Ragdoll

[smellyjelly 0]

Now, all of these take pretty long, and that might not be very feasable in ArmA due to huge distances, you cant make sure you have a hit and you still see the guy standing (WTF, BUG?! reactions), but still a lot of diversion in this would be great.

I see that as more of a pro than a con, because it forces me to be more aware of the enemy and his movements.

Anyways, overall ragdoll isn't a big deal to me, especially if BIS add's more animations.

[sic-disaster 311]

Now, all of these take pretty long, and that might not be very feasable in ArmA due to huge distances, you cant make sure you have a hit and you still see the guy standing (WTF, BUG?! reactions), but still a lot of diversion in this would be great.

I see that as more of a pro than a con, because it forces me to be more aware of the enemy and his movements.

Anyways, overall ragdoll isn't a big deal to me, especially if BIS add's more animations.

Yeah i see it as a definite pro as well, i love those kind of lengthy animations as they testify of a lot of work going into them, and they are just plain fun to watch.

But it might be confusing, though that wouldnt be such a bad thing either but still, not everybody would like it i think.

[dentist guba 0]

maybe they could use something like euphoria (the dynamic animation system being used in GTA IV). it would probably be expensive and use a bit of processing power but it should certainly solve all the problems with fixed animations and may even make it easier to navigate buildings e.t.c.

[Enixj7 0]

Ragdoll is a Must If the engine is not upto scratch thats poor coding, if a Playstation 2 game like Mercenaries can run large levels with full physics support on 32mb/294 MHz then theres no excuse for no support on a single core / 512mb, forget dual core/quad core! And like the original poster said they should always be an option to turn it off.

The typical argument is performance trade-off, can ragdoll be simulated on a such a scale ? Question is such the scale necessary, I find playing Map Facts South Sarhani much better then playing United Sarhani performance wise for the obvious reasons. In either case physics engines have been out for the better part of 5 years and have been heavily optimized, its time BIS took note.

[Grey Wolf 0]

what they could do is something similar to Turok (the new one) where the bodies are ragdoll for a certain amount of time then they become static objects in what ever position they were last in. This may stop the lag issue and make the deaths more realistic looking. However they also could get in some odd positions in turok at least.

[Uziyahu--IDF 0]

Most UbiSoft Tom Clancy games have really good ragdoll physics.

Ragdoll calculations are usually calculated client-side, and there could be something in the code that would decrease the calculations per second dependent on the number of units killed at the same time.

The delayed death animations in ArmA really don't compare to the immediate ones in OFP:R, in my opinion.

It's high time there was ragdoll physics in the OFP franchise.

Some complain about unrealistic ragdoll deaths, but compare that to the stiff mannequin reactions of current animations to the concussive force of an explosion. The latter is awful looking.

[dentist guba 0]

Ragdoll is a Must If the engine is not upto scratch thats poor coding, if a Playstation 2 game like Mercenaries can run large levels with full physics support on 32mb/294 MHz then theres no excuse for no support on a single core / 512mb, forget dual core/quad core! And like the original poster said they should always be an option to turn it off.

The typical argument is performance trade-off, can ragdoll be simulated on a such a scale ? Question is such the scale necessary, I find playing Map Facts South Sarhani much better then playing United Sarhani performance wise for the obvious reasons. In either case physics engines have been out for the better part of 5 years and have been heavily optimized, its time BIS took note.

i don't really think the scale makes much difference anyway, it's probably more to do with how much is happening at once on screen which would slow the game down. Plus, battlefield 2 has 60 players max, all with ragdoll e.t.c. if the ragdoll is set up properly the games shouldn't have to do a lot at once, it would be unlikely to have everyone dying at once so if the ragdoll is only activated when someone dies it should be fine.

GTA4 has euphoria acting on loads of peds at once so i think it is possible for ragdoll in this game.

[mr.g-c 6]

The IMHO major Problem with Ragdoll and other popular systems like that is that the hit/death effects and animations are mostly extreme unrealistic up to a level, in my opinion unfitting for Arma2 which is called " The Ultimate Military Simulation".

Also in the Games i played with Ragdoll-like effects, the bodys often felt way to lightweight - just like if they were filled with feathers.

But yes they should really do something about the current ultra-stiff and clunky (death and hit) animations - they are a no-go for Arma2 i think.

