Has 4.0 made the stock campaigns unplayable?

[db_zero]

Perhaps the fleeing is somewhat realistic? I reading a book and around Villers Bocage a British platoon fled and was forced back into position at "gunpoint" only to flee again. It was pretty much accepted that many Allied generals would lose about a divisions worth of men before they finally began to catch on to the art of waging war and many never really did. 

Under those circumstances the largely conscripted armies of the day probably did melt away more than we think. They had been in combat for extended periods of time and there is simply so much one can take.

Even in elite German SS units I'm reading form the same book "The Guns at Last Light" German commanders resorted to things like tossing live grenades at reluctant troops to get them to move forward and fight

[Holman]

9 hours ago, db_zero said:

Perhaps the fleeing is somewhat realistic? I reading a book and around Villers Bocage a British platoon fled and was forced back into position at "gunpoint" only to flee again. It was pretty much accepted that many Allied generals would lose about a divisions worth of men before they finally began to catch on to the art of waging war and many never really did. 

Under those circumstances the largely conscripted armies of the day probably did melt away more than we think. They had been in combat for extended periods of time and there is simply so much one can take.

Even in elite German SS units I'm reading form the same book "The Guns at Last Light" German commanders resorted to things like tossing live grenades at reluctant troops to get them to move forward and fight

There has been a lot of discussion about this.

There has been a change (I'd say a dramatic change) in game behavior, and it seems to involve units fleeing where before they would merely cower in place for a short time.  Units in good cover (a heavy building, a trench) coming under fire will now get up and run into the open rather than hunkering down in place as they used to.  This is as true of Elite British paras as it is of Italian conscripts.

[Hilts]

2 hours ago, Holman said:

There has been a change (I'd say a dramatic change) in game behavior, and it seems to involve units fleeing where before they would merely cower in place for a short time.  Units in good cover (a heavy building, a trench) coming under fire will now get up and run into the open rather than hunkering down in place as they used to.  This is as true of Elite British paras as it is of Italian conscripts.

Agreed...... I have seen a lot of this lately. It doesn't take much to make good quality troops vacate buildings or foxholes and I'm not talking just about HE . Small arms fire will suffice. It's sometimes kind of cool if the troops retreat out the back into cover when under fire but half the time they run out through the front into a hail of bullets or tank shells, well that just looks f*****g ridiculous!

[db_zero]

I have seen some odd behavior in my 4.0 Campaign game like German troops running down the hill slope, being shot at. Retreating and then running down the slope again only to be shot at, retreat back up, then come back down.

This is continuous behavior and seems to only stop then they get shot

[Bulletpoint]

I wish Battlefront would just come out and comment on this issue, saying whether it's a bug that's being looked into, or if things are working as BF thinks they should.

And if so, I'd love to understand their thinking. Why their AI behaviour seems the opposite of most wartime accounts I've ever read.

World War 1 would have been over very quickly if soldiers got out of their holes to flee across the fields when the first shells started landing.

Edited August 8, 2017 by Bulletpoint

[DougPhresh]

I think that if they wanted to better simulate morale having soldiers be pinned is a much better way to do it than a rout.

Even in antiquity everybody knew that the army that broke formation in battle would be massacred. It's as true in a company attack as it was in a phalanx.

Even with conscripts I think they'd be more likely to lie at the bottom of their trenches and surrender when the position was taken than head for the hills into a barrage or a hail of small arms fire. As others are saying CMFI is particularly  impacted because much of the scenarios and campaigns are attacks against dug-in positions in defensive terrain. Just compare any scenario set at Monte Cassino to accounts of the real battle and you'll see the issue.

[Holman]

9 hours ago, Bulletpoint said:

I wish Battlefront would just come out and comment on this issue, saying whether it's a bug that's being looked into, or if things are working as BF thinks they should.

I couldn't agree more!

[Erwin]

I am now accepting bets re you hearing that from BF.   Currently 1:18 against.  Electronic deposits welcomed.  PM for details.

