Yes i can!
I will give you a Feedback soon
and its working fine!
its not so stucky now, but a little bit it is.
the last bug is the problem with the gibs ^^
Has anyone got ACF motors working with this? I hear engine sounds, and some sort of weird movement, but nothing really happens.
On a better note, everything else seems to be working much better. Wheels don't snag on the ground. Man, do I love me some Bullet welds. Such strength. Much less weight stool.
[QUOTE=Skyfire12;45473503]Has anyone got ACF motors working with this? I hear engine sounds, and some sort of weird movement, but nothing really happens.
On a better note, everything else seems to be working much better. Wheels don't snag on the ground. Man, do I love me some Bullet welds. Such strength. Much less weight stool.[/QUOTE]
Nope! How are they supposed to move? SetAngularMotor on axis constraint? ApplyTorque? If the developer wants to pop in, that'd be great.
Also, funny thing about bullet constraints: They're too strong for our use! Some maps have them spawned in a way where they cannot fully resolve the error, and in the case of bullet they start freaking the fuck out and bashing the attached props against the ground.
I'll have to figure out how to make them weaker and expose that in a convar so people who are using VPhysics for building can enjoy their full strength. It'll probably have to default to the weak setting so clueless people can enjoy any map they want without problem.
[QUOTE=Grayman;45492621]Nope! How are they supposed to move? SetAngularMotor on axis constraint? ApplyTorque? If the developer wants to pop in, that'd be great.[/QUOTE]
Looking through the ACF code, [I]Phys.ApplyForceOffset()[/I] is what's being used.
[quote][img]http://puu.sh/aqrK6/565b337c83.png[/img][/quote]
[QUOTE=MrWhite;45495245]Looking through the ACF code, [I]Phys.ApplyForceOffset()[/I] is what's being used.[/QUOTE]
Why do they use ApplyForceOffset to apply torq... Oh. Garry's Mod doesn't expose PhysObj:ApplyTorqueCenter. Welp, guess that's a new feature request!
P.S: [URL="https://github.com/nrlulz/ACF"]Is this the correct link to the source code?[/URL] What entity is it/where is that code you linked?
[QUOTE=Grayman;45495912]Why do they use ApplyForceOffset to apply torq... Oh. Garry's Mod doesn't expose PhysObj:ApplyTorqueCenter. Welp, guess that's a new feature request!
P.S: [URL="https://github.com/nrlulz/ACF"]Is this the correct link to the source code?[/URL] What entity is it/where is that code you linked?[/QUOTE]
The snippet I posted is in /lua/entities/acf_gearbox.lua.
You've got the right source link, yep.
[QUOTE=MrWhite;45496342]The snippet I posted is in /lua/entities/acf_gearbox.lua.
You've got the right source link, yep.[/QUOTE]
Okay, so I've messed with it and... I have no idea what I'm doing. Can you (Skyfire) post a video of normal behavior versus the behavior you're experiencing in VPhysics?
Also give me an easy-to-construct test case so I can look at this (picture with parts labelled maybe?)
[video=youtube;-MtAShtgqJw]http://www.youtube.com/watch?v=-MtAShtgqJw[/video]
[I][If video isn't available, then try again later. I'm uploading the video while I nap.][/I]
Minor update: Looks like I've mostly fixed being able to drag objects through other objects.
[vid]http://puu.sh/auE1G/c68815f6fa.webm[/vid]
However, as you can see, it's not completely gone. I think now the issue is just that bullet doesn't have continuous collision detection, and it'll take a lot more time to implement that.
Remember Gravity Hull Designator? Is that implemented in a non-hacky way with Bullet possible?
When is this gonna be fixed?
[vid]http://puu.sh/auWD8.mp4[/vid]
[QUOTE=Ott;45525014]Remember Gravity Hull Designator? Is that implemented in a non-hacky way with Bullet possible?[/QUOTE]
Well... It'll probably still be hacky cause they're doing tricks with player movement physics, but it is indeed possible to simulate gravity differently for props with a motion controller and disabling the environment gravity for those props.
In fact, you could simulate custom gravity with valve's VPhysics by using a motion controller.
The only problem really is the player movement physics. I can't change that at all because that code lives in the game. So yeah, it'll probably still be just as hacky. If I think of anything I'll post it here.
[QUOTE=WhiteHusky;45526641]When is this gonna be fixed?
[vid]http://puu.sh/auWD8.mp4[/vid][/QUOTE]
Hah. Got a formula for calculating the buoyant force of a single point on an object? IVP did buoyancy by triangle, but our objects are made of point clouds not triangles.
Yeah... I'll have to think of something. May involve switching our collision shape format over to using triangles.
Pardon my misunderstanding, but how is triangle buoyancy any easier than point buoyancy?
