kuIoodporny 45 Posted October 2, 2013 Arma 3 Tanks Config Guidelines Feel free to use it according to our modding license. It roughly describes new features for tanks in Arma 3, You still would need to know the basics from A2/OA and how to set up cars in A3 first. Share this post Link to post Share on other sites
maturin 12 Posted October 2, 2013 (edited) Awesome, I'm sure a lot of potential modders just got excited. Thanks for the article. But if you don't mind me pushing the issue, could you explain just what is going on with tank damage handling at the moment? So we have a baseline of expected behavior before diving into the crazed world of config variables? For example, I just shot at the front, sides, rear and top armor of a T-100 using RPG-42 HEAT. I found some good stuff, like rear hull shots disabling the engine (yaayyyyy). However, it was usually mixed with disappointment, like rear turret hits disabling the turret but not doing any really critical damage. Top armor hits make no appreciable difference, despite what you said in the other thread. Top rear hull seems a lot like a normal rear hit, and top turret and hull are not really any better than front armor hits. Worst of all, pretty much every hit seems to injure the crew, always the same, minimal amount. Like 10% damage or some such. They won't even complain or turn red unless you ask them about it. It seems a lot like the crappy old ArmA feature of transferring a percentage of vehicle damage to the crew (the reason pilots kept blacking out in ACE wounds after their plane took small arms fire). TLDR: Enough ranting, sorry. Here is my real question: Why does it seem like AT weapons reliably damage hitpoint components (turret and gun) without penetrating? A couple hits to the front turret with light AT, and the tank is disabled in the components. (However, the T-100 can take a huge number of front hull hits because the gun and turret can't be damaged this way. As a result, it's much better to be hit in the hull than the turret, which is the opposite of reality. The crew and fuel live in the hull, after all). So are all these weapons penetrating and doing very light damage (like the KE ammo)? Or should the RPG be unable to penetrate the front armor, and is simply damaging components and crew too reliably and predictably? I assume the RPG is penetrating the rear and side armor. The bottom line is that whenever there is penetration, the death of some or all of the crew should be extremely common. And unless a hit penetrates, turret, gun and crew damage should be very uncommon. Non-penetrating hits should not even touch the hull hitpoints. And finally, a big problem is the all-or-nothing damage done to components. I can have gun and tracks red, yet both systems work fine. What happened to ArmA's nice intermediate stages of tank mobility kills? I think we would all love a penetrate-or-not debug tool, like ACE has. Edited October 2, 2013 by maturin Share this post Link to post Share on other sites
kuIoodporny 45 Posted October 2, 2013 (edited) It's constantly being worked on: - turret/gun hitpoints are still to be improved in nice fashion when the time allows - I focused on hull damage recently. Crew damage is problem, because: - turned in driver is not visible in fire geometry - explosions cannot hurt vehicle crews directly - need programmers to fix these issues. A hint: - I just managed to destroy T-100 with one RPG-42 from elevated position. - don't aim at engine, but area under the ammo storage (tank must turn his turret around). Penetrate-or-not debug tool: cursorTarget getHitPointDamage "HitHull" Intermediate mobility kills: - engine power does not decrease with damage received, - tracks reliability does not worsen with damage received, - need programmers to improve these issues. Edited October 2, 2013 by zGuba Share this post Link to post Share on other sites
maturin 12 Posted October 2, 2013 (edited) That's good news. I have liked what I have seen of hull damage so far. We're getting close the near-invulnerability to light AT that should be there. ACE's system was awesome because it was so unpredictable and bombastic (tanks running away on fire, catastrophic explosions that are now incredibly rare in A3). But the most important part was actually the crew damage. Are you guys at BIS completely opposed to scripted damage? Any penetrating hit of the crew compartment should be rolling dice and applying ad hoc damage to the crew from spall. If a shaped charge jet or sabot passes within a few inches of a damage proxy, that guy should be incredibly dead. Why exactly does the driver have no proxy? Is this the case for wheeled APCs as well? As for top-attack, does the turret roof have thinner armor? I guess the front hull glacis is going to be as tough as ever. ACE couldn't fix that either. Edit: I just got your one-round kill. If only that were likely to happen from all turret penetrations, and especially rear/top hits. (Although the external ammo bustle makes catastrophic kills unlikely, I'll warrant). explosions cannot hurt vehicle crews directly So when I smack the front of a Varsuk with an RPG, doing the vehicle absolutely no harm (progress!!!!!), why is the whole crew injured? Edit again: You say programmers are needed. I won't try to dig up gossip on things inside BIS, but I can't help asking: is help from other devs in the pipeline? Does BIS realize how much we care about this issue? Edited October 2, 2013 by maturin Share this post Link to post Share on other sites
