pzgndr
-
Posts
2,255 -
Joined
-
Last visited
Posts posted by pzgndr
-
-
Italy does surrender once Allies land several units on the mainland. Germany needs to be prepared for this and have units in Italy once Allies start to make landings in Sicily.
What difficulty settings are you using? I tried to shoot for balance at the Int +1 level, but I'm finding I need Expert +1.5 for challenging game. My latest update in progress has a few things to toughen it up some more. I thought I may have it out by now but I'm want to review my changes some more. Maybe by end of the month.
-
My assertions were never outrageous or incorrect and I challenge you to prove they were.
Just one, then we're done. You asserted without basis that continued AI development to implement full FOW in EOS was trivial. Knowing better, I stated that it would involve hundreds or thousands of hours of additional effort to accomplish what you wanted. Lots of time to recode the program, debug any issues that arose, playtest to verify AI performance was actually enhanced by the changes and not made worse, make necessary adjustments, etc. You disagreed. Brit finally responded to your call out and told you pretty much exactly what I had told you. I stated facts based on personal experience actually programming code and managing software development projects; you respond that it's all BS based on zero knowledge of what you're talking about. Are we missing something here?
You know, wise man says, "When you find yourself in a hole, stop digging." But you still have a death grip on your shovel and you're still at it. Too funny. Go for it, keep digging if you want. LOL
-
This thread has been an eye opener for me. Simply posting my opinion, which goes against the mainstream, left me open to personal attacks by infantile posters.
Rich, for the record, it was never about your personal opinion. Regarding FOW, I agree with you that ideally the AI should play by the FOW rules and not be all-seeing. We would all love to have that in every game we play. Wouldn't that be nice?
The problem was with your outrageous and incorrect assertions about things. Any rational response to you by me and others pointing out facts contrary to your assertions was met with hostile responses and personal insults by you, to everyone including the game designer. There is no excuse for your behavior here, none. End of story. Time to move on...
-
Brit, you have succeeded in making a fun game. Maybe not perfect, as nothing is perfect, but certainly good enough for most and worth a purchase. Thank you for that. It's a shame sales didn't reward your efforts. At this point you have nothing to apologize for. Best of luck in your new job and hopefully you will continue to provide some modest improvements to EOS when you can.
-
Call J.G. Wentworth?
-
If I had even a small amount of programming experience... I think you and others way overestimate what Brit needs to do to fix the AI and how long it would take. What he needs to do is...
ROTFLMAO. Someboby with zero computer programming experience dishing out advice (directives?), and refusing to listen to anything to the contrary. Too funny. This provides some comic relief to an otherwise obnoxious "discussion."
-
in any other forum would have been locked long ago
Shauny1987, Moon knows. The clock is ticking. It's not worth trying to discuss anything further on this thread.
-
It's way out-dated, by my old Strategy Guide for SC is still available at http://www.furysoftware.com/documents/SC-StrategyGuide-Rev1.pdf
Much has changed, but some of the fundamentals have not...
-
This is so sad it's almost funny. Almost.
The really sad thing is that any aspiring game developer who might be considering a jump into this niche market is observing an ugly side of the business. If your first effort isn't brilliant and perfect, be prepared to be abused. Cuz some kid who spent his lunch money on a toy wants it to be perfect.
Ah well. No horse is too dead to beat. Rich, go ahead and lob a few more insults and get yourself in the last word. Enjoy yourself. I'll just sit back and watch....
-
Rich, I was your age for a whole year once. Maybe someday you'll grow up and act a little more mature than you have been here. If you think insulting folks and throwing temper tantrums is going to get you anywhere in life, then I recommend you check out a Dale Carnegie book or two.
The AI already knows how to explore, expand and attack. All it needs is how to search. Brit wrote about that a few months ago. It's not a big deal. He was going to do it. If he was going to do it then, he could do it now.It is obviously a bigger deal than you are capable of understanding, and not going to happen for free. Sure Brit "could" do it, he could probably work himself into total bankruptcy and early death too, but it ain't gonna happen. Wake up and smell the coffee.
