Frustration with CMCW - Russian side

[domfluff]

They do not have no memory of previous spots. This is (partly) what spotting contacts are, and why you upgrade from a partial spot to a full one easier.

[womble]

47 minutes ago, domfluff said:

They do not have no memory of previous spots. This is (partly) what spotting contacts are, and why you upgrade from a partial spot to a full one easier.

Sometimes their memory of previous spots persists way past usefulness, into the realms of distracting.

[FinStabilized]

8 hours ago, domfluff said:

They do not have no memory of previous spots. This is (partly) what spotting contacts are, and why you upgrade from a partial spot to a full one easier.

 

[Butschi]

The discussions about spotting that pop up here frequently voften come down to good/bad luck with RNG. As far as I can tell, the random part of spotting works like this: There is a dice roll every second about whether you get a contact icon. The probability is determined by... whatever affects spotting. Once you have a contact icon, there is another dice roll every 7 seconds. (starting at 0 seconds, so you can go immediately from no contact at all to full actually seeing the enemy unit). Probabilities are, again, determined by whatever affects spotting. Incidentall, that is what makes a functioning C2 chain important. If your tank gets the contact icon via from somewhere else, you skip the first dice rolling part.

An M60 has a higher probability than a T-62 but both can end up in the long tail, where spotting takes minutes because of bad luck. It just happens more often for the T-62. It's really like waiting to roll a 6 on your average die. You can get lucky and roll the 6 the first time or don't get one 33 times in a row. The M1 gets to roll with a D6 while your humble T-55 rolls a D20.

Edited April 22, 2023 by Butschi

[chuckdyke]

It is not the M60 which spots quicker it is the result of more effective combined operations. Figure out a way so that the full contact of Joe the Grunt gets passed on to the FO inside a couple of minutes.

[Andrew Kulin]

3 minutes ago, chuckdyke said:

It is not the M60 which spots quicker it is the result of more effective combined operations. Figure out a way so that the full contact of Joe the Grunt gets passed on to the FO inside a couple of minutes.

I agree with this in the sense that more eyes passing on information effectively should certainly help in spotting.

The issue at hand however is that a single tank, on its own, with supposedly modern sensor capabilities, and crewmen with their own eyeballs, including an unbuttoned TC, should, while hidden unmoving in a treeline, be able to spot moving vehicles, out in the open, during daylight hours in a reasonable amount of time.  They are not.  In the one battle, I would have had around 6 vehicles (T72s, BMPs) with eyes on an open field with moving Abrams and Bradleys and TOWs and not spotting a single one over many minutes of battle.  But the American forces are picking off my vehicles with alarming frequency.  I read in earlier posts that this enhanced spotting capability is because American force doctrine is having their TCs unbuttoned and that buttoning them up will bring their spotting ability down.  If that is the case, then surely unbuttoning Soviet vehicles should do same, no?

26 minutes ago, Butschi said:

An M60 has a higher probability than a T-62 but both can end up in the long tail, where spotting takes minutes because of bad luck. It just happens more often for the T-62. It's really like waiting to roll a 6 on your average die. You can get lucky and roll the 6 the first time or don't get one 33 times in a row. The M1 gets to roll with a D6 while your humble T-55 rolls a D20.

If BFC is using something like a random number generator in spotting, with a rule along the lines of if RND() > XXX % then vehicle is spotted (Where XXX = a number between 0.000 and 1.000 (0 - 100%)) then I would suggest that they revisit whatever cutoff value they have set for the Soviets because it is too low.  I presume that the checking done every 7 seconds or whatever it is in game time is done against every unit (tank, IFV, squad, gun, etc.) within LOS of the Soviet tank/IFV.  Which would mean every such unit fails the spotting test say 9 times/minute.  Statistically that is one hell of a lot of spotting fails, especially when you now multiply that by multiple Soviet vehicles also all drawing blanks on the same turn, and over many turns thereafter.  Yet soviet infantry in the area are able to spot enemy vehicles in a reasonable time frame.  So what is the difference in spotting mechanics of enemy vehicles used by Soviet infantry vs. Soviet vehicles?  Is that something that can be fixed?

[chuckdyke]

9 minutes ago, Andrew Kulin said:

Soviet vehicles should do same, no?

