very quickly.

That's incorrect. I've watched mirror matches between the Skirmish AI and the Standard AI and posted the replays and results here in the past. Standard AI won, I guess it's time to run that again...

For the most part, the game only cares about stuff in data/attrib with regard to sync errors. Does that open a huge hole for abuse in auto? Yes, obviously. But luckily the population is too small to have technical abusers who exploit it

Now I'm off to watch two AIs fight to the death IN A STEEL CAGE

http://dow.lerp.com/...0211.005730.rec
http://dow.lerp.com/...0211.004212.rec
http://dow.lerp.com/...0211.002519.rec
http://dow.lerp.com/...0211.000606.rec
http://dow.lerp.com/...0210.234053.rec

Alright, the cage match of the century is complete! The Skirmish AI only lost 1 game to the WA AI! These are all replays of mirror matches (one for each race) of 1.65b7 vs the default WA 1.41 AI.

Edited by Excedrin, 11 February 2006 - 09:02 AM.

[LarkinVB]

BTW: If you move the skirmish AI to DOW pro, you'll basically be on an island surrounded by sharks. Someone would take up a new AI project for Winter Assault, and of course all the race mods which run on it.

I will add the AI to DoWpro but I guess the Dawn of Skirmish AI will stay based on Relics latest code.

I haven't played any 1.7 beta yet so I can't comment on details but IF Thudo has removed units like chimera, landspeeder, horrors or any uber I disagree with this decision.
I agree with removal of rhino and mines, this is true since v1.0 of the mod and no one ever complained about that - until now. A mine or two can be usefull if cleverly placed at the right moment but I can't see the AI handling that effectivly.

I don't think anyone has crippled orks on purpose so Thetas observations are either based on a) a bug or b) orks well known weaknesses.
They are weak in WA 1.41 AND the AI is better in using ranged troops.

Weak orks are no reason to build a rhino.

There should be only one agenda for the mod. Will a certain decision make it play better (=more successfull) ??? This decision should be based on a match versus a good human player.

PS: I checked the code and horrors are not removed. Same for vypers and chimera. If Thudo will readd one landspeeder and the ubers (don't know if they are really missing) all should be fine.

PSPS: Theta, do you know that guardians and rangers are also removed apart from the starting squads ? Guess why ?

PSPSPS: I'm very happy to work with the mod and DoWpro. Now I can allow all units except rhino as all can be used now and then.

[Excedrin]

I agree 100% with everything Larkin said.

There should be only one agenda for the mod. Will a certain decision make it play better (=more successfull) ??? This decision should be based on a match versus a good human player.

This point is really key, and thinking about it, which units the AI uses are a very minor point in comparison to the other huge deficiencies in its current play.

Also, many units that are weak now will be improved in the next patch.

[ThetaOrion]

I just saw an Eldar Vyper in Beta 7. It was rather a surprise to see it, but a pleasant surprise nonetheless.

I don't know which of the Ork equipment the 'gurus' consider unacceptable or a waste of resources, but I thought there was one of the trucks that was on the blacklist. Is the Ork Mad Doc being produced and attached in Beta 7? As long as everything Ork is being produced at some point in some build program, including the Squiggoth, that's the direction I think it should be going.

But, I have yet to see a BaneBlade or a Squiggoth in Beta 7, and I have had some big maps and long games. The absence of the Uber is noticed. And, I do personally believe that it would be a big mistake to release 1.70 of the AI Mod with half the Ubers absent.

It makes you start asking what else is missing!

And for the 1.70 final release, mines for the Imperial Guard would be a nice touch. The IG seemed to be designed to take advantage of mines. Mines are not as important to me as the Ubers or the Titan Avatars, but I think some people in the general public will eventually notice that the mines are missing and will complain. Mines should be last in priority on the build schedule, but mines give the AI the chance for an infinite defense if the human is slow to move, so mines shouldn't be taken off the table, especially if the goal is to allow the AI to become as strong as it can be. If mines were removed because the code for mine laying is buggy, that's one thing. But, if they were removed because Thudo doesn't like them, that's another thing, and only serves to increase the feeling of imbalance or selective discrimination.

I don't think anybody will notice that the Rhino is missing, but some will notice that the LandSpeeder is missing. I guess I want Thudo to consider putting back some of what I think some people will notice and miss.

[ThetaOrion]

Theta, do you know that guardians and rangers are also removed apart from the starting squads ? Guess why ?