[IICptMillerII]

I too am eagerly awaiting some feedback from BFC on this issue. Hopefully a fix is not as far off as it may appear. I seem to recall hearing that one of the big behind the scenes updates with the new launcher system, that 4.0 introduced to all the titles (except CMSF) is that they can now release an update to the engine that can be installed once to all owned CM games. If thats the case, when the fix is ready, it should be easier now to apply the patch. 

Hopefully we hear something soon. 

[Txema]

23 hours ago, Bulletpoint said:

I wish Battlefront would just come out and comment on this issue, saying whether it's a bug that's being looked into, or if things are working as BF thinks they should.

And if so, I'd love to understand their thinking.

Please, Battlefront, we need your comments on this topic !

[Battlefront.com]

On 8/8/2017 at 7:43 AM, Bulletpoint said:

I wish Battlefront would just come out and comment on this issue, saying whether it's a bug that's being looked into, or if things are working as BF thinks they should.

It appears there's room for improvement, for sure.

On 8/8/2017 at 7:43 AM, Bulletpoint said:

And if so, I'd love to understand their thinking. Why their AI behaviour seems the opposite of most wartime accounts I've ever read.

When artillery comes in soldiers make a decision about what to do OR they panic.  If they make a decision it is EITHER to hunker down and ride it out where they are, or relocate to some place else.  Engine 4 introduced new behavior that allows soldiers to do the latter to an extent never before seen in CM.  It seems that there's some situations where the new behavior (relocate) is getting triggered when the old behavior (hunker) might be better.

Do not confuse relocating with panicking.  If a soldier is standing out in the open and there's a brick building 10m away, he's probably going to make a break for it (perhaps not that second).  This is a conscious decision which may or may not be the right one.  If a soldier is standing out in the open and panics, he makes an unconscious decision to do anything but stay in place.  That too could be the right or the wrong thing to do.  Sometimes panicking units survive simply because making a run for it was the correct course of action.

With this in mind, we think there's likely some tweaking to do.  However, 20 years of debating this very topic with you guys leads me to conclude that some of the criticism is based on unreasonable/unrealistic expectations for how units behave under fire.  There's plenty of accounts of units getting wiped out because they did the wrong thing, like run when they should have stayed or stayed when they should have run.  Therefore, expect to see things you do not understand or necessarily agree with in terms of TacAI behavior even after some things are tweaked.

On 8/8/2017 at 6:52 PM, Erwin said:

I am now accepting bets re you hearing that from BF.   Currently 1:18 against.  Electronic deposits welcomed.  PM for details.

Man, I hope people took you up on that offer!  Your incessant pessimism and complaining might be taken down a notch if you had to pay up

Steve

[Battlefront.com]

On 8/8/2017 at 9:50 PM, IICptMillerII said:

I too am eagerly awaiting some feedback from BFC on this issue. Hopefully a fix is not as far off as it may appear. I seem to recall hearing that one of the big behind the scenes updates with the new launcher system, that 4.0 introduced to all the titles (except CMSF) is that they can now release an update to the engine that can be installed once to all owned CM games. If thats the case, when the fix is ready, it should be easier now to apply the patch.

Yes and no.  The move to unified code and installers for Engine 4 does not make patching any easier than they were before.  What it makes easier is having those fixes applied to across all Families sooner than they could be done before.

The primary problem with patches is threefold:

1.  We have to make and test five separate builds.  We can not assume that if a tweak is made for one that it works for all, nor can we assume that avoiding unintended negative consequences in one game means all games will avoid new problems.  This is time consuming to do.

2.  We have to update all the full game installers (perhaps not right away), which is a separate process than #1 and also very time consuming.