Anything else, kid? This is entertaining in its own perverse little way. We all got time...
-
As stated several times in this thread, Brit said he was going to fix the all-seeing AI. Watch my lips on this one. Brit said he was going to fix the all-seeing AI. So it was broken. Brit admitted it. Then all of a sudden he's quitting the game so it's too much work to do.
So what? "Fix" as in make a feature enhancement to implement FOW, or fix as in it was there and then he "broke" it so now he has to repair it? Methinks it is you spreading misinformation here, not I. Every game programmer would ideally like to provide the perfect computer opponent, but most maintain more realistic expectations. And so do most players for that matter. Why not find yourself some pbem friends??
Question for pzgndr. Does Strategic Command use an all seeing AI? Apparently you were a beta tester so you should know. Does it?See post #48. And what do you mean by "apparently" and "were"? My posts clearly show me as an SC Beta Tester, but you simply cannot accept anything provided to you in black&white? I know a thing or two about the challenges of AI programming and AI scripting based on everything Hubert Cater's been through since about 2002, as well as my own Advanced Third Reich mod efforts. It's not trivial.
When Brit and others tell you point blank how challenging it is to reprogram an AI originally built to not use FOW, listen to them. And to think Brit will commit to hundreds or thousands of hours of additional effort for free for some piss-ant whining about "moral obligation" to provide some marginally better AI, you must be high on drugs. Too bad for you, eh?
I wasn't going to comment anymorePlease don't. Bye bye.
-
The question about the omniscient AI: I don't think it's feasible for me to get together an AI that's using real fog of war. It's something that's a real challenge to do, and getting it working properly would be quite a bit of work. And, by "working properly" I mean that the AI can compensate for this lack of knowledge with a much increased skill in playing the game.
This comment from Brit leads me to believe the AI in EOS never used FOW, so it's not like Brit "broke" it and now needs to "fix" it. If he did have it using FOW at one point, then it should be relatively easy to turn it back on again and improve upon it from there. That does not appear to be the case? So, backfitting the AI to implement FOW would be a real challenge as Brit says. And not worth it at this point. The computer opponent provides decent gameplay as it is.
-
You're a real piece of work Rich. You still don't get it. Probably never will either...
-
In the quote Brit provided the strategic AI knew where the enemy was somewhat. But the tactical AI still needed to search just like a human. I don't find that too unrealistic for a ww2 game like WitE. The game is divisions and corps. In RL intel would generally have a basic idea where the enemy's large units like that were within a realistic distance.
Rich, this was Joel Billings' quote about WITE? The AI does know tactically and strategically where all enemy units are. How could it know strategically but not tactically? If you reread his comments, the AI still has to fly air recon to increase the detection levels of enemy units, which it already knows where they are, not to "find" them but to affect ground combat results. There's no "somewhat" about the AI knowing where all enemy units are in WITE. In case you're interested in knowing what you're talking about...
-
Unless the developer wants to invest a serious amount of time into this, which I'm pretty sure he doesn't, no amount of complaints into the issue will help.
Well said. I'm sure Brit was well-intentioned when he commented about resolving the AI FOW issue. But that was then and this is now, sales have been sluggish, Brit is moving on, and that's real life. What's disturbing here are a couple of folks unwilling to accept this, insisting that Brit "owes" them hundreds or thousands of dollars at personal loss in additional development work for a $45 (now $20!) game. C'mon.
In fairness, I also appreciate an AI that uses FOW. The Strategic Command series does this and it works well, but as stated this is really something that should be coded from the start and not backfitted later. Perhaps a fix could be implemented to restrict specific unit spottings, such as for subs and paratroops, etc, in response to some of the harsher criticisms. But this would have to be something Brit works on as a hobby development effort from now on. Far better to help encourage him to continue doing what he can when he can, rather than kicking him when he's down and making a fuss.
