Hello guys!
So AMD is creating an open-source free GPU renderar and more alternatives is always great, at the moment it is available in Maya, Max, Rhino And Solidworks, is soon available in Blender and Cinema 4D.
http://www.amd.com/en-us/innovations/software-technologies/radeon-pro-technologies/radeon-prorender [amd.com]
Any chances it might be coming to Houdini/Houdini Indie? Would be sweet to have a free GPU-renderer for single artists who can't afford the likes of RedShift and Octane.
What do you think about ProRenderer? Does it have a place in Houdini?(As third-party alternative, not included)
Best regards
//Morgan
AMD ProRenderer for Houdini/FX and Houdini Indie?
6938 9 2- MorganNilsson
- Member
- 45 posts
- Joined: 2月 2014
- Offline
- Enivob
- Member
- 2625 posts
- Joined: 6月 2008
- Offline
A little late to the game I think. Why would Blender need another GPU render when they have Cycles? Because of the close ties with Adobe and Apple's dumb move of including AMD card in their Pro Trash Can line of computers Maxxon has recently adopted this as the new GPU renderer for C4D.
If they really wanted it to be Open source they would publish the scene definition language so anyone could write an exporter. When I visit the developers page I can't find that. Even Arnold publishes that [support.solidangle.com] and they are not even open source.
I guess another freebie is always welcome. I am not sure how useful it will be against heavy scene or instancing, however. But for simple product shot rendering it might be cool if you are stuck on a mac.
If they really wanted it to be Open source they would publish the scene definition language so anyone could write an exporter. When I visit the developers page I can't find that. Even Arnold publishes that [support.solidangle.com] and they are not even open source.
I guess another freebie is always welcome. I am not sure how useful it will be against heavy scene or instancing, however. But for simple product shot rendering it might be cool if you are stuck on a mac.
Edited by Enivob - 2016年12月24日 09:26:30
Using Houdini Indie 20.0
Windows 11 64GB Ryzen 16 core.
nVidia 3050RTX 8BG RAM.
Windows 11 64GB Ryzen 16 core.
nVidia 3050RTX 8BG RAM.
- A-OC
- Member
- 253 posts
- Joined: 7月 2006
- Offline
- anon_user_37409885
- Member
- 4189 posts
- Joined: 6月 2012
- Offline
@A-OC not sure what you mean; it's openCL, like, so it'll run on all Nvidia as well.
According to Maxon: “ProRender will fully support Nvidia as well as graphics cards on Windows, and as the vendor of choice for Apple hardware, AMD is in the best possible position to support .”http://www.cgchannel.com/2016/10/maxon-picks-amds-radeon-prorender-for-cinema-4d/ [cgchannel.com]
- A-OC
- Member
- 253 posts
- Joined: 7月 2006
- Offline
- anon_user_37409885
- Member
- 4189 posts
- Joined: 6月 2012
- Offline
- Dan Bertilsson
- Member
- 28 posts
- Joined: 7月 2015
- Offline
NVIDIA crippled their 10series to force people into quadro range for heavy computing.
They expressed negative opinion about consumer grade gpus being able to run pro computation for a long time so it was all coming.
Redshift devs discovered this forced downclock at the release of 10series and several drivers and months later its still unresolved so…
Add to this the fact that SLI is far more limited than previous generations and the 1080 having less cuda cores than a 780 makes it all more than a coincidence - There is a pattern of purposeful downgrading.
As CUDA is proprietary NVIDIA hardware only, they can force downclocks all they want, they risk no rival outperforming them on CUDA.
OpenCL however is hardware agnostic so theres healthy competition… which obviously results in manufacturers keeping their stuff competitive to avoid risk of being outperformed.
They expressed negative opinion about consumer grade gpus being able to run pro computation for a long time so it was all coming.
Redshift devs discovered this forced downclock at the release of 10series and several drivers and months later its still unresolved so…
Add to this the fact that SLI is far more limited than previous generations and the 1080 having less cuda cores than a 780 makes it all more than a coincidence - There is a pattern of purposeful downgrading.
As CUDA is proprietary NVIDIA hardware only, they can force downclocks all they want, they risk no rival outperforming them on CUDA.
OpenCL however is hardware agnostic so theres healthy competition… which obviously results in manufacturers keeping their stuff competitive to avoid risk of being outperformed.
- pezetko
- Member
- 392 posts
- Joined: 11月 2008
- Offline
- Dan Bertilsson
- Member
- 28 posts
- Joined: 7月 2015
- Offline
- pezetko
- Member
- 392 posts
- Joined: 11月 2008
- Offline
-
- Quick Links