Grog vs. Gamer

[deanco]

What?? Graphics AGAIN?? We just got done with the group hug over in the Gunny Bunny thread, and I come over here and find THIS??

A couple of points:

I don't agree with the definition of Group 3 either, as some other people have already eloquently stated. The most interesting thing in CM BY FAR is the tactical possibilities of the units, and I'm a Gamer. Please stop lumping us all in one category. You even said, "By this I don't mean accurately modelled TO&E or proper penetration for the long 88, I mean turrets flying off tanks or men reduced to bloody pulps or aircraft crashing onto the battlefield." I think most "Gamers" here would disagree with you there, I know I do. Even if we may not know the stats by heart, the realism is the reason the game is so finely balanced. Each unit has a unique and varied set of capabilities, because they are taken from real life. If one tried to "make up" a set of units like this, I imagine one would have to do a lot of thinking.

2) Graphics and gameplay go hand in hand. If it is nighttime and a vehicle is on fire, and a platoon walks near the flames, they should have a greater chance to be spotted. If and when dynamic lighting is implemented, this will add to CM's realism. Just 1 example, there are more.

------------------

DeanCo--

CM interface mods: http://mapage.cybercable.fr/deanco/

so many games...so little time

[Gun Dog]

I believe that CM has found it's niche market and that market is most of us. (Gorgs of one degree or another).

Other publishers have had the opportunity to create a game of this detail and shied away to appeal to the 'gamer' audience.

Let Hasbro and the others market to the masses. I dont like the gamey games that they offer anyway.

In the meantime, I'll be happy as a clam playing my CM. I am doubly happy because the game specifically does appeal to the small market of which I am a part of.

BTS, keep up the GREAT work. My next wargame purchase will be your product.

[Hamstersss]

Originally posted by deanco:

I don't agree with the definition of Group 3 either, as some other people have already eloquently stated. The most interesting thing in CM BY FAR is the tactical possibilities of the units, and I'm a Gamer. Please stop lumping us all in one category. You even said, "By this I don't mean accurately modelled TO&E or proper penetration for the long 88, I mean turrets flying off tanks or men reduced to bloody pulps or aircraft crashing onto the battlefield." I think most "Gamers" here would disagree with you there, I know I do. Even if we may not know the stats by heart, the realism is the reason the game is so finely balanced. Each unit has a unique and varied set of capabilities, because they are taken from real life. If one tried to "make up" a set of units like this, I imagine one would have to do a lot of thinking.

2) Graphics and gameplay go hand in hand. If it is nighttime and a vehicle is on fire, and a platoon walks near the flames, they should have a greater chance to be spotted. If and when dynamic lighting is implemented, this will add to CM's realism. Just 1 example, there are more.

Deanco, it is not the definition of group 3 that you disagree with, it's the title I chose for it. If you would prefer to refer to them as "Graphics Hound" or some other term that sounds more appropriate to you, so be it. The definition, however, does describe a main aspect of CM's appeal, regardless of whether you like the title. In my often wrong opinion, you sound like you're bigger on catagory 2 than anything else. Maybe we should just lose the titles and refer to them as catagory 1, 2 and 3.

Also, this wasn't meant to be another graphics debate, there is a much more interesting debate regarding rare and exotic vehicles brewing.

------------------

Car Wars sucked hard. In fact all Steve Jackson games sucked hard. What the hell was wrong with you, back then? Didn't you ever hear about the d20?!?!

No d20 back then fool. Cars Wars was only 4 dollars, what do you expect?

[Gun Dog]

Sorry. I meant 'Grogs', not Gorgs.

[L.Tankersley]

Originally posted by jshandorf:

But if BTS models tank shell ballistics accurately and realistically, then if you put a Panther A that you modeled from the paper specs into the game wouldn't this flaw appear by itself?/B]

Not without increasing the complexity of the ballistic and vehicle models by at least an order of magnitude (and probably two). Right now the "shape" of vehicles is abstracted into a single silhouette number, and armor configuration is abstracted into 9 faces (turret, upper hull and lower hull for front, side and rear). At present things like shot traps are modeled by setting a flag on certain vehicles to say "this vehicle has a shot trap so for hits on the frontal armor add a check for it to hit the trap and encounter less armor" (or something similar).