Best Regards, Christian

[baddo 0]

It is not the fault of the technology if it is used wrong.

You can make a ragdoll behave unrealistically or you can make it behave realistically.

Look at your body, what does it have. Now go and simulate it in a game. You will find that you need ragdoll physics. It is a problem to tune it to behave realistically.

I think the main realism problem some ragdoll implementations have is that they have little or none simulation of the natural resistance, stiffness of the body, i.e. the muscles and else which act as resistance to the movement of the joints. So maybe they are more like skeletons without any flesh.

Nothing in ragdoll technology forces you to make ragdolls behave like a skeleton and not like a body with flesh.

[dentist guba 0]

the best examples of realistic ragdoll i have seen were in Brothers in arms and ghost recon, they seemed to have a good amount of weight and they were last gen games.

[yourgrandma 0]

I agree with GRAW was an ok game but had some great effects on how bullets hit enemies such as if you shot them while running they would tumble and roll but it is animated assisted with ragdoll. Or even if you shot someone and it didn't kill them they would get knocked to the ground was great (but didn't apply to you???)

But it put it plainly ragdoll is useless and usually would be completely unrealistic in a game like arma. Guess what when people die they fall to the ground very quickly they don't tumble for 10 seconds or fly 15 feet even with a grenade. I hate how people like to complain a lot about crashing physics in racing simulators or flight simulators ... but you know what they are simulating the experience not the crash! Arma is supposed to be a combat simulator not a death simulator.

[baddo 0]

But it put it plainly ragdoll is useless and usually would be completely unrealistic in a game like arma. Guess what when people die they fall to the ground very quickly they don't tumble for 10 seconds or fly 15 feet even with a grenade.

Oh you already saw BIS ragdoll in action? How else can you know how it would look like.

There is a problem with this kind of thinking. It is that an assumption has been made that it is the ragdoll physics technology itself which makes things look wrong, always. I'm telling you it is the developer who makes it look wrong or right.

Ragdoll physics would be only one component of the total unit system. It would be combined with animations. Also there are more than one currently-known ways to implement ragdoll physics. Some ways are better than others don't you think.

[AWDrift 0]

Another thing is it's pretty hard to find a shooter without ragdoll now. If ArmA 2 comes out in 09 or 10 with death anims it's going to look dated in that aspect. It's kind of like shading in that it's not necessary, but having it in the game is part of moving forward with the rest of the gaming world.

[Lou Montana 101]

Exactly... what if they kept OFP shadows, the ones that didn't cast on close walls ?

there are people like that :

1. OMG add every poss|b13 n3w f3atures (ok, very bad leetspeak) (== add full blowning ragdoll of the death)

2. No, crap, every new thing is horrible, just add new scripting functions (let the community do the job), do the engine, even if it's graphically horrible if you can add steerable parachutes and all new realist things, it's OK (== no ragdoll !)

3. use the new technology to at least provide immersion to the player, and work on stability of the game (== ragdoll would be great, but tweak it correctly)

I personnally am looking for a totally stable version (a bit enough of ArmA because of that, but still playing <span style='font-size:7pt;line-height:100%'>because I'm addicted to...</span>) ;

I say too that if they can implement ragdoll, as they are requesting quad-core in less than a year, if they can they should.

Now three ways of thinking for ragdoll people :

1. oh crap, I see ragdoll used in other games, people will fly to the sky for at least 30s with a single hand grenade !

2. OMG Y34H blow every people I shoot at and throw them 20m back !

3. ragdoll can be a good visual effect, but it HAS to be tweaked correctly (i.e good limb weight, nice transition between anim and calculated death position, etc)

I let you guess in what kind of category I am. and I think other should be (yes, I want everyone with my opinion, hahaha)

but, the final problem is, will it be possible to implement the ragdoll technology in such a huge visualdistance game like ArmA ?

CoD4 will have a maximum of 10-15 bodies, and ArmA can take 15x more ; should ArmA do disappearing bodies ? *Yuck*

As the soldier body can be used (as cover, as ammo pouch, as everything you want), it has to be sync by the server.

Now the true question : Can an ArmA2 server keep going on with sync hundred bodies through 100km2 of map ?

[baddo 0]

As the soldier body can be used (as cover, as ammo pouch, as everything you want), it has to be sync by the server.