Soviet tanks fight with a different doctrine. They spot when the enemy fires at them and let attrition work in their favor. 

[Andrew Kulin]

14 minutes ago, chuckdyke said:

Soviet tanks fight with a different doctrine. They spot when the enemy fires at them and let attrition work in their favor. 

Huh?

They are trained to not look for tanks?  Just sit back, enjoy the birds and butterflies rather than waste their time looking for threats, and only start to look for enemy when one of their friends goes BOOM?  Because getting in the second shot is the better approach?

[domfluff]

11 hours ago, FinStabilized said:

 

No, I'm sorry, you're 100% wrong.

A spotting contact *is* the unit's memory of a previous spot.
A unit with a spotting contact will upgrade this to a full spot faster than one without a spotting contact.

The spotting contacts are then the representation of a unit's "memory" (or, equally, intra-unit communication, or more broadly the ISTAR picture in general.)

Now, that's the system as presented. Taking exception with that system is one thing, and one can certainly have opinions about that from a design perspective - any wargame design is a conflict between fidelity of representation and how that fidelity will end up being used, whether that's "playability" or something to do with the questions being studied - but to claim that this doesn't exist is just nonsense.

The specific claim was "CMCW units cannot remember what they have spotted". This is trivially simple to disprove, and if you're using as a basis for an argument, your argument can carry no weight.

Edited April 22, 2023 by domfluff

[Grey_Fox]

4 hours ago, Andrew Kulin said:

If BFC is using something like a random number generator in spotting

My understanding is that they are modeling the individual eyeballs of the troops and where exactly they are looking at any given time.

For instance if somebody is using binoculars to observe something from a few kilometers away, that is a pretty narrow cone of vision to look through.

Viewports and the position of troops within the vehicles are modelled - in fact there was a bug where a crewman was positioned facing the inside of the tank rather than their assigned vision ports and so couldn't actually see anything.

If there is an obstruction between the viewer and the target, say a cluster of trees or whatever, then there is a percentage chance that that the viewer will be able to see past each individual tree towards the position of the target.

You also have to remember just how limited the fields of view armoured vehicles have. You're talking about narrow slits in the armour, with often fairly big gaps between them. This is why it's often highly recommended to turn your tank crews out as often as possible, and why it's the practice in western armies.

Soviet doctrine was fight buttoned-up, and rely on mass to overcome the reduced fields of view that would be the result.

Edited April 22, 2023 by Grey_Fox

[domfluff]

Also, of course there's an RNG aspect to spotting - that's how spotting has been modelled since there have been models for spotting.

As the basic example, Koopman in Search and Screening (1946), who theorised that the detection rate is proportional to the solid angle subtended at the point of observation of the target.

Since your chances of finding something is going to be harder the larger an area you're looking at, that's a base-e relationship of some kind. The Koopman probability of a detection in time t is P(t) = 1 - e^(-yt), with your y in Koopman theory being y=kh/r^3, h and r describing the height and distance to the target, and k being a value for how complex the search operation is.

Call that a "dice roll" if you like, but that's how modelling this kind of thing usually goes.

 

[Andrew Kulin]

3 minutes ago, domfluff said:

Also, of course there's an RNG aspect to spotting - that's how spotting has been modelled since there have been models for spotting.

As the basic example, Koopman in Search and Screening (1946), who theorised that the detection rate is proportional to the solid angle subtended at the point of observation of the target.

Since your chances of finding something is going to be harder the larger an area you're looking at, that's a base-e relationship of some kind. The Koopman probability of a detection in time t is P(t) = 1 - e^(-yt), with your y in Koopman theory being y=kh/r^3, h and r describing the height and distance to the target, and k being a value for how complex the search operation is.

Call that a "dice roll" if you like, but that's how modelling this kind of thing usually goes.

 

Wow.  I'm an engineer and I did not even understand that.