To actually model the penetrator-armor interaction at a level of detail that would allow things like shot traps to fall out naturally would be prohibitively difficult. You would need EXACT vehicle dimensions and armor thicknesses instead of the abstractions now used, and the ballistics model would have to be much more complex as well since you would have to track the shell to determine exactly where on the vehicle the hit occurred, and what the angle of incidence was with each surface contacted.

This would be such a fundamental change to the CM engine (and would impose such horrendous requirements for data and model building, not to mention computer resources for storing all this data and accessing it during play) that it is many, many times farther out of reach than any of GB's wildest graphics fantasies.

That's my opinion, anyway.

------------------

Leland J. Tankersley

[deanco]

Everything cool, Elijah. I'll shut up now.

------------------

DeanCo--

CM interface mods: http://mapage.cybercable.fr/deanco/

so many games...so little time

[L.Tankersley]

Originally posted by Elijah Meeks:

Deanco, it is not the definition of group 3 that you disagree with, it's the title I chose for it.

Personally, I disagree both with the title you chose for group 3, and (more importantly) with the blatant disregard for and snubbing of my group 4, you scurvy dog!

------------------

Leland J. Tankersley

[Hamstersss]

Originally posted by L.Tankersley:

Personally, I disagree both with the title you chose for group 3, and (more importantly) with the blatant disregard for and snubbing of my group 4, you scurvy dog!

Dammit, Tank, you may be a highly decorated CPX commander but I think we're all gamers. We can't seriously debate that some of us are more fun than others, can we? Well, maybe some of us.

As to your idea that we couldn't model ballistics more correctly, can't we move forward without going all the way to on-the-fly physics? Isn't there some kind of compromise between what we have now and the Cray-Necessary Armor Resolution System? (Not to be confused with the E10K Necessary Graphics System proposed by certain rodents)

------------------

Car Wars sucked hard. In fact all Steve Jackson games sucked hard. What the hell was wrong with you, back then? Didn't you ever hear about the d20?!?!

No d20 back then fool. Cars Wars was only 4 dollars, what do you expect?

[jshandorf]

Originally posted by L.Tankersley:

Not without increasing the complexity of the ballistic and vehicle models by at least an order of magnitude (and probably two). Right now the "shape" of vehicles is abstracted into a single silhouette number, and armor configuration is abstracted into 9 faces (turret, upper hull and lower hull for front, side and rear). At present things like shot traps are modeled by setting a flag on certain vehicles to say "this vehicle has a shot trap so for hits on the frontal armor add a check for it to hit the trap and encounter less armor" (or something similar).

To actually model the penetrator-armor interaction at a level of detail that would allow things like shot traps to fall out naturally would be prohibitively difficult. You would need EXACT vehicle dimensions and armor thicknesses instead of the abstractions now used, and the ballistics model would have to be much more complex as well since you would have to track the shell to determine exactly where on the vehicle the hit occurred, and what the angle of incidence was with each surface contacted.

This would be such a fundamental change to the CM engine (and would impose such horrendous requirements for data and model building, not to mention computer resources for storing all this data and accessing it during play) that it is many, many times farther out of reach than any of GB's wildest graphics fantasies.

That's my opinion, anyway.

I am not sure if you are correct here. I remember a lot of discussions where Steve and Charles have commented as to certain details of armor hit and penetration. As of right now I know that the shell's angle of impact IS calculated. And that Steve and Charles have model the surface of the vehicle (not down to the finest detail) but it is modeled in 3D IN the engine. It is NOT abstracted as you have put it.

Also I know that ricochets are computed as is the collateral damage from a shell breaking up.

I don't think you are right in saying there are 9 armor faces. Since the vehicles are in 3D inside the engine you don't have to compute facing. You only have to compute where the shell strikes and at what angle of incidence and at what speed, etc...

If CM can't realistically simulate a shot trap how much further must the engine go to do this?

