Waiting for 980?
Forum rules
Please add your OS and Hardware Configuration in your signature, it makes it easier for us to help you analyze problems. Example: Win 7 64 | Geforce GTX680 | i7 3770 | 16GB
Please add your OS and Hardware Configuration in your signature, it makes it easier for us to help you analyze problems. Example: Win 7 64 | Geforce GTX680 | i7 3770 | 16GB
Hope 980 works fine with Octane:) Please post some results as soon as you get the card.
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
hi,
I was also very excited to see the arrival of the GTX 980, and very excited by reading tests.
But I quickly became disillusioned when I read this one:
geforce gtx-http://www.tomshardware.com/reviews/nvidia-980-970--maxwell,3941.html
look on page 12. Their conclusions regarding consumption when used GPGPU is very worrying.
I was also very excited to see the arrival of the GTX 980, and very excited by reading tests.
But I quickly became disillusioned when I read this one:
geforce gtx-http://www.tomshardware.com/reviews/nvidia-980-970--maxwell,3941.html
look on page 12. Their conclusions regarding consumption when used GPGPU is very worrying.
Mac Pro 2010 6 cores 3,33 - Mac OS 10.11.6 - 16Go - GTX 780 3Go and GT 640 4Go - Archicad 19 - Cinema 4D R17 Visualise - Octane Render V3
namely?ashurao wrote:Their conclusions regarding consumption when used GPGPU is very worrying.
IMO without real octane test we cannot say too much
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
I quote:
"The values above have potential consequences for the everyday operation of these graphics cards, as they represent what can be expected when running performance-hungry compute-oriented applications optimized for CUDA and OpenCL. The reference card’s two 6-pin PCIe power connectors start looking a bit out of place in this context, as they might just not be enough."
Of course, they probably did not test Octane render. But I guess all GPGPU applications consume a lot of energy.
"The values above have potential consequences for the everyday operation of these graphics cards, as they represent what can be expected when running performance-hungry compute-oriented applications optimized for CUDA and OpenCL. The reference card’s two 6-pin PCIe power connectors start looking a bit out of place in this context, as they might just not be enough."
Of course, they probably did not test Octane render. But I guess all GPGPU applications consume a lot of energy.
Mac Pro 2010 6 cores 3,33 - Mac OS 10.11.6 - 16Go - GTX 780 3Go and GT 640 4Go - Archicad 19 - Cinema 4D R17 Visualise - Octane Render V3
you are right - 285W for 980 seems to be odd in the context of what is being advertised
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
- Jaberwocky
- Posts: 976
- Joined: Tue Sep 07, 2010 3:03 pm
Best guess on the 980 octane performance in the benchmark scene is this , based on the new maxwell drivers issued today V344.11
750ti = 640 cuda cores
980 = 2048 cuda cores
eg 3.2 times the performance of the 750ti
first timings are the 750ti , the second are the theoretical 980 timings
D/L = 6.70 m/sec x 3.2 = 21.44 M/sec
PMC = 1.81 m/sec x 3.2 = 5.79 M/sec
Pathtracing = 2.17 M/sec x 3.2 = 6.94 M/sec
I know the 980 has a 256 bit bus and the 750ti only has a 128 bit bus , but once the scene is loaded into the GPU's memory , the bandwidth should not make to much difference.
Anyway that's my best guess until someone actually gets one to try out.
750ti = 640 cuda cores
980 = 2048 cuda cores
eg 3.2 times the performance of the 750ti
first timings are the 750ti , the second are the theoretical 980 timings
D/L = 6.70 m/sec x 3.2 = 21.44 M/sec
PMC = 1.81 m/sec x 3.2 = 5.79 M/sec
Pathtracing = 2.17 M/sec x 3.2 = 6.94 M/sec
I know the 980 has a 256 bit bus and the 750ti only has a 128 bit bus , but once the scene is loaded into the GPU's memory , the bandwidth should not make to much difference.
Anyway that's my best guess until someone actually gets one to try out.
CPU:-AMD 1055T 6 core, Motherboard:-Gigabyte 990FXA-UD3 AM3+, Gigabyte GTX 460-1GB, RAM:-8GB Kingston hyper X Genesis DDR3 1600Mhz D/Ch, Hard Disk:-500GB samsung F3 , OS:-Win7 64bit
980 is 2x slower than 780 ! Look here
http://render.otoy.com/forum/viewtopic.php?f=9&t=42475
Where did you order you 780 6gb?
http://render.otoy.com/forum/viewtopic.php?f=9&t=42475
Where did you order you 780 6gb?
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
That is most likely because it does not have the right drivers. Maybe try and look at this:
http://www.extremetech.com/computing/19 ... 0-review/3
And you will see compute is anything but bad with the new cards. So there is some hope for you.
Also the 256 bit bus isnt what it used to be. Effectivly it is more like 300 something, thanks to some optimisation to the compression algorithm and so on.
It is not so easy to see just by comparing some stats. You need to take in all the little changes into account and the picture will get clearer.
Overall it seems like a very good card to me. Maybe not ideal, but very good nontheless.
Either way im definitely gonna look forward to when it works properly in Octane.
http://www.extremetech.com/computing/19 ... 0-review/3
And you will see compute is anything but bad with the new cards. So there is some hope for you.
Also the 256 bit bus isnt what it used to be. Effectivly it is more like 300 something, thanks to some optimisation to the compression algorithm and so on.
It is not so easy to see just by comparing some stats. You need to take in all the little changes into account and the picture will get clearer.
Overall it seems like a very good card to me. Maybe not ideal, but very good nontheless.
Either way im definitely gonna look forward to when it works properly in Octane.