CMx2 and Steve's mention of target "memory"

[c3k]

Guys,

Did anyone else pick up the mention in one of these threads about CMx2 units having "memory" of targets which have gone out of LOS?

(See, this is where one of you multi-thousand posters should post a LINK to the thread. With my meager post-count of ~400, I have no idea how to do that.)

Now, let's discuss how GOOD this inclusion is.

Steve - thanks.

Regards,

Ken

[MikeyD]

I suppose 'target memory' would have the AI actually put those 'last seen' icons currently in CM1 to some practical use. Strict human vs human players would probably feel less of an impact since they actively overide the AI much of the time anyway.

I'm trying to grasp the infantry implications (or would 'target memory' be vehicle-only?). Relative spotting combined with target memory should make for a VERY different feel in game-play.

[ September 13, 2005, 11:37 AM: Message edited by: MikeyD ]

[c3k]

A self-aggrandizing [bump]

Edited because certain symbols don't appear...

[Battlefront.com]

The memory thing solves a bunch of the targeting and behavioral problems previously (and incorrectly) blamed on the TacAI's competence. The truth is that the TacAI pretty much had all the skills it needed, but not necessarily all the information it needed to produce realistic results.

Target Memory (or put into child development psycho babble... object permanence) didn't exist so the TacAI couldn't possibly remember what to do and what not to do based on changing events. The lack of separation of Suppression and Morale meant being told "the unit is getting hammered" and having to react instead of being told "the unit is getting a lot of suppression" vs. " the unit is freaking out".

If we went back into CMx1 and changed these two things, and nothing else, the TacAI would probably go automatically from decent to spectacular. So expect good things from CMx2's TacAI, though since we are starting fresh with the TacAI and the game as a whole I can't make any predictions about its overall level of competence. In theory all bets are off

Steve

[aka_tom_w]

So did Steve Just Say the TacAI is going to go automatically from decent to spectacular in CMx2 !!!!

Given this statement:

"If we went back into CMx1 and changed these two things, (Target Memory AND seperate the Morale and Suppression modifiers or Databases) and nothing else (!) , the TacAI would probably go automatically from decent to spectacular."

So bring on the Spectacular Tac AI for CMx2!!

That IS what he just said right?

-tom w

[ September 13, 2005, 11:44 AM: Message edited by: aka_tom_w ]

[Zalgiris 1410]

So why not spend just a little time BFC and develope an improved CMx1 version 1.4 with such minor enough changes in the interium?

[Martin Krejcirik]

Originally posted by Zalgiris 1410:

So why not spend just a little time BFC and develope an improved CMx1 version 1.4 with such minor enough changes in the interium?

Stave said a zillion times that was not possible with the old engine, that is why they are making a new one.

[Battlefront.com]

So why not spend just a little time BFC and develope an improved CMx1 version 1.4 with such minor enough changes in the interium?

Uhmmmm... because these are not minor changes in terms of what it takes to implement them. If they simple, don't you think we would have patched it in years ago? Or do you think we purposefully withheld such improvements from you guys because we actually LIKE to have you bitch and complain about stuff? Or might it be that these changes are conceptually simple, but to get them in would require a code rewrite, which is what CMx2 is? You be the judge

Plus... I don't know how some of you guys do your time calculatoins. There is no such thing as "interium" when you're the only guy coding. Charles has been coding practically games non-stop for the last 14 years of so. Therefore, 4 months to recode something that is 6 years out of date means 4 years more wait to get the game that is state of the art.

CMx1 is dead, burried, and forgotten about as far as development is concerned. That's been true since the last CMAK patch came out.

Steve

[c3k]

Steve,

The separation of morale state and suppression is great. The target memory (objective permanence? sheesh...) is, IMHO, even better.

I think it (target memory) may be the single best advance between CMx1 and CMx2.

But, do Space Lobsters even _have_ memories?

Thanks,

Ken

[Battlefront.com]

But, do Space Lobsters even _have_ memories?