Jeff

------------------

First of all, David, you stupid sot, if names were meant to be descriptive, everyone would have the, culturally appropriate, name of, "Ugly little purple person that cries and wets itself." -Meeks.

[Guest Germanboy]

Originally posted by jshandorf:

I am not sure if you are correct here. I remember a lot of discussions where Steve and Charles have commented as to certain details of armor hit and penetration. As of right now I know that the shell's angle of impact IS calculated. And that Steve and Charles have model the surface of the vehicle (not down to the finest detail) but it is modeled in 3D IN the engine. It is NOT abstracted as you have put it.

Jeff - I believe it would not be possible. IIRC the debate about the Tiger mantlet correctly, there are some underlying abstractions. I may well be wrong though. I think Leland covers this point quite well and indeed correctly.

Jeff H. - if your statement was correct (i.e. that you don't need field experience except in a few cases), then why does the Panther A have a shot-trap at all? Surely the German engineers would have noticed it while designing it, and done away with it before it hit the road, as they did in the later models. Unless Steve and Charles actually reconstruct the tanks, and preferably go across them with a tape measure, I believe you need real life reports to see how they would perform, if you want to do it realistically. You may get a good approximation without that, I would agree with that.

------------------

Andreas

Der Kessel

Home of „Die Sturmgruppe“; Scenario Design Group for Combat Mission.

[L.Tankersley]

Originally posted by Elijah Meeks:

Dammit, Tank, you may be a highly decorated CPX commander

I can neither confirm nor deny that statement at this time.

but I think we're all gamers. We can't seriously debate that some of us are more fun than others, can we? Well, maybe some of us.

In a sense. My point was really that not everyone that plays/enjoys CM need fit into the three categories you defined, and your initial post seemed to be focused on addressing the needs of these specific groups to the exclusion of others.

[On reviewing that initial post, I see that may not have really been a fair reading. But darnit, I still don't like using the term "gamer" to refer to people whose primary focus is graphic enhancements.]

As to your idea that we couldn't model ballistics more correctly, can't we move forward without going all the way to on-the-fly physics?

Of course; I was referring specifically to the "shot-trap" argument. Of which I will speak more in my next post. You do need to have caution, though, when adding fidelity to one area of a simulation to be sure you don't invalidate assumptions made by other areas. To raise the spectre of a past discussion, it might be argued that improving the ballistics model without explicitly modeling optics and sighting systems would actually decrease realism.

------------------

Leland J. Tankersley

[dieseltaylor]

Having read the thread I am conscious of a niggling feeling all the way from the initial post.

The feeling that the game is a success because - not because of its graphics, not because of its realism or modelling of the period - it is a success because it is an intellectually satisfying challenge. Second guessing the AI or a human - or failing - to makes it addictive particularly as you have the 1 minute film slowly revealing the success or failure of your own decisions.

So what are the main feature - playability and suspense - and all of this in a reasonable time frame. The way I look at it is that all else should be viewed as subsidiary to this.

More vehicle types nice but not necessary,

better graphics nice but what trade of in network and PBEM play times, completely accurate armour simulated - nice but not necessary.

I think Mr Meek in his initial post missed a trick by ignoring a group who play for enjoyment. I believe that actually probably includes everyone!!! but the subjects of play "balance" and game "functionality" as final arbiters of what should be included in a game seemed to be absent.

And yes I believe the designers have got it right. Yes it seems there are interest groups who want this that and the other included - and yes I am praying that this game & BTS avoids a computer bloatware edition a la ASL. Amen.

Regards

p.s love the maths and syntax grogs I LOL

[Michael Dorosh]

Originally posted by Germanboy:

Jeff H. - if your statement was correct (i.e. that you don't need field experience except in a few cases), then why does the Panther A have a shot-trap at all? Surely the German engineers would have noticed it while designing it, and done away with it before it hit the road, as they did in the later models.....I believe you need real life reports to see how they would perform, if you want to do it realistically

The best evidence I ever saw of the Panther's shot trap was Radley Walters, a WW II armour vet, sitting on a Panther in a French museum, showing a television interviewer how the shot trap worked. Allied tankers, in 1944, knew that the shot trap was the only chance they had on a Panther. If the Panther never saw combat, I wonder whether the ability of the 75mm round to bounce off the underside of the mantlet and into the driver's head would have been noticed by historians (or game makers).