kuIoodporny 45 Posted October 2, 2013 ACE's system was awesome because it was so unpredictable and bombastic (tanks running away on fire, catastrophic explosions that are now incredibly rare in A3). But the most important part was actually the crew damage. Are you guys at BIS completely opposed to scripted damage? Any penetrating hit of the crew compartment should be rolling dice and applying ad hoc damage to the crew from spall. If a shaped charge jet or sabot passes within a few inches of a damage proxy, that guy should be incredibly dead. BIS won't use scripted solutions, because they turned out to cause performance problems in A2. Why exactly does the driver have no proxy? Is this the case for wheeled APCs as well? The proxy is there, but it's hidden both from view and fire geometry when driver is turned in. It's a program limitation (though VBS2 has it working). As for top-attack, does the turret roof have thinner armor? Yes, but hitpoints aren't configured up to standards so damage received may differ. I'd improve it when time allows. Front armor is made thick enough to stop 120mm APFSDS at least on same level. Sloping means, if you attack tank from elevated surface you should be able to improve your odds of penetrating front armor. Share this post Link to post Share on other sites
maturin 12 Posted October 2, 2013 (edited) I just noticed the terrible accuracy of damaged tank cannons. That's awesome. As for the front armor stopping sabot, I think that point blank shots <1000m should penetrate, for reasons of both gameplay and realism. I think just about anyone you ask will agree with me. Edited October 2, 2013 by maturin Share this post Link to post Share on other sites
Smurf 12 Posted October 3, 2013 Looking at the PhysX LODs, no way to "soft" the tracks\suspension? Do this ingame and you will know what I'm talking about: Are these even PhysX related? And as far component damage goes, will stay as is? Or things like optics, eletronics (EMP...2035...) and more could be expected? Even I that act as a foot soldier see a complex system being more fun, imagine for tankers is even more. Share this post Link to post Share on other sites
kuIoodporny 45 Posted October 3, 2013 As for the front armor stopping sabot, I think that point blank shots <1000m should penetrate, for reasons of both gameplay and realism. I think just about anyone you ask will agree with me. Penetration can still occur, just not necessarily destroy tank totally because round speed is greatly reduced when it leaves armor plate. Please note damage has to be balanced so it works also for missiles and AT mines, not everything can be done ideally. Share this post Link to post Share on other sites
tempestriser 10 Posted October 3, 2013 Very good, i was waiting for these guidelines :) thanks zGuba. Share this post Link to post Share on other sites
Sealife 22 Posted October 3, 2013 Current public Oxygen version doesn't support showing correct name of this lod and displays only Geometry instead. Don't be afraid of having two geometries, it's going to be fixed with new tools. WHy is this Mantra still being pedalled without an explanation as to why it causes crashes if you binarise with currently available tools ? hopefully this can be resolved as there are already many ships and cars awaiting to utilize Physx but crash when Binarised with this Lod :( . Share this post Link to post Share on other sites
maturin 12 Posted October 4, 2013 Penetration can still occur, just not necessarily destroy tank totally because round speed is greatly reduced when it leaves armor plate.Please note damage has to be balanced so it works also for missiles and AT mines, not everything can be done ideally. Does it really require some sort of script to filter damage from different types of weapons and apply modifiers? And missiles and AT mines hurt the same no matter their velocity. I remember in A2 certain HEAT weapons would have different effects at different times in their trajectory. If that is still a feature, it needs to be taken out back and shot in the head. Share this post Link to post Share on other sites
kuIoodporny 45 Posted October 8, 2013 To all issues mentioned: we need a repro and ticket on FT. I'll gladly answer questions directly related to tank configuring ;) Share this post Link to post Share on other sites
soul_assassin 1750 Posted October 24, 2013 Has anyone really got PhysX to work with A2 tools? I'm just curious because mo matter what I try this all goes whacky and I'm forced to return to simulation = "tank". Share this post Link to post Share on other sites
koffeinflummi 96 Posted October 25, 2013 Has anyone really got PhysX to work with A2 tools? I'm just curious because mo matter what I try this all goes whacky and I'm forced to return to simulation = "tank". I got it to work. In order to prevent the vehicle from going batshit, lower the MOI or increase the dampingRateInAir of the wheels, then adjust enginePower/maxOmega/etc. The only problem that remains for me is that the dampers refuse to work at all. I know the engine and wheel values in the config are right since it's basically a copy of the Kamysh's one, the model fits all requirements that are listed in the article. I'm assuming the problem is in my model.cfg, although I've tried checking the damper animationSource with: X animationPhase "wheel_1_1_damper" That always returns 0, so I'm absolutely clueless. Is there a chance we could get a sample model with a full model.cfg? Share this post Link to post Share on other sites
