Talk:Game Treatment

From Code43 Urban Warefare Wiki wiki.code43.net

I still need to go through the document again adding footnote numbers for the references links. --- JWB;2009-05-31

Gameplay Highlights Comments

Bullets ricochet - While I like this idea and the challenge it can create, programming it, I am worried that players will start to complain when they are hit by a ricochet from a player they can not see as he is not in the line of sight. It will create the impression that the other player was cheating even if it was a ricochet that bended the bullet path and hit him. --Dekki 15:16, 13 October 2009 (UTC)

One more thing to be aware of is how different surfaces affect (produce?) the ricochets - Take smooth concrete and brick wall (made of solid bricks) for example. Both have similar density and hardness (roughly). But there is one important difference - smooth concrete will ricochet bullets more or less evenly because of the even surface; unfortunately this isn't the case with bricked wall. Bricked wall has smooth (hard) surfaces which represent the major part of the wall and shallow grooves that represent the minor part of the total surface. And these grooves are the source of the problem because at sharp angles they dispose the sided of the brick to the bullet. This means that bullet won't hit the surface at the same angle it was shot from the weapon, but at considerably greater one - (angle from which the bullet was shot at the wall + 90°). The consequence is that the bullet will rather stop than ricochet. This means that only the smooth part of the wall will deflect (ricochet) the bullets. That's why I propose a special ricochet coefficient for different surfaces. If the brick-wall surface has 90% of even-area and 10% of the groove-area, the calculation of ricochet would be something similar to: Equation 1:

0.9 (amount of surface that can produce ricochets) * material ricochet coefficient (tells us how effectively the even area deflects bullets - will/would be tricky to get a realistic value!) * cos(angle of attack) * ricochet coefficient for the bullet-type (lead core, iron core, etc) * speed of the bullet * (mass of the bullet)^(-1) 

for speeding the calc process we could simplify and join some coefficients - mass and bullet type; or material ricochet coeff. and amount of surface that can produce ricochets for example.

Of course if this won't hurt the speed of the engine. --sxy 14:12, 24 October 2009 (UTC)

We're always going to deal with people screaming "cheater" when they died for completely legitimate reasons, I don't feel that danger is reason enough to elminate a feature like ricochet.

I agree with Marty's post on the forum http://forum.code43.net/viewtopic.php?f=8&t=269 that we should keep this simple. Remember, our goal is the illusion of reality. We're making a vido game, not a simulation. To better model our illusion on reality, we should research actual ricochet conditions (ballistics tables might provide some data). We might be surprised to find such information readily available online somewhere. If not, it might be obtainable elsewhere without having to resort to a testing lab. ---JWB 14:53, 4 November 2009 (EST)

Uninhibited movement - I do not understand how a player would control this. Free look can be triggered by a key and the player could run and look in a different direction. But altering the stance just sounds too complicated for the player to do during a game. Unless he spends his time setting it up for the ultimate camp position, it feels that this feature would hamper game flow instead of improving it. The quick keys for predefined stances sounds good but not the free form. --Dekki 15:16, 13 October 2009 (UTC)

Have you checked out the pages describing Uninhibited Movement in the "Best of Tactical" article on the dslyecxi.com website? He provides a decent description along with several videos. Take a look http://dslyecxi.com/bestoftactical.html#I and let us know what you think. ---JWB 14:53, 4 November 2009 (EST)

Statistics tracking - This should be server side so that we can give players awards to keep them in the game. e.g move this to Online Highlights --Dekki 15:16, 13 October 2009 (UTC)

It should be both. Each server should track its users statistics, and the local one described in the game highlights section is the combination of each server's stats. This provides a broader, more accurate, view of one's performance.

The local copy of the stats could have multiple sorting functions: by server, all together, game type, etc., to provide a better understanding of how one performs. ---JWB 14:53, 4 November 2009 (EST)

Damage based on bullet type - While the GT talkes about accurate weapon and damage behaviors there are no explanation how this would work. I propose calculating damage base on bullet type instead of weapon types as most game do. In a bullet type system the base damage per hit is calculated on the bullet used for that weapon. Barrel length only affects the aim property and not the amount of base damage. As an examples, the glock hand gun and the mp5 submachine gun both use the 9x19mm bullet. These would therefore have the exact same base damage, while the mp5 with the longer barrel would have better aim. Similarity a sniper rifle that uses the 5.56x45mm bullet would do the same base damage as the m4 assault rifle. Many games instead increase the sniper rifles base damage because they do not have auto fire and are harder to hit. --Dekki 15:16, 13 October 2009 (UTC) (this was originally by Dekki, right?)

"While the GT [talks] about accurate weapon and damage behaviors there are no explanation how this would work." That's correct, and exactly how it should be. As described elsewhere, the G.T. is not a document that contains such explicit details. Those types of details appear in the Game Specification, and the various architecture documents for the features in question.

This discussion about damage, as well as sxy's below, should be removed until it can be placed in its proper document: either moved to the forum for the time being, or saved to each person's collective hard drive.

Again, this type of detail belongs in the Game Specification, and possibly not even there, belonging more appropriately in the architecture documents that accompany these features (but I'd be happy with it being in the Game Spec.). ---JWB 14:53, 4 November 2009 (EST)

//-sxy on Damage based on bullet type Not quite correct. Although both (M4 and M16) share the same caliber and bullet-type there is huge difference not only in external but also in terminal and wound ballistics. The reason for this is the difference in barrel length - 508mm for M16 and 368mm for M4. This different barrel length results in different muzzle velocities - 880m/s (1550J for 4g bullet) for M4 and 975m/s for M16/A1 (1900J for 4g bullet) and consequently different kinetic energies - 1550J for M4 and 1900J for M16/A1 (the same 4g bullet). This difference means that the maximum range at which the standard 4g 5.56 NATO bullet will still fragment is 45m for M4 and 120m for M16 which is huge difference! (take note that there is the same critical speed needed for fragmentation and this critical energy is specific for each bullet-type and doesn't change with different barrel lengths.) (source: Wikipedia http://en.wikipedia.org/wiki/5.56x45mm_NATO). Fragmentation creates much greater shock and wound than a simple pass-through and the same holds true for the energy. If the bullet will only wound the target (passage through one lung for example) and it won't create the shock, then there is a big possibility that the target will keep shooting at you (but will still die in 45mins if shot through 1 lung).

When it comes to the pistol rounds, the difference between different barrel lengths is not as important as for rifle rounds. The main reason for this is that propellants in pistol rounds burn faster. (the burn time is shorter!) The reason for this is simply that pistols and SMGs have short barrels and there would be no sense in burning propellant after bullet has left the barrel. That is why there is very small difference in muzzle speed between MP5 and glock19 - almost all propellant has already burned when bullet leaves the barrel of glock19. MP5 gets better spin and is therefore more stable and has better ballistics, but since the bullets speed and kinetic energies are almost the same for MP5 and Glock19, the wound ballistics and the shock that they produce are similar. --sxy 16:34, 24 October 2009 (UTC)

Online Highlights Comments

Online account / private key registration - Create a unique key for every player to keep track of them, prevent user stealing, allow effective banning and other actions. Idea for a solution (first draft) is in my own wiki. http://dejan.se/wiki/index.php/First_person_shooter_(fps)_anti_cheat_concepts --Dekki 15:16, 13 October 2009 (UTC)