-
Rich, Joel Billings is one of the 2by3 Games developers, so if he says something then it's worth paying attention. Being one of the more successful wargame development teams in this niche market, it is instructive to see what they're doing that actually works versus idealistic expectations for something more. If AI performance is optimized with FOW disabled, then fine. That's not exactly a huge design flaw or game breaking bug, and not worth ranting about. What matters is whether the computer opponent(s) provide competent and challenging gameplay. IMHO, EOS provides decent gameplay. It's a fun little game. Again, thanks to Brit for doing what he's done so far with his limited resources to bring back an updated Empires game.
-
Sigh. Perhaps you guys missed Joel Billings comments on Dec 22? Here's the quote, to back up exactly what I said, for a game I do have and have started to play some:
ORIGINAL: willgamerJust so we're all on the same page, can we talk about what are all of the advantages enjoyed by the AI.
I'm pretty sure it's documented that the AI always plays with FOW off. Does this imply it has maximum recon of everything as well? Does it fly recon anyway?
The AI knows strategically where all enemy units are. However, it plays by the tactical rules of the detection level of units. So it must fly recon to increase detection levels of enemy units just as a human player should. Is knowing where the enemy a big handicap? Yes. Is it needed. Yes. We think the AI plays in it's own fog of war and doesn't need additional fog of war. It does not use the information nearly as well as a human would. In fact, other than trying to get a handle on where major German armor concentrations are located, it really doesn't use the information at all (and most of the time it's fairly obvious where the German armor is). If you want to play against a capable AI this cheat is imperative. It may prevent some ability for strategic surprise, but the computer is not thinking very strategically anyway, and it is very hard to pull off strategic surprise. Tactically, it really isn't taking advantage of the extra information. This isn't the first Grigsby game where the AI ignores FOW.
Does the AI follow the same air rules? Does it use the Air Doctrines at all? Does it follow the "must be first mission, or no fly" rules?
Yes, AFAIK, the AI follows all air rules that the human player follows.
Does the AI pay exactly the same admin costs... is it constrained by admin at all?
Yes and no. It pays for some things, but others are "free". It abuses this in two ways. If set to 110% morale it can build up to a given number of Soviet units for free. It will build a fortified line along the Svir river line even if it doesn't have the admin points. On the other hand, I don't think the AI is trying to replace it's leaders. I think it just takes what it has. A human player can choose to spend admin to change leaders and this can be very important.
Does the AI set locked/unlocked for SUs? Does it set the number of SUs/HQ and let the default up/down migration of SUs occur?
I'm not 100% sure. I don't remember if the AI just leaves all HQ's with a support level of 3 or if it makes some adjustments. I think it just leaves everything at 3 and lives with the results of the automatic system.
Does the AI observe the same supply rules for both distance and MPs?
Yes, unless you change the logistics level, in which case it gets some help.
Does the AI use refit?
I don't remember. I think it does, but can't tell you how or if for sure it does. I need to check with Gary on that.
I'm sure I've missed some.....
I'm not going to rehash "Can the AI escape pockets in some enhanced way?" since that's heavily discussed else where.
The AI does not always move it's units like a human player. It does do some warping if it judges that it is in a supplied situation near a rail line and not moving through zocs (not sure exactly what the limits are, but we worked hard to try to avoid warping out of possible pockets). This warping is needed because the AI cannot do the unit shuffle when forming a line like a human player would naturally do. On the downside, you may have noticed that the AI has a very hard time keeping it's unit integrity intact, so it is constantly suffering combat value reductions in combat. We tried to make an AI that could play well but not cheat in ways that would make valid strategies ineffective. Based on some of the posts about the AI, we've done a decent job. But I also see some players already crushing the AI after only a week or two of playing, and when that happens increasing the play level is called for. Eventually, all human players that invest time in playing the game will get good enough to beat the AI at the normal level, because humans learn and the computer does not.
Maybe Joel doesn't know what he's talking about? You fellas can take it up with him. Enjoy!