But all things being equal, who would have higher probability of spotting?  Crew in a moving tank spotting a stationary vehicle located just inside a forest, or a crew in a stationary vehicle spotting a moving vehicle out in the open?  h and r would be essentially the same (tanks about same size respectively).  I would assume the k term for complexity (however that is calculated) would be smaller for a stationary, hidden target compared against a moving unhidden target (and note, our eyes are attracted to motion - so moving things should be modeled with higher potential to be spotted).  Presumably the area being searched is the same across all units so however that is accounted for in the equation should be a wash for both sides.  And not that I expect it to be modeled such in the game, in the case of the Soviet tanks sitting in ambush positions near edge of woods, eyes should be forward, not needing to do 360 degree scans like probably would be the case for the moving units.

 

[domfluff]

We know that the spotting model in CM is done from pairs of eyeballs (so an infantry section will have perhaps 10 different chances to spot something), and that there's a larger chance of spotting directly ahead of the spotting eyes, than to the periphery. Moving targets and large targets will inevitably be spotted faster. The individual optics for each position are modelled, so (for example) a tank with a decent thermal optic on the gunner's sight will have the gunner spotting better than the driver. Depending on the vehicle, commander's often have access to the gunner's sight, and sometimes have an independent sight of their own (to a large extent that's what the M60 cupola is doing).

Plenty of other factors will be at play here, including the individual line of sight of each spotter (it's entirely possible for a tank commander to see something and the gunner's sight to be blocked by complex terrain). and the soft factors involved, not to mention the spotting contacts that have been shared across the C2 network.

So sure, all things being equal, a stationary unit should tend spot a moving unit before the moving unit spots a stationary unit. I cannot imagine what hoops you'd need to jump through to engineer a situation where all things were actually equal, and even if this was the case, it can only be a probability and a tendency, there will always be outliers.

In the case of armour specifically, the chances of spotting is also going to be atrocious (even in something ultra-modern like an Abrams or a Bradley, by comparison to infantry), so your initial chances of anyone spotting anything are going to be terrible. Your baseline in all situations, but especially in CMCW and the WW2 games, are that tanks are blind. Infantry are your spotters, and that's one of the many reasons why "combined arms" is a thing.

[Grey_Fox]

12 minutes ago, Andrew Kulin said:

Wow.  I'm an engineer and I did not even understand that.

But all things being equal, who would have higher probability of spotting?  Crew in a moving tank spotting a stationary vehicle located just inside a forest, or a crew in a stationary vehicle spotting a moving vehicle out in the open?  h and r would be essentially the same (tanks about same size respectively).  I would assume the k term for complexity (however that is calculated) would be smaller for a stationary, hidden target compared against a moving unhidden target (and note, our eyes are attracted to motion - so moving things should be modeled with higher potential to be spotted).  Presumably the area being searched is the same across all units so however that is accounted for in the equation should be a wash for both sides.  And not that I expect it to be modeled such in the game, in the case of the Soviet tanks sitting in ambush positions near edge of woods, eyes should be forward, not needing to do 360 degree scans like probably would be the case for the moving units.

 

One anecdote isn't proof of anything. There's always going to be a chance that something like this could in fact happen at least a few times, and we also don't know if the tank already had a spotting contact, be it from pre-battle intel or a scout. 

[chuckdyke]

8 hours ago, Andrew Kulin said:

Because getting in the second shot is the better approach?

I play mostly WW2 and had a formation of JS2 vs Tigers and Panthers in a Hull Down position. I spotted the Tigers and Panthers before they spotted the JS2s. Reason I used infantry to do the spotting a Regular Ivan without binoculars spotted a Panther in a Hull Down Position from over a Km away. He ran back to the nearest JS2 and when they moved, they got the first shot in. I find it hard to accept that BF lowered the standards in CW. CM comes close to real life but there are differences.

[The_Capt]

19 hours ago, Grey_Fox said:

One anecdote isn't proof of anything. There's always going to be a chance that something like this could in fact happen at least a few times, and we also don't know if the tank already had a spotting contact, be it from pre-battle intel or a scout. 

This.  And in Valley of Ashes I had a moving Soviet T64 spot and engage a static M60a3 in a wood line at 2+ kms.  This does not prove or disprove anything beyond the fact that stuff on the end of either spectrum can (and will) happen.  The problem the developers have is that these wild situations are what players remember.  They do not remember the other 80% of the time when engagements happened pretty much as one would expect.  And then once a bias sets in it becomes all one sees.