Now the true question : Can an ArmA2 server keep going on with sync hundred bodies through 100km2 of map ?

I proposed to delay the synchronizing as much as possible in

http://www.flashpoint1985.com/cgi-bin....;st=435

Something along those lines. I'm actually not satisfied with that post, but the main point I made in it stands: delaying synchronization as much as possible, or not doing it at all if by any means possible.

Everything does not need to be synchronized all the time I think. A lot of network traffic can be eliminated completely by carefully choosing what to synchronize and when.

The server should have the only 'correct' state of the game at any one time.

The clients will be corrected by the server as required. But ONLY as required.

It will be interesting to see how far BIS takes ArmA 2. I am a little bit pessimistic about actually seeing ragdoll physics in that game, but let's wait and see what comes out.

[Big Dawg KS 6]

Another thing is it's pretty hard to find a shooter without ragdoll now. If ArmA 2 comes out in 09 or 10 with death anims it's going to look dated in that aspect. It's kind of like shading in that it's not necessary, but having it in the game is part of moving forward with the rest of the gaming world.

And it's pretty hard to find shooters that don't revolve only around scripted narratives and cinematic gameplay. Whether or not such a design concept is effective, ArmA (II) is not one of those.

In today's gaming world, the borders that define genres are dissolving. There are certainly instances where ArmA plays more like a real-time strategy (or real-time tactics) game than a shooter, and when was the last time you saw ragdoll in an RTS? This genre focuses on a larger scope than most shooters, and narratives and cinematics aren't driving characteristics of this type of gameplay.

Saying that standardizing certain technologies is part of moving forward with the gaming world is quite inaccurate. As the gaming world moves forward, more diversity appears in the tech as more advancements are made and ideas are explored. While it's true that a lot of game developers try to follow a standard, it's not so much the gaming world advancing but the business aspect of it staying put.

The gaming world is not moving forward in a linear path either, rather it is branching out in all different directions. There are still audiences for games like ArmA, and many of them are hoping BIS is doing their best to keep it from merging with those other branches simply out of obligation. Ragdoll isn't somethign a game like ArmA is in dire need of, and if the need should arise then let it come (and be resolved) on its own terms and not simply because everyone else is doing it.

As a final note, if someone believes they need ragdoll to satisfy their desires, then it's like you said, there are plenty of shooters out there with ragdoll; choose one.

[baddo 0]

Ragdoll has been, at least on my part, discussed here as a possibility to fix obvious problems in the games BIS creates.

It is not necessarily 'ragdoll' what we need. If you can point some other technology to us and to BIS which fixes the obvious problems with static death animations, then we welcome that technology.

It goes like this: there is something in both OFP and ArmA which we dislike, in the form of static death animations which can make things look very unrealistic a lot of the time when playing the game. The ragdoll technology is one potentially good way to solve that. Whether BIS actually uses it is of course beyond our control, and it shouldn't be in our control.

As an example of how obviously unrealistic things can get in ArmA. I was editing a new mission the other day, which has enemy AI guards in watch towers. When I shoot such an enemy all the way to death, the visual result of it is far from good. Every player can easily say that that is totally wrong. The body of the dead unit is usually over half resting over nothing, in horizontal position. Just one example where it would be glaringly obvious to benefit from a technology like ragdoll. With a ragdoll chances would be good that the dead body would not be hanging in air supported by nothing.

It is the feeling of realism in the game which we would like to see improved. It is not that we want BIS to follow other game developers.

[AWDrift 0]

Another thing is it's pretty hard to find a shooter without ragdoll now. If ArmA 2 comes out in 09 or 10 with death anims it's going to look dated in that aspect. It's kind of like shading in that it's not necessary, but having it in the game is part of moving forward with the rest of the gaming world.

And it's pretty hard to find shooters that don't revolve only around scripted narratives and cinematic gameplay. Whether or not such a design concept is effective, ArmA (II) is not one of those.

In today's gaming world, the borders that define genres are dissolving. There are certainly instances where ArmA plays more like a real-time strategy (or real-time tactics) game than a shooter, and when was the last time you saw ragdoll in an RTS? This genre focuses on a larger scope than most shooters, and narratives and cinematics aren't driving characteristics of this type of gameplay.

