Still disliking Artillery

[Steiner14]

I doubt many would agree it's that important. And even if they did, there are still the other problems which are not fixed by being WeGo only. Nor the other features that wouldn't get in the game because we'd be spending a considerable amount of time on making this work.

Effectively the game would have to store two maps in memory; one with FOW deformations, one without. That's a huge footprint.

This is like trying to swat a fly with a 155 Howitzer. Sure, the current implementation isn't visually as pleasing as we would all like it to be, but there's practically no game play problem with the way it is now. Given everything in the balance, it's pretty much a no-brainer to leave it alone. Which is exactly what we're going to do.

Steve

Steve,

thanks for the reply, but i was referring to the protection issues with units that are entrenched, not terrain-FOW.

If placing a unit below surface and using a 2D-texture (or 3D-models being not part of the map?) for trenches, foxholes or pits right now is not possible because of the LOS-mechanism, that needs units to be placed on the 3D-ground, then a more complex LOS-mechanism not only checking the action spots at grond level but also several levels above each, allowing to see units only if their heads would be visible, could theoretically be possible with WEGO, while RT forbids it because of consuming too much CPU power for the additional LOS-checks.

[Sgt Joch]

MikeyD,

Don't know whether this would apply to CMFI or not, but it's certainly directly applicable to artillery dominance in CMBN. The Americans suffered through eleven months of ammunition shortages, some in the scary range where all but the most juicy and critical targets had to be ignored.

http://www.ibiblio.org/hyperwar/USA/USA-E-Logistics2/USA-E-Logistics2-9.html#fn1

JK, nice find. However, I disagree with the interpretation. What I glean from the article is that a supply shortage occurred in early-june 44, right after the landings and in october 44, when the Allies outran their supply lines. If you look at table 6, you can see there was more than enough ammo in theater to meet average daily expenditure. The problem was getting it to the units at the front. The article also states than even in the worst period, 12th army group had a minimum of 15-20 day's supply. During the heavy fighting in december 44 and january 45, artillery fire was "unrestricted".

No direct mention of Sicily, but the article does mention that the U.S. had built up a huge stock in all theaters by late 43, especially in North Africa.

[Battlefront.com]

Steve,

thanks for the reply, but i was referring to the protection issues with units that are entrenched, not terrain-FOW.

But the two issues are inseparable. If we have deformed mesh then we lose FoW. This has NOTHING to do with RealTime. CMSF had deformed mesh for foxholes and trenches and it worked fine in RT. It didn't have FoW for foxholes and trenches for the reasons I stated.

If placing a unit below surface and using a 2D-texture (or 3D-models being not part of the map?) for trenches, foxholes or pits right now is not possible because of the LOS-mechanism, that needs units to be placed on the 3D-ground, then a more complex LOS-mechanism not only checking the action spots at grond level but also several levels above each, allowing to see units only if their heads would be visible, could theoretically be possible with WEGO, while RT forbids it because of consuming too much CPU power for the additional LOS-checks.

LOS checks in RT are not holding back anything. Nothing. So your basic premise is technically incorrect. You need to forget about this because it simply isn't true.

We have had this discussion about alternatives many times already. Internally, during CMBN development, we had a debate that makes anything seen here in public look like idle chatter. The simple fact is there is no other way to go about it. Having soldiers visually below the surface of the mesh doesn't work for many reasons.

For example, what happens when a soldier goes prone (hiding)? He will disappear from the map because he'll now be below the surface of the map. Either that or the soldier will have to first elevate (or warp) to the actual ground level before going prone. Then lower himself (or warp) to the sub-ground plane when he goes to kneeling again.

Again, this horse has been beaten to death, butchered into small pieces, put into a food processor, molded back into the shape of a horse again so it can be destroyed another way. If anybody wants to waste their time going over this again, be my guest. But I'm done with it (again).

BTW, we can make adjustments to the levels of protection in the game currently because there are abstract values compensating for the visual inconsistencies. It's just in a few dozen discussions we haven't seen any reason to tweak them further than they have already been tweaked.

Steve

[LongLeftFlank]

Gotcha. Any chance, then, of some "camouflage" for designated units (e.g. adapt the dormant CMSF "pop density" function where Uncons are harder to spot at a distance until they commit a hostile act). Right now, it's basically impossible to ambush from fortifications since these are too easy to spot at range, even in heavy cover.

[Broadsword56]

Gotcha. Any chance, then, of some "camouflage" for designated units (e.g. adapt the dormant CMSF "pop density" function where Uncons are harder to spot at a distance until they commit a hostile act). Right now, it's basically impossible to ambush from fortifications since these are too easy to spot at range, even in heavy cover.

