GPU benchmarks - discussion
0
-
garrison wrote:
Christian Gruner wrote:
@Garrision
1: Done and posted.
2: I will have investigate that further to make sure I am correct.
3: The forum is not the right place for a detailed explanation, but with the exception of the above, then all steps are either multi-threaded or are run in parallel.
4: See above, we load the next raw file on CPU, while the current raw is being rendered by GPU.
5: Primarily the Cuda cores for Nvidia or the Stream processors for AMD. However, all the rest of the card's specs will also chime in on the final verdict. So the benchmark made by CO is indeed a better value for judging relative performance in CO.
Thank you, Christian!
We will be eagerly waiting for information from you about JPEG export.
And another question for you (hope I have not bothered you yet 😊 ).
Could you explain why CPU & GPU are not loaded to about 100% even during TIFF export? It can be seen on this graph (made on my PC with this export benchmark ):
https://s10.postimg.org/lz5yiaejt/forum4.png
It seems that SSD has performance reserve too, so it cannot be weak point of overall process.
If CO could export raws using, for example, 2 or more parallel pipelines instead of one, would that help to optimize the process and load the hardware to its full abilities? I guess than despite stages of single export pipeline are optimized, there are situations when the hardware is idle and waiting until other stage is done. That could explain why the hardware is not fully loaded (I may be wrong, of course).
I don't know if it matters that while export the CPU is at 100%
To me at the moment export times are Great, no matter if it is Jpeg or Tiff....
There's a Question that comes to my Brain right now, i've read that OpenCL will merge into Vulkan, are there any plans to move to this API?0 -
garrison wrote:
Christian Gruner wrote:
@Garrision
1: Done and posted.
2: I will have investigate that further to make sure I am correct.
3: The forum is not the right place for a detailed explanation, but with the exception of the above, then all steps are either multi-threaded or are run in parallel.
4: See above, we load the next raw file on CPU, while the current raw is being rendered by GPU.
5: Primarily the Cuda cores for Nvidia or the Stream processors for AMD. However, all the rest of the card's specs will also chime in on the final verdict. So the benchmark made by CO is indeed a better value for judging relative performance in CO.
Thank you, Christian!
We will be eagerly waiting for information from you about JPEG export.
And another question for you (hope I have not bothered you yet 😊 ).
Could you explain why CPU & GPU are not loaded to about 100% even during TIFF export? It can be seen on this graph (made on my PC with this export benchmark ):
https://s10.postimg.org/lz5yiaejt/forum4.png
It seems that SSD has performance reserve too, so it cannot be weak point of overall process.
If CO could export raws using, for example, 2 or more parallel pipelines instead of one, would that help to optimize the process and load the hardware to its full abilities? I guess than despite stages of single export pipeline are optimized, there are situations when the hardware is idle and waiting until other stage is done. That could explain why the hardware is not fully loaded (I may be wrong, of course).
This is a groovy observation. When I watch my uploads to my online galleries to Pixieset Galleries, there are dual uploads happening simultaneously. Is this even a possibility with Capture One Pro? Would it require 8+ cores and dual GPUs to truly activate that kind of acceleration?0 -
CraigJohn wrote:
garrison wrote:
Christian Gruner wrote:
@Garrision
1: Done and posted.
2: I will have investigate that further to make sure I am correct.
3: The forum is not the right place for a detailed explanation, but with the exception of the above, then all steps are either multi-threaded or are run in parallel.
4: See above, we load the next raw file on CPU, while the current raw is being rendered by GPU.
5: Primarily the Cuda cores for Nvidia or the Stream processors for AMD. However, all the rest of the card's specs will also chime in on the final verdict. So the benchmark made by CO is indeed a better value for judging relative performance in CO.
Thank you, Christian!
We will be eagerly waiting for information from you about JPEG export.
And another question for you (hope I have not bothered you yet 😊 ).
Could you explain why CPU & GPU are not loaded to about 100% even during TIFF export? It can be seen on this graph (made on my PC with this export benchmark ):
https://s10.postimg.org/lz5yiaejt/forum4.png
It seems that SSD has performance reserve too, so it cannot be weak point of overall process.
If CO could export raws using, for example, 2 or more parallel pipelines instead of one, would that help to optimize the process and load the hardware to its full abilities? I guess than despite stages of single export pipeline are optimized, there are situations when the hardware is idle and waiting until other stage is done. That could explain why the hardware is not fully loaded (I may be wrong, of course).
This is a groovy observation. When I watch my uploads to my online galleries to Pixieset Galleries, there are dual uploads happening simultaneously. Is this even a possibility with Capture One Pro? Would it require 8+ cores and dual GPUs to truly activate that kind of acceleration?
I thought that also some times, splitting a export job in two pieces and send i to a gpu.
Pascal is able to hold 1 Graphic and 32 Compute Queues.
Radeon Cards can hold this even better.
would like to see something like this in a maybe advanced mode of capture one for advanced users to not stress hardware of users that don't have this compute power.0 -
Radeon 7950 Mac version 30% better score as R9 395 Mac version ? strange? 0 -
PC Games Hardware did test the Titan V (3000€/$) GPU in Capture One.
In it's test the V is by far the fastest gpu, 25% faster than Vega 64, but as you know Vega costs far less.
But they measure their benchmark in Mp/s (Megapixels).
As far a i know they use images of Nikon D800.
But nice to see what Performance we can expect in the next GPU generation.0
Post ist für Kommentare geschlossen.
Kommentare
35 Kommentare