http://www.nvidia.com/object/fermi_architecture.html
And the name of this revolutionary render MOOD is an ITALIAN scientist.....
legend of the CUDA.......
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
-
cristian.tumiati
- Posts: 45
- Joined: Wed Jan 13, 2010 10:20 am
- Contact:
This is the FUTURE!!!!!!!!!!!!................the show must go on!!!!!!!!!!
http://www.nvidia.com/object/fermi_architecture.html
And the name of this revolutionary render MOOD is an ITALIAN scientist.....
http://www.nvidia.com/object/fermi_architecture.html
And the name of this revolutionary render MOOD is an ITALIAN scientist.....
win7 64 | Geforce GTX 560 + 2 x GEforce GTX 670 4GB | i7 3930K @ 3.20 GHz 3.20 GHz | 32GB ram
So what's going to happen with openCL!? I've seen a few nay-sayers with regards to CUDA, but surely openCL will have a way to go to match CUDA?
Win 7 Pro (x64) | Geforce 295GTX | Quad 2.4GHz | 4GB PC-6400
-
cristian.tumiati
- Posts: 45
- Joined: Wed Jan 13, 2010 10:20 am
- Contact:
Open cl is the near step ...i think....now is possible program in c into the graphic card...but i'm the graphc designer....not a tech 
win7 64 | Geforce GTX 560 + 2 x GEforce GTX 670 4GB | i7 3930K @ 3.20 GHz 3.20 GHz | 32GB ram
Its gonna be like Direct X versus Open GL
Open gl very good but Direct X is used by Microsoft
Open gl very good but Direct X is used by Microsoft
http://Kuto.ch - Samuel Zeller - Freelance 3D Generalist and Graphic designer from Switzerland
"The sleeper must awaken"
Yes but even with current GTX200 line Nvidia is the winner because of CUDA
I don't care if I can't get Fermi ill put four or more (ive seen motherboard with 8 PCIE) GTX260 and that will do the trick.
Or I will buy a rack of GPUs
I don't care if I can't get Fermi ill put four or more (ive seen motherboard with 8 PCIE) GTX260 and that will do the trick.
Or I will buy a rack of GPUs
http://Kuto.ch - Samuel Zeller - Freelance 3D Generalist and Graphic designer from Switzerland
-
cristian.tumiati
- Posts: 45
- Joined: Wed Jan 13, 2010 10:20 am
- Contact:
Hey....or this...if the fermio will be a good BOY!!!!!
but....this is the EXTREME-hardware solutions....i think that whit 3 or 4 GTX 285 when the pricewill be quite low....is the better solutions
but....this is the EXTREME-hardware solutions....i think that whit 3 or 4 GTX 285 when the pricewill be quite low....is the better solutions
win7 64 | Geforce GTX 560 + 2 x GEforce GTX 670 4GB | i7 3930K @ 3.20 GHz 3.20 GHz | 32GB ram
Excuse me being techno NOT! But from seeing that article cristian posted, is it suggesting that fermi will actually allow other apps to utilise the GPU without rewriting?
Is c++ what other stuff written in?
"Man I have to get with the program - I don't know shit!"
Is c++ what other stuff written in?
"Man I have to get with the program - I don't know shit!"
IMHO CUDA is a deadend. It is a propietary technology fixed to one manufacturer and it is not Open even though nVidia PR likes to spin sometimes it is. It feels like the Cg for GPGPU.
Having read that article on SemiAccurate from Charlie (a known critic of nVidia's corporate strategy). I sort of agree with him. Fermi looks like a medium disaster in many aspects it has this GeforceFX (5 series) smell on it all over the place. People try to bullshit him but he has been right on nVidia in many occations.
Basically nVidia hasn't been doing anything decent since the G80, the G92 was a respin, the G200 was basically another respin of the same thing. And then the terrible rebranding of old parts strategy all over the shop.
The green team is really screwing customers since the last 2-3 years.
Having read that article on SemiAccurate from Charlie (a known critic of nVidia's corporate strategy). I sort of agree with him. Fermi looks like a medium disaster in many aspects it has this GeforceFX (5 series) smell on it all over the place. People try to bullshit him but he has been right on nVidia in many occations.
Basically nVidia hasn't been doing anything decent since the G80, the G92 was a respin, the G200 was basically another respin of the same thing. And then the terrible rebranding of old parts strategy all over the shop.
The green team is really screwing customers since the last 2-3 years.
Hi guys,
Just calm down,
be happy things like CUDA exist as thanks to it you can now render on your GPU.
sure opencl might be better as it's open and more compatible,
but we never said we won't be supporting OpenCL.
we're happy to use CUDA at this time as it works,
and we'll be happy to use OpenCL in the near future when it's mature enough.
If you don't want to buy an Nvidia GPU, that's your choice,
but you'll have to be a bit patient then and await OpenCL support in octane to be able to use it.
Radiance
Just calm down,
be happy things like CUDA exist as thanks to it you can now render on your GPU.
sure opencl might be better as it's open and more compatible,
but we never said we won't be supporting OpenCL.
we're happy to use CUDA at this time as it works,
and we'll be happy to use OpenCL in the near future when it's mature enough.
If you don't want to buy an Nvidia GPU, that's your choice,
but you'll have to be a bit patient then and await OpenCL support in octane to be able to use it.
Radiance
Win 7 x64 & ubuntu | 2x GTX480 | Quad 2.66GHz | 8GB