...and maybe connect that to a "camouflaged" model that would look a bit different from the uncamouflaged version. For example, cammo netting over an AT or MG emplacement, where you'd just see the gun poking out, low to the ground, and only darkness inside (the crew would not be visible but would be abstracted and shown as usual on the GUI).

So, you wouldn't have to deform any of the terrain mesh -- just give the player the graphical illusion that the unit is lower down/concealed, and back it up with some protection-level tweaks and/or spottable-distance tweaks under the hood, if necessary.

[Steiner14]

Having soldiers visually below the surface of the mesh doesn't work for many reasons.

For example, what happens when a soldier goes prone (hiding)? He will disappear from the map because he'll now be below the surface of the map.

Thanks.

If the complete hiding below surface creates problems, how about sinking them a bit into the surface? A compromise between fully standing on the ground and the 1:1 representation of the depth of trenches.

I mean sinking them so much into the ground, that in the case of hiding, they will not completely disappear so the AI doesn't get confused because it doesn't see them anymore?

Sinking them and ATGs into the ground, you maybe could make the foxholes more flat, too?

And maybe you could also offer scenario designers several levels of depth, modelling the level of entrenchement?

[Lanzfeld]

^^^ I like these ideas ^^^

[LongLeftFlank]

A bridge too far design-wise, IMHO, with respect to my learned friends. The eye candy of camo nets wouldn't be worth the graphical brain damage to design! What happens when the unit moves? Or takes a direct HE hit? etc etc.

I'd rather use the functionality that already exists (i.e. You rename the Population Density slider "Camouflage Level" and either all the defending side's units automatically qualify as "Uncon", or units put in the last AI group 16 do; that lets you give them all an Ambush or Hide order too. Speaking of which, can we get the shortest Ambush order shortened to 40m? 75m is a holdover from CMSF desert.

Again, I'm talking about something that already exists in the engine being adapted to redress the unrealistically easy spottability of what should be (generally) concealed bunkers, entrenchments, AT guns, etc. Bells and whistles just delay the fix.

[John Kettler]

Broadsword,

You mean something like this, as far as an entrenched, camouflaged emplacement? This one's from Sidi Rezegh.

http://nzetc.victoria.ac.nz/tm/scholarly/WH2-27Ba-fig-WH2-27BaP005b.html

Regards,

John Kettler

[John Kettler]

Sgt Joch,

Fair points, but I would point out that the text makes it clear that ammo rationing was applied at the gun level, as in, x rounds per gun. When we ran out of supply in November, for example, the 105s were down to 18.5 rounds per gun per day. I rather doubt the American players would like to have that for their entire FO allotment. In practice, it would be far less than that, because the commanders would make Midas look generous when it came to approving fire missions. So, while the 12th Army may've had stocks for twenty days' operations, the gunners and FOs were still screwed, because the ammo wasn't available at their level.

A longtime friend's dad was a medic in Patton's Third Army, and he told us about the frantic, triumphant pace after the breakout, then what happened when the gasoline ran out and mobility was lost. Casualties soared because we lost the freedom to maneuver.

Regards,

John Kettler

[JonS]

Maadi is approximately 750km from Sidi Rezeg (about the same distance as Caen to Cologne, or New York to Cleveland).

[poesel]

Effectively the game would have to store two maps in memory; one with FOW deformations, one without. That's a huge footprint.

So that means that we only have to wait two years when the average video RAM will have been doubled? I'm okay with that.

Sure, the current implementation isn't visually as pleasing as we would all like it to be, but there's practically no game play problem with the way it is now.

Well, there is the thing that you can instantly see (and often hear) when someone flattens a fence anywhere on the map. No game breaker of course.

[Broadsword56]

Broadsword,

You mean something like this, as far as an entrenched, camouflaged emplacement? This one's from Sidi Rezegh.

http://nzetc.victoria.ac.nz/tm/scholarly/WH2-27Ba-fig-WH2-27BaP005b.html

Precisely.

[Broadsword56]

A bridge too far design-wise, IMHO, with respect to my learned friends. The eye candy of camo nets wouldn't be worth the graphical brain damage to design! What happens when the unit moves? Or takes a direct HE hit? etc etc.

I know zero about the design aspects of this, so I defer to those who do.

But it would seem analagous to the way units with mortars change from a model showing it "deployed" to "not deployed."

If the cammoflaged/netted unit takes KIA/WIA, nothing changes except the GUI display about crew and weapon status.

If it takes a direct HE hit, you get a crater and the cross or skull-and-crossbones icons.