3.  We have limited time, therefore spending time on patches means not spending time on other things.

This has always been the case, and will always be the case.  It's just that prior to Engine 4 we also had a fourth problem which was different CM games were using different versions of code so either Charles had to "port" fixes from the current code back to old code or would have to spend time bringing older games up to newer standards before a fix could be made.  It took is nearly 4 months to update CMFI's code... just to give you an idea

Steve

[Battlefront.com]

BTW, the reason you don't see me as often as you once did is because if I'm here reading and posting then I'm not doing other stuff.  I have a lot of other stuff to do   Therefore, we rely upon or testers to help us figure out what does/doesn't need our more immediate attention.  It's a good system because our testers have many more eyeballs and collective time than I do, and ironically more gameplay time.  You're in good hands with 'em.

Steve

[Michael Emrys]

2 hours ago, Battlefront.com said:

It appears there's room for improvement, for sure.

Sorry, Steve, but this reminds me of another classic understatement: "The war has developed not necessarily in Japan's favor."

Well...the bottom line is I am confident you will all give it your best shot...for now.

Michael

[Badger73]

2 hours ago, Battlefront.com said:

BTW, the reason you don't see me as often as you once did is because if I'm here reading and posting then I'm not doing other stuff.  I have a lot of other stuff to do

<snipped>

Steve

@Battlefront.com

Steve, can you provide any updates on that "other stuff"?  Maybe the a 90-day window for releasing the w.i.p. final CMFI module? 

If not, good luck and best wishes getting all that other stuff done successfully.  Thanks for the fine games.

[Glubokii Boy]

5 hours ago, Battlefront.com said:

 

When artillery comes in soldiers make a decision about what to do OR they panic.  If they make a decision it is EITHER to hunker down and ride it out where they are, or relocate to some place else.  Engine 4 introduced new behavior that allows soldiers to do the latter to an extent never before seen in CM.  It seems that there's some situations where the new behavior (relocate) is getting triggered when the old behavior (hunker) might be better.

 

Steve

Thanks for commenting on this issue 

IIRC many of the negative comments regarding this does not neccesarely see the 'falling back' as the biggest problem...It's more the direction/location they are falling back to that is the main failure.

If this proves to be difficult to 'get right' perhaps the designers could help the AI somewhat by being given a new command in the editor...

- Fallback location - or perhaps - fallback direction -

Fallback location would allow the designer to designate a specific area that the AI-group would withdraw to if forced out of their possitions.

Fallback direction would work simular to the new FACE command...That is...specify a direction for the withdrawl but not a precise location.

A small suggestion for the tweak 

 

[Bulletpoint]

7 hours ago, Battlefront.com said:

With this in mind, we think there's likely some tweaking to do.  However, 20 years of debating this very topic with you guys leads me to conclude that some of the criticism is based on unreasonable/unrealistic expectations for how units behave under fire.  There's plenty of accounts of units getting wiped out because they did the wrong thing, like run when they should have stayed or stayed when they should have run.  Therefore, expect to see things you do not understand or necessarily agree with in terms of TacAI behavior even after some things are tweaked.

Hi Steve, thanks for explaining things. Looking forward to see what you come up with.

[Erwin]

I am now accepting bets re you hearing that from BF.   Currently 1:18 against.  Electronic deposits welcomed.  PM for details.

Man, I hope people took you up on that offer!  Your incessant pessimism and complaining might be taken down a notch if you had to pay up

Nope no one did - as expected...   But you misunderstand...  Am certain that I am not the only one who feels that there is no need to respond.  Most of us know that you guys do a great job with limited resources and we know that new products/upgrades/patches will get released when they are ready for prime time.  It is amazing that the CEO of a company has the time to respond to detailed complaints/comments.  It's very impressive that you do so. 

Edited August 12, 2017 by Erwin

[IICptMillerII]

13 hours ago, Battlefront.com said:

Yes and no.  The move to unified code and installers for Engine 4 does not make patching any easier than they were before.  What it makes easier is having those fixes applied to across all Families sooner than they could be done before.

The primary problem with patches is threefold:

1.  We have to make and test five separate builds.  We can not assume that if a tweak is made for one that it works for all, nor can we assume that avoiding unintended negative consequences in one game means all games will avoid new problems.  This is time consuming to do.