Saying that standardizing certain technologies is part of moving forward with the gaming world is quite inaccurate. As the gaming world moves forward, more diversity appears in the tech as more advancements are made and ideas are explored. While it's true that a lot of game developers try to follow a standard, it's not so much the gaming world advancing but the business aspect of it staying put.

The gaming world is not moving forward in a linear path either, rather it is branching out in all different directions. There are still audiences for games like ArmA, and many of them are hoping BIS is doing their best to keep it from merging with those other branches simply out of obligation. Ragdoll isn't somethign a game like ArmA is in dire need of, and if the need should arise then let it come (and be resolved) on its own terms and not simply because everyone else is doing it.

As a final note, if someone believes they need ragdoll to satisfy their desires, then it's like you said, there are plenty of shooters out there with ragdoll; choose one.

I was adding on to reasons already listed, not saying they should do it soley because everybody else is. Also if you want Open world or RTS games with ragdoll check out AoE3, Company of Heroes, Soldiers heroes of WW2 1&2, Crackdown, Mercenaries1&2, DoW 2, etc,etc. Also the "if you want this feature then go play another game' excuse is crap. Everbody here has just as much right to ask for new features as you do.

[Bunkers 0]

I'll keep it simple: Ragdoll Rock's!

..if it's affordable.

[Viper23rd 0]

I personally believe a ragdoll physics engine will add to the realism and atmosphere of the game. Especially how they'll fall down to the ground once they've been shot. There's nothing wrong with the current animations however, keeping it simple and stupid, it works so it'll be fine.

If BIS does choose to implement ragdoll physics then it's a question whether it will interfere with current AI physics and if it won't be a large performance hit, which is a complete gameplay killer as well.

[CarlGustaffa 4]

Yeah, I'd rather have performance than ragdoll. Performance is scarce as it is already without it.

As far as I understand:

Problem: Physics uses lots of random appromiation tests, impossible to sync among clients.

Problem: Ragdoll require physics. Maybe not cloth and multibody simulation, but still.

Problem: ArmA makes use of dead bodies for cover, so they must be synched on each client.

Only possible solution I can see for proper multiplayer ragdoll effects with equal effect after done is this: Make every ragdoll effect perform on each client. On one client it moves it this direction due to randomness of physics approximation, on another client it moves in another direction.

Once the ragdoll is finished and all movement has stopped (can be a while if several tanks run over it), THEN it will suddenly snap to servers ragdoll execution.

Would this look any better? Naah, stick with what works.

There was mention of GTA IV has so and so physics. Did you check with pedestrians in multiplayer modes? Gone (or less than regular). Coincidence?

[Viper23rd 0]

Adding ragdoll would be quite a problem indeed, I guess.

Another example would be:

Taking your example of laying behind a dead body for cover, if that body is shot in ArmA it doesn't move at all, with ragdoll physics it would have to move, does that means you can collide with it as well or will it 'pass' through your body as you're laying behind/in it? Adding collision to it will require even more processing and there comes the problem of syncing in multiplayer mode and performance hits.

Yours,

[dmarkwick 261]

As far as I understand:

Problem: Physics uses lots of random appromiation tests, impossible to sync among clients.

Problem: Ragdoll require physics. Maybe not cloth and multibody simulation, but still.

Problem: ArmA makes use of dead bodies for cover, so they must be synched on each client.

All good points

Further thoughts on each:

Really, only the torso needs to be synced. Maybe have just the torso position synced among clients?

Ragdoll does require physics, however ArmA already has physics.

As for bullets hitting bodies & moving them, I'd be happy for bodies to be static once fallen. So the ragdoll would provide the unique pose, the realistic & unique fall, then would switch off once the MP torso sync has reached a satisfactory state. I think each client's machine could cope with the various limb movements, which do not need to be synced. (Not everything needs to be in perfect sync. Smoke for example is done client-side all the time.)

[EDcase 87]

Bodies don't move when hit by small arms so they can become static objects. There is no body collision ARMA anyway.

If caught in a blast a body should disappear.

If hit my large caliber it should break appart but that won't be happening in the game unless scripted later.

[CarlGustaffa 4]

Quote[/b] ]Ragdoll does require physics, however ArmA already has physics.

Yes, but not fully syncronized in all aspects. Throw a smoke grenade and observe that it doesn't land in the excact same place on all clients.