soul_assassin 1750 Posted October 25, 2013 Has anyone really got PhysX to work with A2 tools? I got it to work. The only problem that remains for me is that the dampers refuse to work at all. so no then :P Share this post Link to post Share on other sites
koffeinflummi 96 Posted October 25, 2013 It is fully functional: The dampers 'work' fine, they just aren't visually represented. :) Share this post Link to post Share on other sites
dm 9 Posted October 25, 2013 WHy is this Mantra still being pedalled without an explanation as to why it causes crashes if you binarise with currently available tools ? hopefully this can be resolved as there are already many ships and cars awaiting to utilize Physx but crash when Binarised with this Lod :( . It needs new tools because binarize doesnt understand the physX lod or what to do with it during binarization. So when it converts the p3d to the ODOL format, it causes the pX lod to become corrupt, which is what causes the crash. The new version of binarize understands this lod, so you need the new tools in order to be able to pack. Simple. Share this post Link to post Share on other sites
Sealife 22 Posted October 25, 2013 It needs new tools because binarize doesnt understand the physX lod or what to do with it during binarization. So when it converts the p3d to the ODOL format, it causes the pX lod to become corrupt, which is what causes the crash.The new version of binarize understands this lod, so you need the new tools in order to be able to pack. Simple. Even simpler is for the DEVS to put your explanation in thier posts or the BIKI , this process has been splurged for several months now and frustrated many and i`m sure some have left due to that frustration. My question wasnt really about what the solution is but more about why are people with access to A3 tools writing processes for several months aimed at people with a2 tools :) I cannot register with wiki or i would have written a comment. Share this post Link to post Share on other sites
dm 9 Posted October 25, 2013 why are people with access to A3 tools writing processes for several months aimed at people with a2 tools You'd rather they didnt write any "how to" stuff at all? Because the tools will be released eventually... Share this post Link to post Share on other sites
ravendk 25 Posted October 25, 2013 You are right on this DM, it is good to have this explained, but isnt it a bit like teaching kids how to swim on a football field? :d: Share this post Link to post Share on other sites
dm 9 Posted October 25, 2013 You are right on this DM, it is good to have this explained, but isnt it a bit like teaching kids how to swim on a football field? :d: Seems more like briefing the kids on what to expect in the swimming pool while in the changing room. You're not quite ready to swim, but you can at least understand the theory. Anyho, the logic of the metaphor is flawed - the explained stuff works fine as long as you dont binarize the p3d.... Share this post Link to post Share on other sites
Sealife 22 Posted October 25, 2013 (edited) You'd rather they didnt write any "how to" stuff at all?Because the tools will be released eventually... I would rather you didnt have to come here to explain why the wiki gives instructions that are not actually relevant at this moment in time yes , unless your getting a bit of a thrill out of stating the obvious (to you) and telling people the instructions are complete as long as you dont do something that isnt mentioned in the instriuctions , i dont see the point of your replies to be honest , they do nothing to answer the original question abd merely answer another , which would be , how do i get a physx lod in game if i dont want to binarise it and release the source files unencrypted. either way , this has been my opinion and your opinion resulting from my question of why are the instructions not complete , you have your question answered , i am stil awaiting my answer. Edited October 25, 2013 by Sealife Share this post Link to post Share on other sites
soul_assassin 1750 Posted October 26, 2013 Seems more like briefing the kids on what to expect in the swimming pool while in the changing room.You're not quite ready to swim, but you can at least understand the theory. Anyho, the logic of the metaphor is flawed - the explained stuff works fine as long as you dont binarize the p3d.... thats messed up on so many levels :j: and no it doesnt work, binarized or otherwise. Either that is the problem or some steps are missing. Share this post Link to post Share on other sites
roni 11 Posted October 28, 2013 It's funny that we talk about armor penetration, when in game it's still all about amount of hits and this f***** "health bar" that just players don't see. So please stop talking bullshit about RHA and other shit, when it has nothing to do with game, and you can still destroy tank with grenade launcher and other shit 30mm-like cannons.. 1 Share this post Link to post Share on other sites
[aps]gnat 28 Posted October 28, 2013 Nice tut thanks zGuba Gives hints of the complexity ..... now just waiting for the plane and ship one :) Share this post Link to post Share on other sites