So in beta we have run a lot of spotting tests and the US spotting abilities are superior in the main, pretty much as they were in RL.  Always room for improvement, but no, “Spotting is not broken”…this is a conversation we have had many times.

[Butschi]

3 hours ago, The_Capt said:

They do not remember the other 80% of the time when engagements happened pretty much as one would expect.

Or the situations where RNGesus was on their side.

[The_Capt]

1 hour ago, Butschi said:

Or the situations where RNGesus was on their side.

Exactly.  Take any battlemap and put a Soviet tank platoon in a position about 1000-1500m out.  Then blindly drive US forces right through their field of view.  You are going to wind up with a bell curve of how quickly the US force get spotted - some early, some late.  Now unbutton the Soviet tank platoon.  Now move it closer.  Move it farther.  Change the crew experience.  Change the tank.  You are going to wind up with a bunch of bell curves.  So the actual question for the developer is “where is the middle of the curve?”  If the curve is indeed “They never see them, not once 10 time out of 10!” in a context where they should be closer to 50-50.  Well then we do have problem.  But having actually run a lot of tests like these, I know it is not the case.

We constantly tweak and provide feedback on where that bell curve should be.  So when a player comes on the forum frustrated with an experience the next question has to be “well what was the context and where would it fit on all those curves?”  Big problem in this case is we are not given any in game ranges or data - how far, what conditions, which Soviet vehicles and how many, what crew experience?  For example if this was a  single buttoned T-62s with green crew at 2000ms, we’ll just forget it.  That tank was not designed to engage at those distances and is basically looking through a straw at those ranges buttoned.

In the end a lot of the frustration about CM centres on the fact that the game models tactical friction too well.  Humans in combat do incredibly dumb and counterintuitive things - it is the stress.  So when your stupid tank goes the wrong way around trees, that is not a bug it is a feature.  It is brutally realistic.  The trick is learning to succeed even in the face of all that unscripted chaos.  And for some it is a major draw, others well maybe not so much.

[Sublime]

OP goto youtube search free whiskey domfluff combat mission soviets

there should be a few vids that are excellent. also u can dm me for specific tips. 

You have to accept YOU WILL TAKE CASUALTIES

You HAVE TO USE COMBINED ARMS

The tank sensors WILL NOT SEE THE WESTERN TANKS FIRST, UNLESS YOU ARE SITTING STILL LOOKING AT WHERE THEYLL COME FROM

Learn to use your ATGM inf teams to take out western tanks that are exposed.  If you can tactically at all always try to have the western tank advance or move or enter your tanks LOS, your tanks on the move will never spot the western tank first.   If you cant do this, you need to have infantry with the tanks, and mass your tanks.

I almost always let the tac ai target on its own, but in this case if you are doing massed tank attacks consider things like t62s or maybe not the best sov tanks.  Franjly I think T64Bs are always a better choice than a T80B, perhaps except defense where the reverse speed is like a lousy 4 km/h better.   But in case of massed tank attacks with infantry, use weight of fire.  So lets say you got 12 T62s and see 3 m1s.  First remember western platoons are in 4s with tanks and pact is 3s.  so always look for an extra tank, tho it may not be there. second, weight of fire, tell your tanks like if its m1s you want 3-5 tanks shooting at each m1.  This by weight of fire will kill or mess the tank up so bad it can at least stop a disaster. 

DO NOT TRY POPPING SMOKE if its DEFINITELY A THERMAL EQUIPPED WESTERN WEAPON - THEY WILL BE ABLE TO KEEP FIRING AND NOW U CANT EVEN TRY TO SHOOT BACK

Soviets have a fckton of IFVs, apcs, and tanks.  almost all have coax and a few thousand rounds.  literally shoot eveerything potentially  that could have enemy troops in it with target light for mg fire, multiple vehicles, for a min or two per spot, longer on buildings.   This often also gets a repsonse or makes a human think you know where their men are

Soviet arty respone times suck,. Get TRPs and preplan your arty otherwise

everyones different. the way I play them works great or disastrously-  I like high tempo attacks.  IE Im on you and th ebattles gonna be decided fairly quickly...  If you do this right, or learn to do this right lots of people are terrible at handling this, some ppl play really cautiously or slow.  The entire sov doctrine actually encourages what I do somwhat, you need to basically sorta probe or designate a couple attack lanes and send some BTRS or BMPs, and have some infantry near running wit hthem

