CMx2 4.0 Tech Question

[Ithikial]

Just curious before I raise anything over at the BF forums.

Has anyone else noticed an increase in mouse input lag post 4.0 upgrade? I've just started playing through the Amiens Tonight campaign for CMBN and noticed I'm having some trouble with mouse clicks only registering after a serious delay/command buttons not deselecting etc. It's not constant and a bit adhoc.

Secondly some of the end points for waypoints appear to be being registered on the edges/corners of action squares rather than in the centre. Some weird infantry pathfinding, like heavy forest tiles where infantry now appear to want to go around it rather than through it.

 

[Gnarly]

Can't say I've noticed either, though I haven't looked in such detail, not have I anything like the experience you have with CM. Tried it without any mods first, to remove any variables?

I have noticed that infantry in foxholes are now complete chickens when HE fire arrives: foxholes are close to useless under arty barrage in v4.0, and its not just me noticing this.

 

[Nathangun]

Haven't noticed anything myself that's because I haven't been playing, well waiting several weeks on PBEM's, but I bet foxholes and ATG's are still easy to spot.

 

[PhilM]

Just curious before I raise anything over at the BF forums.

Has anyone else noticed an increase in mouse input lag post 4.0 upgrade? I've just started playing through the Amiens Tonight campaign for CMBN and noticed I'm having some trouble with mouse clicks only registering after a serious delay/command buttons not deselecting etc. It's not constant and a bit adhoc.

Secondly some of the end points for waypoints appear to be being registered on the edges/corners of action squares rather than in the centre. Some weird infantry pathfinding, like heavy forest tiles where infantry now appear to want to go around it rather than through it.

I've not noticed the mouse input lag, but have been seeing (I think!) issues with waypoint registration: previously they were pretty easy (leaving aside vehicles) to get them to "jump" into the intended AS, now they seem a lot more prone to leap an AS away from where the cursor is and end up in the neighboring AS, as though they don't now auto-centre on the AS in the same way?

Also the pathfinding decision making seems to have changed, as you say.

I wondered if they were, respectively, the result of:
- the introduction of the ability to "peek around corners" has somehow "subdivided" (not quite what I mean, but cannot think of a better word!) the AS to allow men to take up particular positions within them, and so has introduced some sort of sensitivity within an AS?

- the introduction of (I think) more varied / less "line ahead" on-foot movement, meaning pathing decisions are now being taken slightly differently?

May well be rubbish of course!

 

[A Canadian Cat]

I have not seen any issues with mouse lag my self.

The "problem" of where the way points land has been around for a long time. I don't really get what is up with that. Some terrain seems to trigger segmented way point landing spots much like a fence or wall running through an action square means there are two three or four way point landing spots. I see it most frequently in forested ares with elevation changes. Since it seemed to have no effect on where the soldiers actually go (as in they position themselves just as well and just as bad as when the way point hits the centre) I have never investigated further.

As for pathing - that I have not noticed a difference but I have always considered my men to be lazy and they constantly try to go around difficult terrain and wonder out of depressions and ditches just cause its easier. So I am in the habit of using more way points to make sure they stay under cover. So, it is possible that something has changed there that I just have not noticed.

 

[Ithikial]

I've been away from CM for a little bit because of RL and certainly not getting to play as many scenarios as say two years ago. (My own game tracker tool is telling me that. )

Just in my last two scenarios I've played since v4.0 released I've hit a few real interesting WTF moments that has led me to 'save scum' after seeing whole fireteams get wiped out for being barn door stupid.
- The forest tile issue in my opening post. British infantry decided to take the long way around a clump of forest tiles and run down an open road in front of a pre-existing fully spotted German MG team. The initial plot path had them running to get into a better fire position two AS away from their current location. Both AS including their final destination was located in heavy forested tiles.
- Another scenario move a US Squad up to the side of a house with a window and door. Before moving into the house from the same side I've issued a pause order for 15 seconds. My troops unload on contacts inside the house and then begin to undertake their next way point of moving into the house itself... not by not using the door directly in front of them but deciding to split up and go around the house on both sides to try and enter from a door on the opposite side of the building. Half a squad gets cut down by fire from the back of the building. Again this second contact is a known contact previously spotted (though admitedly from another unit) and I'm trying to engage them from inside the house.

 

[Baron]

About input lag, it's something I've been experiencing, although I can't tell how much 4.0 has to do with it. It's mostly an issue in larger battles, I've just finished a Huge QB (humiliating defeat, but I digress) and at times the game became unplayable. It seems like input lag builds up over time somehow, but fortunately, alt-tabbing out and back into the game seems to "reset" the issue. This made me paranoid around the Go button, it happened a few times that I thought I had missed the button because of input lag and clicked it again... making me lose a turn.

About "subdivided" Action Squares, I've experienced them as well. I haven't dug into the matter, but I only noticed it when I needed precise waypoint placement : on a crest in a forest.

Fortunately, I've never experienced any infantry pathfinding problems. I do remember that some doors can be blocked by slightly elevation terrain. I must however share your frustration regarding movement inside ditches (and other similar terrain features), I don't know if the pathfinding algorithm takes elevation into account, but perhaps it should (or to a bigger degree).

 

[A Canadian Cat]

- The forest tile issue in my opening post. British infantry decided to take the long way around a clump of forest tiles and run down an open road in front of a pre-existing fully spotted German MG team. The initial plot path had them running to get into a better fire position two AS away from their current location. Both AS including their final destination was located in heavy forested tiles.