------------------

http://wargames.freehosting.net/cmbits.htm

[Guest Big Time Software]

I'll just chip in a few bits worth...

I for one think it is hard to peg people into one slot anymore. It was easier when graphics were generally not all that exciting, but CM kinda wrecked that. In other words, someone can be VERY concerned about historical accuracy AND graphics, but maybe not.

For simplicity sake, here is my take (based on Meeks and Tankersley's addition:

1. Historical - Main concern is fighting historical battles with historical forces under hitorically correct conditions.

2. Accuracy - Same as above, but also likes to fight "what if" battles as well as ones that are sculpted from real life ones.

3. Gamer - Is concerned less with accuracy and more with the game as a whole being fun to play. They don't mind a historical corner cut here and there if the game can be made "funner" as a result. In contrast, #1 and #2 types would raise a big fuss about it.

4. General Gamer - Is not concerned with accuracy at all, really. Just wants to have a fun game that offers rich and CONSISTANT gameplay. What I mean by "consistant" is fair, predictable behavior that is not screwed up by bugs and fun killing features/designs (oh... like say a single winning strategy that stagnates play).

I suspect that all of us are a little bit of each, but lean more towards one than the others. Note that I didn't put in "graphics" into any of the definitions. I think that is a seperate variable. I for one think of myself as a #2, but I also think that graphics are critically important. In other words, if there were a game EXACTLY like CM in terms of the historical accuracy and even gameplay, but was 2D... I would be far less interested in it than CM.

Just my two bitties worth,

Steve

[L.Tankersley]

Originally posted by jshandorf:

As of right now I know that the shell's angle of impact IS calculated. And that Steve and Charles have model the surface of the vehicle (not down to the finest detail) but it is modeled in 3D IN the engine. It is NOT abstracted as you have put it.

Much of the "3d modeling" is done with textures. As you mention, the actual 3d polygon-boxes that the graphics engine draws the textures onto aren't super-detailed. The angle of impact is calculated, but I'm virtually certain that it is not done using the internal 3d polygonal model. There may or may not be multiple mathematical models used for determining hit location on different vehicles; someone from BTS would have to speak to this.

It would be possible to calculate which polygon of a vehicle model is first intersected by the path of a round and then determine where in that polygon the impact occurred, but I'm comfortably sure that CM is not doing this now. For one thing, I'm pretty sure that there is a random "to-hit" roll made internally against the calculated hit chance when a shot is fired. If the shot hits, then a determination is made as to what armor facing is hit (based on target position, silhouette, orientation, elevation and so forth). If it misses, then a "miss" trajectory for the shell is generated. After all this is done the graphical representation of the shot is displayed. This makes no sense if the actual ballistic trajectory of the round is being used to determine whether it intersects the target or not (although the ballistics equation does come into play in determining angle of impact and energy of the penetrator, for example).

I don't think you are right in saying there are 9 armor faces. Since the vehicles are in 3D inside the engine you don't have to compute facing. You only have to compute where the shell strikes and at what angle of incidence and at what speed, etc...

I thought that the armor values in the data screen were essentially all there are in the processing engine (other than some tweaks that are probably handled in-code, like the variable-thickness Tiger mantlet that was added in a recent patch). As I mention above, though, the engine could be modified to use the actual polygons to determine hit location (I don't think it does now). I would say that this would actually be a pretty valuable enhancement, actually, because it would really give you varying degrees of being "hull-down." Right now I'm pretty sure hull-down is just a binary determination but BTS could tell me I'm wrong..

If CM can't realistically simulate a shot trap how much further must the engine go to do this?