I dunno, but a lobster fisherman around here pulled up a friend's trap and dressed a lobster in Barbie clothes, put it back in the trap, and then tossed the trap back in the water. The friend was quite puzzled, but instead of keeping it he put the dressed up lobster back in the water. It apparently was repeatedly caught and tossed back in, losing a little bit of clothing each time. Friggin' hysterical

Steve

[jim crowley]

Originally posted by Battlefront.com:

If we went back into CMx1 and changed these two things, and nothing else, the TacAI would probably go automatically from decent to spectacular. So expect good things from CMx2's TacAI, though since we are starting fresh with the TacAI and the game as a whole I can't make any predictions about its overall level of competence. In theory all bets are off

Steve

This sounds very promising indeed.

However, in CMX1, the AI tends towards choosing cover over platoon integrity. In other words, be it on attack or defence, the components of a platoon will more often than not be set up using any and all available cover, often out of C&C with the HQ.

If there is a greater emphasis on C&C in the new engine, this behaviour would need, at least, to be moderated.

[dalem]

I just peed Steve's lobster trap.

-dale

[Dogface]

***Ow OW OW OW OW****************

talk about passing a stone.

[Philippe]

Cross-dressing Space Lobsters of Doom.

This is getting a bit too weird. It hadn't occured to me that cross-dressing was even possible for a lobster.

[Philippe]

On a more serious note, if I understand it correctly (and I probably don't), target memory shouldn't have much of an impact on infantry under fire doing weird things like crawling away from the cover of a wall and exposing themselves in the open. I really hate seing minefield markers pop up outside of my line of site for two reasons: 1) I shouldn't be able to see them; and 2) it means that my computer opponent is doing something really stupid. Doesn't mean it wouldn't have happened in real life from time to time, but panicked troops shouldn't continue to take refuge in a minefield once someone realizes that there are things that go boom if you step on them out there.

I really hope that AI improvements will cover this kind of thing.

[Jack Carr]

This is good news! If I'm reading this right. It seems to me that this is exactly what was needed, retention of spotted enemy units.

[Battlefront.com]

On a more serious note, if I understand it correctly (and I probably don't), target memory shouldn't have much of an impact on infantry under fire doing weird things like crawling away from the cover of a wall and exposing themselves in the open.

The Morale/Suppression problem is likely responsible for most of the things like this, but it could also be memory. If the unit remebers that there is a threat to that open ground, then it could avoid going there. But if it forgets about such things, then it assumes it is fine until it is out there and getting shot up. Then its too late.

Steve

[stoat]

Originally posted by dalem:

I just peed Steve's lobster trap.

My pants are dry. *reads recent posts, gets excited*

Never mind.

[aka_tom_w]

still and interesting thread if we could stay on topic....

-tom w

[Michael Dorosh]

tom just peed on our fun.

[Barrold]

Pee grog

BFS5

[Brent Pollock]

Probably kept crawling back into traps as a deathwish because all the other crustaceans made fun of it.

Originally posted by Battlefront.com:

</font><blockquote>quote:</font><hr />But, do Space Lobsters even _have_ memories?

I dunno, but a lobster fisherman around here pulled up a friend's trap and dressed a lobster in Barbie clothes, put it back in the trap, and then tossed the trap back in the water. The friend was quite puzzled, but instead of keeping it he put the dressed up lobster back in the water. It apparently was repeatedly caught and tossed back in, losing a little bit of clothing each time. Friggin' hysterical

Steve </font>

[c3k]

And I thought this thread would have some legs. I just didn't think there'd be 8(?) pairs of them and they'd be red and covered in shell.

Ken

[stoat]

Originally posted by c3k:

8(?) pairs

That's 16 legs. I didn't think this was common but these are Space Lobsters of Doom.

[Mord]

Originally posted by Battlefront.com:

Or do you think we purposefully withheld such improvements from you guys because we actually LIKE to have you bitch and complain about stuff?

Steve

YES! All you game making types are masochists. You can see the stiletto imprints on Charle's brain if you look real close....

Originally posted by dalem:

I just peed Steve's lobster trap.

-dale

LMFAO! You my friend are gonna dehydrate...atleast you're aiming in other directions beside Mord's...my rash has finally healed.

Mord.