[QUOTE=Ott;45527241]Pardon my misunderstanding, but how is triangle buoyancy any easier than point buoyancy?[/QUOTE]
The formula we're using now requires a volume, which triangles have.
Points are rather... uh, pointy, though and don't exactly have a volume.
Are there any downsides to polygon collision meshes?
[QUOTE=Ott;45527705]Are there any downsides to polygon collision meshes?[/QUOTE]
According to ancient bullet documentation they're somehow slower... But more importantly I'm too lazy to write the code to preserve them from IVP meshes (actually it's already written it just doesn't work).
Might fix it in a little bit.
Great work on this! It's nice to see that after more then a year you are still working on it. Will this fix crazy physics on props and ragdolls crashing servers?
[QUOTE=YourStalker;45536693]Great work on this! It's nice to see that after more then a year you are still working on it. Will this fix crazy physics on props and ragdolls crashing servers?[/QUOTE]
Well, it probably makes physics on props crazier but I've definitely already fixed the issue with ragdolls crashing servers.
I need to look into player controller behavior, there's a feedback loop that makes players continuously speed up if they're in contact with a prop.
[QUOTE=Grayman;45537250]Well, it probably makes physics on props crazier but I've definitely already fixed the issue with ragdolls crashing servers.
I need to look into player controller behavior, there's a feedback loop that makes players continuously speed up if they're in contact with a prop.[/QUOTE]
Thanks! The prop crashes have became real issues on my TTT server lately.
I have a small bug to report after trying to use a stargate i just walk right though as if its not there and if you send a prop it sends it
[QUOTE=justinboggs;45558294]I have a small bug to report after trying to use a stargate i just walk right though as if its not there and if you send a prop it sends it[/QUOTE]
That's weird. Are you sure it isn't the complete opposite, where players are teleported but props aren't?
VPhysics doesn't send any trigger entered events to the game, so if stargate uses triggers to detect anything entering them, only players would be detected. Those are on the todo list and will be implemented later though.
Wheels and welds are MUCH better than with the default vphysics, but if two props collide for too long (more than just a bounce) it drops my client to about 1-2 fps because of all the dust particles and banging sounds. Any chance this could be fixed? I have an nVidia 650M card...lag is not common in any game except for this case where there are tons of particles spawned every tick.
SNIP
[QUOTE=briman0094;45711575]Wheels and welds are MUCH better than with the default vphysics, but if two props collide for too long (more than just a bounce) it drops my client to about 1-2 fps because of all the dust particles and banging sounds. Any chance this could be fixed? I have an nVidia 650M card...lag is not common in any game except for this case where there are tons of particles spawned every tick.[/QUOTE]
Although this is a slight issue, a 650m is really nothing to write home about. If this dragged my 7970ghz edition down, id be worried about it. At the moment it seems that performance isnt number one on the priority list as getting it to fully work should come first, then optimizations.
This could probably be solved pretty easily by putting limit to how often the physics engine can tell the game to spawn particles or create sounds.
That collision dust issue is temporary. I don't have the collision event thing tracking the delta collision time between two objects so the game thinks every collision reported is 10 seconds or whatever I say it is apart.
And no, running VPhysics on your client will not get you VAC banned. I've been told that VAC does not look for VPhysics so you're fine. Plus, in the 2 years I've been working on this project I haven't been VAC banned (and yes, I've connected to VAC-enabled servers with my replacement running).
[QUOTE=ricky23;45723188]Although this is a slight issue, a 650m is really nothing to write home about. [/QUOTE]
It is if it doesn't happen with the default physics.
Couldnt the point buoyancy be solved by taking the mass times four (just an example) and then applying that force divided by the number of points in the object, per point thats in contact with water perpendicularly?
[QUOTE=ricky23;45808198]Couldnt the point buoyancy be solved by taking the mass times four (just an example) and then applying that force divided by the number of points in the object, per point thats in contact with water perpendicularly?[/QUOTE]
Wait what? [img]http://puu.sh/b9ULT/559a314713.png[/img]
[QUOTE=ricky23;45808198]Couldnt the point buoyancy be solved by taking the mass times four (just an example) and then applying that force divided by the number of points in the object, per point thats in contact with water perpendicularly?[/QUOTE]
That's a good idea, but I think the volume is needed. Feel free to prove me wrong, preferably with a link to a detailed explanation.
I'm not majoring or minoring in physics (oddly enough) so I wouldn't know.
[QUOTE=Grayman;45815518]That's a good idea, but I think the volume is needed. Feel free to prove me wrong, preferably with a link to a detailed explanation.
I'm not majoring or minoring in physics (oddly enough) so I wouldn't know.[/QUOTE]
I doubt anyone at VALVe was either when vphysics was written, so no issues there!
Sorry, you need to Log In to post a reply to this thread.