A shot trap results from a peculiarity of the 3d geometry of the tank. For example, a shot hitting low on the turret front might tend to be deflected downward into the relatively thin top armor of the hull rather than in some other direction, and certain hull/turret geometries could make this a relatively likely occurrance. Right now, BTS calculates a hit location, determines the effective armor protection at that point and determines whether the shot penetrates or not. To really model a shot trap, you need to (1) replicate the geometry that results in the real-world effect and (2) replicate the physics that lead the geometry to have that effect. My contention is that neither the geometric nor the physics models in CM are anywhere near the level of detail required to model a shot trap "naturally" in this way. (You could probably gin up something that would demonstrate the same basic result of a shot being deflected off the turret into the top armor, but actually getting it to happen at even roughly the right frequency would be a real bear and require extremely detailed models of the vehicles. And there are a lot of vehicles.)

------------------

Leland J. Tankersley

[Jeff Heidman]

Originally posted by Big Time Software:

I suspect that all of us are a little bit of each, but lean more towards one than the others. Note that I didn't put in "graphics" into any of the definitions. I think that is a seperate variable. I for one think of myself as a #2, but I also think that graphics are critically important. In other words, if there were a game EXACTLY like CM in terms of the historical accuracy and even gameplay, but was 2D... I would be far less interested in it than CM.

Just my two bitties worth,

Steve

I would concur with Steve except to say that in theory the 2D, 3D element is unimportant. What I care about is immersiveness. If you can immerse me with a 2D game, then that would be fine with me, although that is going to be difficult to do after CM.

I am looking with interest at the new 2D tactical wargame that Matrix Games is working on. What it might lack in 3D it might just make up for in a higher level of detail in the sprites.

Of course, if it does not have the attention to historical accuracy and detail, all the pretty in the world won't make a difference.

Jeff Heidman

Jeff Heidman

[Guest Big Time Software]

As for the discussion about the level of detail underneath the graphics layer, there is a LOT of info already posted on it. But in short:

CM has no exact concept of a vehicle's shape. It is displayed graphically, and visually behaves as one would expect it to (turret rotates, at such and such a speed, etc.), but there isn't some underlying model that knows a Panther's mantlet is Xmm by Ymm, curved in such and such a manner. Instead, there is a table based representation of the Panther, much like you see in a detailed book on the subject. Maybe a little more. X plate at Y angle of Z thickness.

When a "hit" is determined, CM figures out which part of the vehicle was likely hit. Not the EXACT part, but the general part described in the table. This is determined by the vector of the shot. Damage is then determined by looking at the particular part hit, the angle of the impact in relation to it, thickness, type of shot, etc., etc. This then determines the chance the round penetrates, bounces, breaks up, etc. It also determines the chance of hitting a shot trap.

So think of the underlying model as one that uses odds and not mm for mm representation. Currently computers are not capable of simulating things to this level. Fortunately, the system that CM uses is more than a fair approximation of combat. The critical thing is not WHERE the shot hits (since how can say where a unique shot should land) but HOW and under what circumstances. CM has all that detail allowing the level of realism, abstractions and all, to be very high.

By the time we go to rewrite Combat Mission's engine we will most likely be able to make this stuff a little less abstracted. But we will have to wait and see

Steve

[Michael Dorosh]

Here is the litmus test to see what kind of gamer you are.

If your skin crawls when you see the "word" (for lack of a better term) 'zook' used in a post, you know you're a grognard.

If your skin crawls when you notice someone insist on using the complete and cumbersome designation "2.85 inch rocket launcher" every time he discusses US LATWs, you know you're a gamer.

If your skin crawls when you see either term used in casual conversation, you are probably a healthy mix of both.

[Gremlin]

Steve wrote:

....I suspect that all of us are a little bit of each, but lean more towards one than the others. Note that I didn't put in "graphics" into any of the definitions. I think that is a seperate variable. I for one think of myself as a #2, but I also think that graphics are critically important. In other words, if there were a game EXACTLY like CM in terms of the historical accuracy and even gameplay, but was 2D... I would be far less interested in it than CM.

That whole post was music to my ears Thanks in part to all the "noise" in some of these graphics/gameplay threads, I thought my take on some of these issues was substantially different from BTS's. I'm very happily proven wrong.

------------------

New to Combat Mission?

Visit CM Boot Camp at Combat Missions for tips.