I actually noticed the missing Eldar Rangers or snipers last night, but I forgot to put them on the list of missing items. They were duds except for their cloaking ability.

On the other hand, a guardian squad attached to the FarSeer Commander is considered by some to be the toughest or strongest opening act of any faction in the game. I too have been surprised at times at how well it can do. But, Banshees attached to the FarSeer is an even better combination, probably the best early rushing combination in the game.

[ThetaOrion]

The cage match up above gave me a new respect for AI vs AI that I never had before. I didn't know that an older AI could be matched up with the latest and greatest.

They should probably have a thread or a seminar where they teach us all how to do cage matches.

With the Final 1.70 Beta, Excedrin should probably run it off against 1.51 of the AI Mod and 1.60 of the AI Mod along with 1.41 Plain Vanilla.

Sure, such tests won't really solve my SM vs Orks gripe about imbalance. I'll have to wait for 1.50 of the Relic Patch to fully scratch that itch.

But, if they keep running the current betas against the older AI's that have been released and keep tweaking and doing so until all of us like the looks of what we are seeing, then that tweaking or refinement of the 1.70 AI would probably indeed eliminate 3/4ths of my disappointments that I was feeling in regards to my AI allies.

The Beta 7 Ork AI allies were such a disappointment that I didn't want that embarrassment being set in stone and presented to the world, so I became a broken record in an attempt to try to get my point across. I really felt as if the Beta 7 Orks would embarrass the whole team if they were presented to the world 'as is'. The Beta 7 Orks were noticeably bad. I was losing because of them and their unwillingess to engage the enemy. It just didn't feel right for a FINAL release.

But, if Excedrin will run some more of these tests with the current Beta against the older AI's, and LarkinVB give them a look, and then they teach some of the rest of us how to make these tests, then maybe we really could fix it up nice before releasing 1.70 to the world.

There was a time when I personally didn't notice much imbalance between the 1.41 Plain Vanilla Space Marines and the 1.41 Orks, because the 1.41 Ork AI was functioning well enough 90 percent of the time so as to make the imbalance unnoticable. If we can get the 1.70 Orks doing the same, then they won't be an embarrassment to the team, and I really shouldn't have any complaints.

[Finaldeath]

That's incorrect. I've watched mirror matches between the Skirmish AI and the Standard AI and posted the replays and results here in the past. Standard AI won, I guess it's time to run that again...

For the most part, the game only cares about stuff in data/attrib with regard to sync errors. Does that open a huge hole for abuse in auto? Yes, obviously. But luckily the population is too small to have technical abusers who exploit it

Fine, it still doesn't stop it being a mod.

if you add the Ai to the WXP or DOW folders, it will make other parts of the game, notably the singleplayer, cease to work correctly. Having it install as a map, as in the actual folders itself, is therefore not a good idea.

Considering the AI should run somehow choosing one data set over another online, I do wonder how you think you could add random code to the AI and not have all clients crash.

I wonder how you got the skirmish and noramal AI fighting, but I guess perhaps you changed the default.ai script or something.

In any case, this is a mod, if you changed the games default files, be it AI or Attributes, it is modding the game. Technically, to not call it a mod what would it be called?

Edited by Finaldeath, 11 February 2006 - 12:57 PM.

[LarkinVB]

I noticed that you need two computers to test with two different AI, one AI type running on each of them. I would like to see 1.5 matched against 1.7beta. Sad enough my second computer misses the necessary 3D card.

I have a tip for you Theta. Condense your posts and provide a replay pointing out obvious and blatant flaws and I'm sure you will make more impact than writing lengthy posts with vague feelings.

Edited by LarkinVB, 11 February 2006 - 01:00 PM.

[Klementh]

Hey Malkor, I still play BroodWar. Where can I find a good AI mod for it?

I tried to use pm but the board didn't let me use that feature.

[Excedrin]

Here's how to have different AIs fight:
http://forums.revora...ndpost&p=179204

Edited by Excedrin, 11 February 2006 - 05:38 PM.

[Malkor]

Hey Malkor, I still play BroodWar. Where can I find a good AI mod for it?

I tried to use pm but the board didn't let me use that feature.

Hi!

There's only one good AI mod for it, and that's RequiemD! That's what I'm making/made. Of course, what I can do is very limited, but it should give you a run for your money.

www.staff.samods.org/iskatumesk/RequiemDvA04A.exe

Running this will launch starcraft and temporarily patch memory for the duration. Note that if you play with buddies over battle.net they'll need it as well. To return to normal starcraft, simply run starcraft normally.