Discuss stuff not about Indigo.
-
Zom-B

- Posts: 4701
- Joined: Tue Jul 04, 2006 4:18 pm
- Location: ´'`\_(ò_Ó)_/´'`
-
Contact:
Post
by Zom-B » Mon Jan 10, 2011 2:14 am
Nvidia has some hard time doing OpenCL, with CUDA they are way faster.... smells fishy for me!
You can read a lot of technical indepth information in the SLG Forum

polygonmanufaktur.de
-
CoolColJ
- Posts: 1738
- Joined: Mon Jun 25, 2007 1:47 pm
Post
by CoolColJ » Mon Jan 10, 2011 8:11 pm
I think I'll go for a GTX 570.
Last time I had an Nvidia, currently using a borrowed ATI 4670, and I'm not liking the ATI drivers much.
A taste of CUDA is nice too
ATI cards have more ram though, up to 2GB
Last edited by
CoolColJ on Wed Jan 19, 2011 10:01 pm, edited 1 time in total.
-
lycium
- Posts: 1216
- Joined: Wed Sep 12, 2007 7:46 am
- Location: Leipzig, Germany
-
Contact:
Post
by lycium » Tue Jan 11, 2011 12:57 pm
SLG was developed on AMD hardware, so I don't find it too surprising

Their architecture offers extremely high peak FLOPs throughput compared to nVidia, if you can make full utilisation of it (i.e. optimise to use of SIMD).
It's great to see ray tracing benchmarks in such GPU reviews, that's the main reason I'm buying them these days

-
lycium
- Posts: 1216
- Joined: Wed Sep 12, 2007 7:46 am
- Location: Leipzig, Germany
-
Contact:
Post
by lycium » Wed Jan 19, 2011 6:32 pm
I bought the broken silicon (GTX 470)

-
neo0.
- Posts: 1784
- Joined: Thu Feb 28, 2008 7:11 am
- Location: the US of A
Post
by neo0. » Thu Jan 20, 2011 5:33 pm
LOL I have the 250. It's so powerful, it's not even on the chart. How do you guys get all the best cards? You must be rich!
-
CoolColJ
- Posts: 1738
- Joined: Mon Jun 25, 2007 1:47 pm
Post
by CoolColJ » Thu Jan 20, 2011 6:58 pm
GTX570 is not that expensive

-
PureSpider
- Posts: 1459
- Joined: Tue Apr 08, 2008 9:37 am
- Location: Karlsruhe, BW, Germany
-
Contact:
Post
by PureSpider » Fri Jan 21, 2011 6:24 am
My 1GB GTX460 is cool for me... and still on the chart!

-
galinette

- Posts: 923
- Joined: Sat Jan 09, 2010 1:39 am
- Location: Nantes, France
-
Contact:
Post
by galinette » Fri Jan 21, 2011 6:47 am
Where is the Tesla C2050?
Etienne
-
PureSpider
- Posts: 1459
- Joined: Tue Apr 08, 2008 9:37 am
- Location: Karlsruhe, BW, Germany
-
Contact:
Post
by PureSpider » Fri Jan 21, 2011 7:03 am
Probably on top of 3 times the height of that diagram

-
lycium
- Posts: 1216
- Joined: Wed Sep 12, 2007 7:46 am
- Location: Leipzig, Germany
-
Contact:
Post
by lycium » Fri Jan 21, 2011 11:58 am
Etienne: as far as single precision performance is concerned, it hangs with the GTX 470/480.
-
CoolColJ
- Posts: 1738
- Joined: Mon Jun 25, 2007 1:47 pm
Post
by CoolColJ » Sat Feb 05, 2011 11:08 am
I write and mix music, my main hobby, and I'm drooling at the thought of Cuda accelerated audio plugins and software now! Much cheaper than the curently available DSP cards and probably much faster.
ie
http://www.liquidsonics.com/software_reverberate_le.htm
once they work out the kinks, since it's so time dependant
GPU Edition: Status Notes
Zero latency mode is not supported by the GPU Edition due to a combination of factors including the mechanisms involved with transferring data to and from the GPU being much more efficient with larger blocks, a current requirement to run CUDA VST plug-ins in a separate thread (making larger block processing more efficient) and the lack of coherency in buffering schemes used by various different VST hosts complicating the above issues. It may become more practical to implement this feature in future using newer GPU architectures.
-
dcm
- Posts: 663
- Joined: Sun Jan 03, 2010 1:55 am
Post
by dcm » Sat Feb 05, 2011 1:29 pm
CoolColJ : im music producer too, where i can listen your music ?
-
StompinTom

- Posts: 1828
- Joined: Mon Sep 04, 2006 3:33 pm
Post
by StompinTom » Sat Feb 05, 2011 4:04 pm
CoolColJ wrote:I write and mix music, my main hobby, and I'm drooling at the thought of Cuda accelerated audio plugins and software now! Much cheaper than the curently available DSP cards and probably much faster.
ie
http://www.liquidsonics.com/software_reverberate_le.htm
once they work out the kinks, since it's so time dependant
GPU Edition: Status Notes
Zero latency mode is not supported by the GPU Edition due to a combination of factors including the mechanisms involved with transferring data to and from the GPU being much more efficient with larger blocks, a current requirement to run CUDA VST plug-ins in a separate thread (making larger block processing more efficient) and the lack of coherency in buffering schemes used by various different VST hosts complicating the above issues. It may become more practical to implement this feature in future using newer GPU architectures.
Waiting for the same thing here! Screw the mega-expensive proprietary DSP cards if we can use our GPUs for it.
Who is online
Users browsing this forum: Google [Bot] and 8 guests