Where you fin dthe enemy weakest you want to RUSH your men into that hole, and try to blow into their rear where they hopefully arent even sure where u are exactly.   Then You should also on the move be able (time is key, dont stop and mass) pick the avenue you want to strike from now and either roll up th esides, or maybe you wanna rush your tanks over into keyholed spots spots with only a small LOS window as it were ) and have your infantry take up good cover and now the western defender has to come dig u out

thats off the top of my head sorry g2g booty call

[Sarjen]

I do share this frustation in many battles when i play the soviets. Just now i experienced a harrowing view. My T80B1 is driving into the enemy. Many of its company are already dead from invisble enemies. C2 didn't help. Infantry spotted and relayed the contacts to the HQ, but the info didn't trickle down to the tanks. So i just drove my tanks to near (20m) range of the enemy, still seeing nothing. I have recorded this from both views. Playing the soviets is no fun at all.
 

View from the T80B1:

and the view from the US side:

 

[Grey_Fox]

1 hour ago, Sarjen said:

I do share this frustation in many battles when i play the soviets. Just now i experienced a harrowing view. My T80B1 is driving into the enemy. Many of its company are already dead from invisble enemies. C2 didn't help. Infantry spotted and relayed the contacts to the HQ, but the info didn't trickle down to the tanks. So i just drove my tanks to near (20m) range of the enemy, still seeing nothing. I have recorded this from both views. Playing the soviets is no fun at all.
 

View from the T80B1:

and the view from the US side:

 

Yes, tanks have very narrow fields of vision, which means that the closer they are to something the less likely they are to see it. Remember, you're essentially looking at things through letterbox-sized periscopes or viewports, and in the case of the gunner through a straw.

To overcome this, they should be used en-masse where possible.

Western doctrine is to fight with the commander at least partially turned-out. Soviet doctrine was to fight turned-in and en-masse.

[Sarjen]

10 minutes ago, Grey_Fox said:

Yes, tanks have very narrow fields of vision, which means that the closer they are to something the less likely they are to see it. Remember, you're essentially looking at things through letterbox-sized periscopes or viewports, and in the case of the gunner through a straw.

To overcome this, they should be used en-masse where possible.

Western doctrine is to fight with the commander at least partially turned-out. Soviet doctrine was to fight turned-in and en-masse.

While driving the company into the enemy the soviets killed two Bradleys with dismounted infantry from the scouts. The advancing tanks didn't even once see an enemy but got killed. All from 500m start to death. The US side in the above videos were not turned-out.

[Grey_Fox]

18 minutes ago, Sarjen said:

While driving the company into the enemy the soviets killed two Bradleys with dismounted infantry from the scouts. The advancing tanks didn't even once see an enemy but got killed. All from 500m start to death. The US side in the above videos were not turned-out.

It's also night-time isn't it? Bradleys have thermal optics, while the T-80s don't, they have IR lamps which only provide visibility out to something like 100m. And I'm not even sure if they're used.

Edit: yeah it's 4.30AM, so visibility would be very low for non-thermal optics.

Edited May 10, 2023 by Grey_Fox

[snarre]

 there is smoke around area , night time +smoke + nou thermals 

Edited May 10, 2023 by snarre

[Sarjen]

30 minutes ago, Grey_Fox said:

It's also night-time isn't it? Bradleys have thermal optics, while the T-80s don't, they have IR lamps which only provide visibility out to something like 100m. And I'm not even sure if they're used.

Edit: yeah it's 4.30AM, so visibility would be very low for non-thermal optics.

Yes. I sometimes just share the frustation with the thread starter. I think about the scenarios for Single Players as a game, where the player should have fun. While i design a scenario for the "red" player, i have to give the soviets every advantage i can think of: Short range encounters, not night time , many more tanks than "blue" side. It still is a game. I really appreciate the spotting and the whole mechanics underlying to give us a near realistic experience - but still the soviets feel at large at a disadvantage in the 80`s.

Edited May 10, 2023 by Sarjen