Real-Time Physics Simulation Forum

Hi,
What's up with Bullet 3 with GPU acceleration? It's been a very long time since it was teased. And yet no release dates...
And one more thing, it appears that you are using OpenCL for GPGPU on version 3 ... then what happens to platforms that does not support it? For example I don't think XBOX One supports OpenCL. Wouldn't it be better if you used compute shaders instead?
Thanks

Nothing is 'up'. The OpenCL work was mainly sponsored by Advanced Micro Devices, but AMD didn't allocate any resources on the project.

Bullet Physics development is controlled by what contributors such as myself are working on during working hours. My employer gets to decide what I work on, even though it has been almost 100% Bullet open source work during working hours:

From 2003-2010, Sony Computer Entertainment employed me, working on Bullet, and their focus was video games, in particular for Playstation.
From 2010-2013, Advanced Micro Devices employed me, and they wanted me to work on OpenCL acceleration for Bullet.
From 2014 onwards, Google employs me, and currently I am working on Robotics simulation for Google, so my contributions to Bullet are on mainly on improving simulation quality.

As far as I am aware, there are no contributors working on Bullet OpenCL at the moment.
Hope this clarifies things,
Erwin

Hi Erwin,
Thanks for the update. I know you are having a hard time keeping this up and I really appreciate what you have done.

A few things:

* Open source projects are mostly kept in non-working times (don't get offended I don't mean you should spend your time on this, I am a programmer myself an I understand what it means. I mean you have done a very great job and let's keep this up! Perhaps it may one day make you more money than any of your employers give you). It's a bit sad that it's affected by the one who employed you, it's like your son ... you can't just abandon it. And finally so many people are dependent on this project and really count on it.

* I think you need to do better in your social and public affairs. I mean make your web site more usable, add more info and encourage people and companies to contribute. Currently when I see http://bulletphysics.org/ I see a set of good news and I think there are a group of shadow good people working on Bullet everyday and version 3 will be added soon. I did not know you were in short of developers to add GPU support, so do many more people. Many people read blog posts and think bullet is hyper active and it's going to compete with Physx and Havok soon. It will ... but you need to do more to make it happen.
So perhaps if you officially state that this project needs more contribution, more people might come to help. Also it's worth looking up for more sponsors.

Thanks

xfactor5 wrote: * Open source projects are mostly kept in non-working times (don't get offended I don't mean you should spend your time on this, I am a programmer myself an I understand what it means. I mean you have done a very great job and let's keep this up! Perhaps it may one day make you more money than any of your employers give you). It's a bit sad that it's affected by the one who employed you, it's like your son ... you can't just abandon it. And finally so many people are dependent on this project and really count on it.

Not too sure about that Blender has a full paid development staff (and interns, volunteers, etc), Android and Chromium have Google's own staff programmers working on it, ROS was made using Stanford University funding and then developed by staff at Willow Garage, basically any big open source project has a person doing it as a job at it's center, even if people add fixes as a hobby. If you really feel that someone devoted to the engine 100% of the time is necessary, why not see about providing funding for it? Blender's main developers are at least partially funded by donations ( and I'm sure that Erwin wouldn't object to the help, though the effort in setting up such a system will certainly not be something anyone actually wants to do, especially with a full-time job and code maintenance) .

xfactor5 wrote: * I think you need to do better in your social and public affairs. I mean make your web site more usable, add more info and encourage people and companies to contribute. Currently when I see http://bulletphysics.org/ I see a set of good news and I think there are a group of shadow good people working on Bullet everyday and version 3 will be added soon. I did not know you were in short of developers to add GPU support, so do many more people. Many people read blog posts and think bullet is hyper active and it's going to compete with Physx and Havok soon. It will ... but you need to do more to make it happen.
So perhaps if you officially state that this project needs more contribution, more people might come to help. Also it's worth looking up for more sponsors.

Website is a bit bare bones, but it's not really important since this isn't a commercial product like PhysX and Havok. As for development speed, it's a bullet-train (pun intended) compared to ODE and Vortex, and the engine itself has seen about as many improvements as PhysX, despite a much smaller group of active contributers. The main issue is that many of the better funded users don't contribute back as much as with GPL licensed projects, so instead you have exactly a dozen secondary contributes (on github) with more than one commit, none with more than 20 despite the large number of bug fixes and performance improvements that could have been added by developers that use the engine.

Not too sure about that Blender has a full paid development staff (and interns, volunteers, etc), Android and Chromium have Google's own staff programmers working on it, ROS was made using Stanford University funding and then developed by staff at Willow Garage, basically any big open source project has a person doing it as a job at it's center, even if people add fixes as a hobby.

Somehow, those are mostly commercial opensource projects. But most other projects specially libraries are taken care of as a hobby. If I was in Erwin's shoes I would start monetizing bullet by some special models (for example licences for larger studious) or looking for sponsors. If so many more people use it, then bigger players might start sponsoring. Don't know...

If you really feel that someone devoted to the engine 100% of the time is necessary, why not see about providing funding for it?

If I get money, I would. I am currently on the verge of starting my start up which is web not rendering. I am going to spend my extra time and money on 3D graphics.

NOW ... The problem is that this project seems left out, I mean there is even no donate or contribute icon in the website.

The main issue is that many of the better funded users don't contribute back as much as with GPL licensed projects, so instead you have exactly a dozen secondary contributes (on github) with more than one commit, none with more than 20 despite the large number of bug fixes and performance improvements that could have been added by developers that use the engine.

That's also true.