Some time ago I noticed that vehicles were driving out and around wooded areas if they were inside the woods but close to the edge. Placing more way points along the path through the woods prevented that from happening. Soldiers have always looked for the easy way - the constantly run out of depressions and along the dry top instead of staying in the rough or muddy terrain in side the depression. It is certainly possible that the v4 tweaks made this worse. It is also possible that the same thing would have happened in CMBN 3.12 too you just never hit that particular scenario. Yeah helpful I know

- Another scenario move a US Squad up to the side of a house with a window and door. Before moving into the house from the same side I've issued a pause order for 15 seconds. My troops unload on contacts inside the house and then begin to undertake their next way point of moving into the house itself... not by not using the door directly in front of them but deciding to split up and go around the house on both sides to try and enter from a door on the opposite side of the building. Half a squad gets cut down by fire from the back of the building. Again this second contact is a known contact previously spotted (though admitedly from another unit) and I'm trying to engage them from inside the house.

So this is very likely a bug / messed up map. There are a couple of things that can happen here. One is the terrain elevation is partially blocking the door - that door will not be usable but players have no indication of that. Solution the map needs to be tweaked or the door removed - author's pick. Or we have had a few cases where a building was just buggered some how (during development a perfectly good and working building was tweaked and now old coppies don't work any more). Solution the building needs to be removed and re placed. Or there could actually be a bug with that building / door placement / terrain placement combination. Again not super helpful to the player I know.

I must however share your frustration regarding movement inside ditches (and other similar terrain features), I don't know if the pathfinding algorithm takes elevation into account, but perhaps it should (or to a bigger degree).

Yes, and no.

The not staying in the ditch / depression thing is usually a smoother terrain vs rougher terrain choice that the TacAI is making. The way to deal with it is to use more way points closer together in the ditch / depression. If the movement segments are short the TacAI will not bother to come out of the ditch and then go back in.

Elevation comes into play in a wider context. If you have a hill that has steep and less steep parts your troops will favour the less steep parts of the hill - if the difference is big enough.

 

[Swank]

So this is very likely a bug / messed up map. There are a couple of things that can happen here. One is the terrain elevation is partially blocking the door - that door will not be usable but players have no indication of that. Solution the map needs to be tweaked or the door removed - author's pick. Or we have had a few cases where a building was just buggered some how (during development a perfectly good and working building was tweaked and now old coppies don't work any more). Solution the building needs to be removed and re placed. Or there could actually be a bug with that building / door placement / terrain placement combination. Again not super helpful to the player I know.

Not sure this clears anything up or not but just bought FI (with 4.0) and when playing the scenario Beyond the Belize I was moving troops into a building that is a bit exposed and has a door on the friendly side and a door on the enemy side. Too my horror the squad separated into two and half went in one door and the other half went in the other door. See image reference for which building if someone wants to see if the building is bugged. I never noticed this as an issue prior to 4.0. unfortunately I didn't think to take a screen shot or save at the time of the incident.

 

[Nathangun]

Has this issue been raised in the BF forums?

 

[DoubleD]

I've seen this, troops use all available doors. I've also blasted through a building wall with a squad to avoid a certain gate in a compound, the squad blew the wall, then split and half the squad ran around to the front of the building straight through the gap in the wall I wanted to avoid... really!

The fix I'm using at the moment is to split the squads and stagger their entrance into a building, micro managing them to the side I want them to enter from. It kinda works but it's fiddly, so I use it sparingly. It looks linked to the number of troops who need to get into a building. Maybe the type of order, fast, quick, normal, hunt etc. Influences, I haven't tried that to see if there's a difference.

Its certainly happening, I just put it down to exuberant pixeltruppen.

 

[PhilM]

Exuberant ... but sometimes dumb!

I know it's easy to criticise, and *most* of the time the AI does a very good job, but ...

The "blasting a gap" circumstance is something that one would think could be hard-wired into the game?

If the movement is for a unit to a waypoint on the far side of a gap that that unit has just blown (which "new" gap the programme must recognise as such, to know that there is now a gap that wasn't there before ...), then surely in 100% of circumstances the unit should use the gap? What logic is there for applying any other outcome?

It may turn out to get them all killed! But it's still the move that was intended, rather than "find your own best (?) way to the new waypoint", which seems to be what is happening** ...

In normal circumstances, it may well be that the quickest / safest way to get x men inside a building with two doors is NOT just to use one entrance, but both of them ... but having blasted a way in is not a normal circumstance.

(**As also in jumping over to the far - dangerous! - side of a low wall and then back again in front of spotted enemy, seemingly because it's a VERY slightly shorter / quicker route to get from A to B even though A and B are both on the same, "safe" side of the wall. I know, I know, use shorter / tighter waypoints ... but this happens even with both waypoints and the visible path all on the "safe" side of the low wall, and some of them jump it anyway ...)

 

[Ithikial]

Maybe the type of order, fast, quick, normal, hunt etc. Influences, I haven't tried that to see if there's a difference.

Its certainly happening, I just put it down to exuberant pixeltruppen.

Never thought of that maybe you need to use the assault command now following some tweaks to the engine and infantry path finding. For the WW2 games especially I'm guilty of using the 'quick' command probably too much including building entry but it's usually the case of getting as many bolt actions into a building as possible to swarm the enemy. Or maybe I'm just a horrible Soviet styled commander at heart.