If the camouflaged unit moves, the cammo model disappears and is replaced by the standard model and the men moving with the packed up weapon, or the gun being hitched, etc.

Anyway -- probably not worth delving into because it sounds like it's not in the cards from BFC's POV.

[Killkess]

I just came home from vacation to see this thread:

In my opinion there are problems with the representation of artillery in CMFI and CMBN in different areas.

First off all the artillery we have in CMFI and CMBN is much to responsive and the player has much to flexible tools to call for fire. I still haven´t seen a proof that calling linear fire missions was common practice during WWII. I can imagine/believe that it was possible for preplanned bombardments but I really don’t believe that it was possible for strikes called by a normal FO "on the fly". One would have to register very precisely each gun and do different calculations for each barrel including shifting the fire of each gun individually within a linear target. And even if one do this calculations "on the fly" the error margin would be so great that the outcome would be anything but a perfect CM2 fire mission.

Second point is the density with which the artillery is used. It seems like many scenarios include too much artillery. And the artillery included is used on to small targets. One often sees that artillery is used against point targets or very small target areas which really wasn´t common practice either. Heck, following some discussions back in the CMBN forum the FOs ordered "FFE" when they got one spotting round in the closer vicinity (50m?) because the inherent errors/failures were so great, that it made no difference to further shift the aimpoint. From a realism standpoint I still prefer the idea of minimum, maybe even fixed, target areas for different support weapons to prevent the unhistorical "over density". Also the ammunition usage on the targets seems to be on the high side. Responding with artillery on each infantry contact seem to be too common. The inability of our pixeltruppen to react accordingly add to this.

Last but not least the whole fortification/trench/foxhole family doesn´t seem to give proper protection. I often see whole trench systems taken out even by small mortar fire which is at least "weird". Foxholes should already give enough protection against small mortars and artillery so that only direct hits should bring casualties.

[John Kettler]

JonS,

The chapter said Sidi Rezegh, so I thought that was sufficient. Didn't have time to read the text--was looking for sangar pics!

Regards,

John Kettler

[Paper Tiger]

Any chance, then, of some "camouflage" for designated units (e.g. adapt the dormant CMSF "pop density" function where Uncons are harder to spot at a distance until they commit a hostile act). Right now, it's basically impossible to ambush from fortifications since these are too easy to spot at range, even in heavy cover.

I could see a way to do this in the game without the need for special models. If the mission is a Probe or Attack mission, defensive fortifications benefit from a camouflage bonus. In an Assault mission, we could assume that some preliminary recon has already taken place and so there is no camo bonus for fortifications.

[John Kettler]

Killkess,

Target Linear IS the de facto default for artillery. Parallel sheaf, regular sheaf and open sheaf are the forms it takes. Parallel sheaf isn't strictly linear simply because the gun line is staggered, at least for the U.S. Otherwise, that would be linear, too. Converged sheaf isn't linear, but neither is it Target Circular. The link shows what artillery footprints should be as a function of the sheaf being fired. Be sure to read Unusual Patterns while you're there.

http://www.poeland.com/tanks/artillery/sheafs.html

Artillery typically went to FFE when within 100 meters of the target; this is because the shells from a given gun land in a box which follows the shell trajectory and whose principal axis is Range and whose narrow axis is Deflection. Combined, these two work to distribute the shells short of, on and past the target, as well as to both sides of the expected flight path. The box in which ALL the shells usually fall for a given weapon, at a given set of settings, is called the 100% zone. The link explains the various sources which contribute to why artillery works the way it does.

http://nigelef.tripod.com/errorsmistakes.htm

Regards,

John Kettler

[LongLeftFlank]

I could see a way to do this in the game without the need for special models. If the mission is a Probe or Attack mission, defensive fortifications benefit from a camouflage bonus. In an Assault mission, we could assume that some preliminary recon has already taken place and so there is no camo bonus for fortifications.

I'll take whatever I can get, but I'd sooner it be left to the scenario designer with the slider than hard coded like that.

[womble]

I'll take whatever I can get, but I'd sooner it be left to the scenario designer with the slider than hard coded like that.

Slider? Slider?! Heretic!

[noob]

I just came home from vacation to see this thread:

In my opinion there are problems with the representation of artillery in CMFI and CMBN in different areas.

First off all the artillery we have in CMFI and CMBN is much to responsive and the player has much to flexible tools to call for fire. I still haven´t seen a proof that calling linear fire missions was common practice during WWII. I can imagine/believe that it was possible for preplanned bombardments but I really don’t believe that it was possible for strikes called by a normal FO "on the fly". One would have to register very precisely each gun and do different calculations for each barrel including shifting the fire of each gun individually within a linear target. And even if one do this calculations "on the fly" the error margin would be so great that the outcome would be anything but a perfect CM2 fire mission.