2.  We have to update all the full game installers (perhaps not right away), which is a separate process than #1 and also very time consuming.

3.  We have limited time, therefore spending time on patches means not spending time on other things.

This has always been the case, and will always be the case.  It's just that prior to Engine 4 we also had a fourth problem which was different CM games were using different versions of code so either Charles had to "port" fixes from the current code back to old code or would have to spend time bringing older games up to newer standards before a fix could be made.  It took is nearly 4 months to update CMFI's code... just to give you an idea

Steve

Ahh thanks for the clarification!

13 hours ago, Battlefront.com said:

With this in mind, we think there's likely some tweaking to do.  However, 20 years of debating this very topic with you guys leads me to conclude that some of the criticism is based on unreasonable/unrealistic expectations for how units behave under fire.  There's plenty of accounts of units getting wiped out because they did the wrong thing, like run when they should have stayed or stayed when they should have run.  Therefore, expect to see things you do not understand or necessarily agree with in terms of TacAI behavior even after some things are tweaked.

I'm glad to hear that you guys are considering some tweaks. For me the main issue is infantry fleeing from cover into the open while under indirect fires, not the other way around. I've made a video and thread about it which I'll link for your convenience:

I know that the video is not a conclusive test, but hopefully it is able to show the specific behavior that I think needs to be tweaked, so that you guys may have a better idea on what to focus on. 

13 hours ago, Battlefront.com said:

BTW, the reason you don't see me as often as you once did is because if I'm here reading and posting then I'm not doing other stuff.  I have a lot of other stuff to do   Therefore, we rely upon or testers to help us figure out what does/doesn't need our more immediate attention.  It's a good system because our testers have many more eyeballs and collective time than I do, and ironically more gameplay time.  You're in good hands with 'em.

In that case I hope we don't hear from you until at least CMSF2 is out! Kidding aside, its glad to know that you are all hard at work. Very much looking forward to the upcoming modules/patches/games!

[akd]

There are two problems here:

1. Core CM problem unrelated to Engine 4: AI has no concept of proactive use of cover and only uses cover reactively.  Foxholes and trenches will not provide good cover for AI until AI has already taken significant casualties from indirect fire / HE.

2. Engine 4 problem that exacerbates the above: AI in cover will proactively displace from cover after receiving very light casualties.

You can see the feedback loop problem that results.  Solution: instead of AI recognizing indirect / HE threat and proactively displacing, AI should recognize foxholes / trenches as good cover versus indirect / HE and proactively use this cover when subjected to indirect / HE (cower / hide to minimize exposure).  Artillery vs. AI in cover then becomes more about suppression than destruction (as is the case in reality).

Edited August 13, 2017 by akd

[Bulletpoint]

On 8/12/2017 at 6:56 PM, Erwin said:

But you misunderstand...  Am certain that I am not the only one who feels that there is no need to respond.  Most of us know that you guys do a great job with limited resources and we know that new products/upgrades/patches will get released when they are ready for prime time.  It is amazing that the CEO of a company has the time to respond to detailed complaints/comments.  It's very impressive that you do so. 

 

 

Edited August 22, 2017 by Bulletpoint

[IICptMillerII]

I should have posted this a while ago. This thread slipped my mind. Apologies for that. 

Here is the link to my thread on a video showing the current TacAI behavior in v4 compared to v3:
 

Its in the CMRT forums because I used CMRT to make the video. Its been up for two weeks now, so you've all likely seen it, but I figured I would post the link regardless for those who maybe do not frequent the CMRT forums. Everything else is explained in the thread/video. 

Edited August 23, 2017 by IICptMillerII

[rocketman]

Hadn't seen that one - very well made and a good illustration.

[Badger73]

20 hours ago, rocketman said:

Hadn't seen that one - very well made and a good illustration.

Me neither.  Good stuff to know.  Thanks for cross-posting here.

[Hardradi]

Great vid Miller.