-
Sorry, I do not plan to convert this again to WWI.
-
The game should never have been published this way. I've never, in 30 years of computer gaming, seen a game with an AI that sees all the player's units and moves. It's a definite game breaker.
Oh please. Gary Grigsby's War in the East, just released, has an AI that is all-seeing and is getting a lot of positve comments. This was explained in some detail as to why it helps make the AI more competent and challenging, and doesn't help the AI all that much with grand strategy advantages anyways. Many games afford the AI a similar "game breaker" and still the AIs are criticized for not being competent enough. EOS does OK.
-
I am revisiting A3R with a modest update for the latest GC v1.04 patch:
=================================================================Advanced Third Reich for SC-WWII Global Conflict v1.04
=================================================================
DATA CHANGES
- revised localization.txt to change "Operate" to "Strategic Redeployment"
- corrected a research chit error for USA in the 1941 campaign
- corrected unit build discrepancies for Russian 3 Infantry Corps and USA ASW fleets
- corrected resource name discrepancies for Cologne and Dusseldorf
- revised bitmaps for Free French units (now blue French units under USA anti-tank slot)
- fixed Vichy France alignment error in 1942 campaign
- increased reinforcement costs from 7% to 8% for all units (v1.00 only changed HQ costs, but change was intended for all units)
SCRIPT CHANGES
- revised strength events to restrict Axis supply shortages in North Africa to summer season only
- added several Axis AI unit events for German defense of objectives during withdrawals 1943-44 (to compensate for AI weaknesses)
- revised Allied variant unit events for Free French (now one 2-3 in Syria and 2-3 in Algeria)
- revised mobilization_1 events to increase chance of Vichy France mobilization upon Allied invasions of North Africa
- added mobilization_3 events for activation of Vichy colonies if Vichy France is activated as an Axis minor
- revised garrison events to improve German position in pre-surrender Italy
I had a pretty good playtest game running the 1942 campaign as Allies vs Axis AI on Expert +1.5. Lots of tough fights leading up to an Allied tactical victory in March 1945. I should have all four campaigns updated in the repository by next week.
-
Unfortunately, I see EOS winding down. I just can't afford it anymore... It's not a living wage.
This is unfortunate news. I did buy a copy of the game to help support development of niche wargames like this. The real-world problem of course is how small indy developers can produce wargames like this and survive? It's not like World of Warcraft with millions of copies sold and a huge revenue stream to support a staff and all sorts of fancy development.
I give Brit credit for doing as much as he's already done. Thank you! And hopefully he will continue to make modest improvements as he can.
-
What Third Reich mod are you referring to?? Not my Advanced Third Reich mod; the only tanks eligible for research upgrades are the Russian 3-5 armor to 4-5.
This has been a controversial topic since SC was released. For fun, it's OK. For more historical "accuracy", one technique is to restrict the number of research chits allowed per tech area to slow down research. Another technique is to restrict the max research levels for some techs. I've done both with Advanced Third Reich, which I prefer.
-
FOW should provide a chance for bad/inaccurate intel. Partially spotted units could have say a 5% chance of displaying a similar but different unit type:eg, infantry instead of armor, fighter instead of tac bomber, etc. Of course, once you move to attack you'd see the correct unit information, but having some chance of being surprised would be nice.
I still play CM and appreciate this feature. Often my guys will spot a Tiger tank and that gets me thinking (or panicking) about how to deal with it, and then it turns out to be a Pz IV or Panther and I rethink my plans. Minor thing, but adds something to a game.
-
I don't think that such a chaining rule has ever existed.
Correct. The manual fully explains the supply rules. HQ supply is based on their range from valid suppply sources, not other HQs. No chaining.
adv reich scenario help
in Strategic Command - GLOBAL CONFLICT
Posted
I provide a copy of my design notes with the mods. Most everything is explained in detail. If you are a veteran player of the old AH boardgames, it should all make sense. If not, well it's explained at any rate. Good luck. Thanks!