Second point is the density with which the artillery is used. It seems like many scenarios include too much artillery. And the artillery included is used on to small targets. One often sees that artillery is used against point targets or very small target areas which really wasn´t common practice either. Heck, following some discussions back in the CMBN forum the FOs ordered "FFE" when they got one spotting round in the closer vicinity (50m?) because the inherent errors/failures were so great, that it made no difference to further shift the aimpoint. From a realism standpoint I still prefer the idea of minimum, maybe even fixed, target areas for different support weapons to prevent the unhistorical "over density". Also the ammunition usage on the targets seems to be on the high side. Responding with artillery on each infantry contact seem to be too common. The inability of our pixeltruppen to react accordingly add to this.

Last but not least the whole fortification/trench/foxhole family doesn´t seem to give proper protection. I often see whole trench systems taken out even by small mortar fire which is at least "weird". Foxholes should already give enough protection against small mortars and artillery so that only direct hits should bring casualties.

Jon S contributed some more realistic restrictions for artillery to an operation i'm intending to play, the restrictions were:

For artillery, only area fire can be used, and a fixed area fire circle based on a 35m spread pattern per gun, giving an area fire circle of for a 4 gun battery of 140m (+/- 10%).

On and off map mortars restricted to point fire only.

[womble]

Jon S contributed some more realistic restrictions for artillery to an operation i'm intending to play, the restrictions were:

For artillery, only area fire can be used, and a fixed area fire circle based on a 35m spread pattern per gun, giving an area fire circle of for a 4 gun battery of 140m (+/- 10%).

On and off map mortars restricted to point fire only.

Bear in mind, though, that as JonS also says, preplanned fire missions were vastly more flexible than the current options. What you lose at the beginning, you're compensated for in the course of the game by the flexible ad-hoc options.

[noob]

Bear in mind, though, that as JonS also says, preplanned fire missions were vastly more flexible than the current options. What you lose at the beginning, you're compensated for in the course of the game by the flexible ad-hoc options.

Well the obvious solution is to allow linear fire for pre planned strikes only, and with the same length restrictions as the diameter restrictions for area fire.

[womble]

Well the obvious solution is to allow linear fire for pre planned strikes only, and with the same length restrictions as the diameter restrictions for area fire.

It's a partial solution, yes. But preplanned arty could manage things like rolling barrages, with constant rates of advance, timed fires on specific targets (beyond 15 minutes) and other stuff you can't set up in the current preplanned fire options; JonS goes into more detail. To me, the "obvious" solution for your Operation is to actually preplan the fire missions, and assign enough TRPs to allow the plan to be carried out, but require that the players of the scenarios generated adhere to the fire plan, apart from the point/area missions you've already limited them to.

[Killkess]

Killkess,

Target Linear IS the de facto default for artillery. Parallel sheaf, regular sheaf and open sheaf are the forms it takes. Parallel sheaf isn't strictly linear simply because the gun line is staggered, at least for the U.S. Otherwise, that would be linear, too. Converged sheaf isn't linear, but neither is it Target Circular. The link shows what artillery footprints should be as a function of the sheaf being fired. Be sure to read Unusual Patterns while you're there.

Did u regognize that the targets are always perpendicular to the firing battery? In CM2 we freely choose linear targets no matter what is the target orientation. If you are out of parallel solutions u will get much bigger problems. With 100m are for FFE, how u want to ensure that each individual gun is properly aimed on a non parallel 200m target?

Artillery typically went to FFE when within 100 meters of the target; this is because the shells from a given gun land in a box which follows the shell trajectory and whose principal axis is Range and whose narrow axis is Deflection. Combined, these two work to distribute the shells short of, on and past the target, as well as to both sides of the expected flight path. The box in which ALL the shells usually fall for a given weapon, at a given set of settings, is called the 100% zone. The link explains the various sources which contribute to why artillery works the way it does.

And thats the primary reason why i think that non parrallel linear missions are not feasible given a "on the fly" mission. At least not with such a precission we see in CM2.

Jon S contributed some more realistic restrictions for artillery to an operation i'm intending to play, the restrictions were:

For artillery, only area fire can be used, and a fixed area fire circle based on a 35m spread pattern per gun, giving an area fire circle of for a 4 gun battery of 140m (+/- 10%).

On and off map mortars restricted to point fire only.

Which is a nice and good idea but i think the game itself should reflect these rules by itself. Which i think is a leftover from CMSF