LuckyStrike
-
Posts
163 -
Joined
-
Last visited
Posts posted by LuckyStrike
-
-
Actually, there is much less going on there than you think. Don't confuse a lot of eyecandy with the stuff that CM has to do under the hood.
Yeah I realise that
Was just wondering if their game-recording approach had any application to CMSF, but I guess not. Given the pretty amazing stuff that CMSF does do, I'm sure if it was in any way worth the time and resources it would have been done by now, so evidently it is, as we say in IT, 'non-trivial' -
Interesting stuff.
-
?!?!
Or maybe you are intending sarcasm? That tank takes at least one serious penetrating hit in that video that would likely cause crew casualties. It's hard to tell exactly how many hits are shown, since they are using multiple camera angles... I think it may be only three. The telltale for the hit that's definitely a kill is the significant smoke coming out of the commander's hatch after the hit. The HEAT jet definitely got inside the fighting compartment on that one.
It's harder to to tell just how damaging the other hits might have been. One looks like a very low side hull hit, even possibly a miss that impacted the ground right next to the tank. I would think this one would be at best an immobilization. What I think is footage of a third hit is very hard to assess -- too much smoke for me to tell whether it was likely a penetrating hit or not.
Granted, the footage here is of a now obsolete IAT weapon being used against a now obsolete tank, so it has very little bearing on modern armor survivability. Nevertheless, it doesn't exactly inspire me to go jump in a tank and get shot at...
Obviously I was being sarcastic. Jeez.
-
Well not just in some game but most games - you have the approach I mentioned, or a more complex approach where everything that happens is explicitly recorded - so apparently CMSF lends itself to the second approach, which is indeed far far more demanding. Obviously I realise if it was easy to implent it would have been done already, and that approach is never easy, it requires huge effort and resources. I have an MS in Comp Sci in AI, so I do understand this issue a little bit
Its a feature we'd all like to see, but if its in the far future, its worth waiting for
-
Brilliant. Should be used as a recruiting film by the Russian Army for their tank crews!
-
Really? Games like IL2 manage it OK, and there is a lot going on there. Just record the player inputs, and the initial randseed value, and all the random numbers generated during the playback should be identical to those in gameplay. There is no need to record everything or indeed really anything that happens, unless the game architecture is radically different from most. Just start the replay with the randseed plugged in, apply the recorded player inputs as and when they occur, and let the game play itself.
-
Fascinating thread, thanks to all!
-
If its 2015 you will need combat robots as well.
-
Beta soon already? Great news!
-
Run, don't walk, to the order page.
-
Things are good in CM land
-
Boneless as a boneless chicken
-
Let's face it. this was obviously done so modders had another paint job to work on. And with that thought, lets start a letter writing campaign to the DOD for them to adopt a German WWII ambush pattern for all AFVs.
Now we are talking!
-
I have a Dell XPSM1710 gaming laptop with GeForce Go GTx 7950 and it runs SF beautifully.
-
Been following CM for 9 years, Brits Module is a very worthy addition to the series - excellent as always.
-
16 hours just to do the first turn so far, plus I still haven't done anything in China yet. I think it would be quicker to get a degree from the naval war college than complete a campaign in WITP-AE.
Doh it just struck me - WITP: War in the Pacific. I bought that just after it came out. That game is a monster. I got about 2 months into the war, and it was fascinating, but it becomes a bit of a struggle with the slow turns etc, and managing the logistics becomes a mission. Have a faster PC now and some of the bugs were likely fixed, might have another go.
I wonder if anyone actually completed the entire war?
Don't worry, the first turn is by far the longest!
Have you player its predecessor Uncommon Valor? Now that is a superb game, and AFAIK its had many of the WITP features added on.
-
Thanks, the models are superb.
-
Picked up what?
WITP-AE
"Walk In The Park"-Aliens Edition
-
'Recon by being shot to pieces'
-
Thanks for this info
very interesting. -
I was moderately interested in the module, but the more I have seen the more I am now determined to buy it.
-
Redwolf,
It depends. If you're interested in the game's topic, I think that is all the appeal that is needed. If you're not, then as a player you're going to be far more "picky" about what the game offers. Which is natural and not at all unexpected.
As stated earlier, this is a game made by Russians for the Russian market. If it has appeal to customers outside of the Russian market, great. If it doesn't, we're fine with that too since it wasn't intended to have broad appeal. Our relationship with the developer is based on these principles too, so they're OK with it as well. [edit... dang, typo. Corrected an "isn't" that should have been an "is"]
Expect more niche within niche games from the CMx2 engine made by people you've probably never heard of. It is a key part of our strategy of providing wargamers with a wide array of gaming options to choose from. Gone are the days when it's just one Combat Mission every couple of years and it's always the same general game. The next few years should be quite interesting for tactical wargaming!
Steve
An unexpected bonus to be sure. Bravo!
-
Agreed, given the complexity of the environment the AI is working with, its pretty amazing.
-
I've worked eight years building industrial AIs (some of them emergent) for a variety of companies, and made frequent use of neural networks.
Building an AI around them is relatively easy. Getting it to behave, to not overtrain, etc., is very hard, especially when you don't have the option of pouring concrete on the "brain" and setting it in stone prior to shipping (per your suggestion about AI that improves itself against players). Even when you do great care must be taken. That goes beyond neural networks, too, and into most AI component systems.
Which gets back to Steve's point regarding research / testing / tweaking time. That's where the pain really is. It's best to keep AI simple, as the returns really do diminish. Believe me, it'd be great if that weren't the case (I'd be able to justify years of extra work on the projects I've done
), but if you want to build something reliable that's the best way to do it.Yes I agree, overtrain the network and you start to get mush on the output end, and knowing when you have reached, or are about to reach that point, is a problem.
Like your analogy about pouring concrete on the brain haha
Was just wondering if their game-recording approach had any application to CMSF, but I guess not. Given the pretty amazing stuff that CMSF does do, I'm sure if it was in any way worth the time and resources it would have been done by now, so evidently it is, as we say in IT, 'non-trivial' 
Things about R.U.S.E that i'll love to see in CMx2 WWII eventually
in Combat Mission Shock Force 1
Posted
IMHO if implemented, the feature would be a lot more popular than many think. It would be an excellent tactical training tool.