Message boards :
Number crunching :
GPU Wars 2016: GTX 1050 Ti & GTX 1050: October 25th
Message board moderation
Previous · 1 . . . 11 · 12 · 13 · 14 · 15 · 16 · 17 . . . 19 · Next
Author | Message |
---|---|
rob smith Send message Joined: 7 Mar 03 Posts: 22216 Credit: 416,307,556 RAC: 380 |
...or given that it is getting "quite warm" in the Northern Hemisphere he may be looking for somewhere to cool down - I know he tends to shut down some of his crunchers when the tempratures start to rise... Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13742 Credit: 208,696,464 RAC: 304 |
Just to go back in time a little bit, some people have been speculating about how well the Pascal cards will crunch Seti WUs. If some of the LINUX benchmark results are any indication, it has the potential to be rather good. Checkout the 2nd Graph, the one for OpenCL FFT single presion work (SHOC Scalable HeterOgeneous Computing)... Phoronix GTX 1080 benchmarks and the OpenCL MD5 results in the graph below it are pretty good too. Grant Darwin NT |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
New NV driver resolves conflict with other gen GPUs Got all my cards back Z |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
Someone wanted the start up with the 1080 info 6/8/2016 7:43:50 AM | | Data directory: C:\ProgramData\BOINC 6/8/2016 7:43:50 AM | | Running under account Zalster 6/8/2016 7:43:50 AM | | CUDA: NVIDIA GPU 0: GeForce GTX 1080 (driver version 368.39, CUDA version 8.0, compute capability 6.1, 4096MB, 3044MB available, 8876 GFLOPS peak) 6/8/2016 7:43:50 AM | | CUDA: NVIDIA GPU 1 (not used): GeForce GTX TITAN X (driver version 368.39, CUDA version 8.0, compute capability 5.2, 4096MB, 3066MB available, 6611 GFLOPS peak) 6/8/2016 7:43:50 AM | | CUDA: NVIDIA GPU 2 (not used): GeForce GTX TITAN X (driver version 368.39, CUDA version 8.0, compute capability 5.2, 4096MB, 3066MB available, 6611 GFLOPS peak) 6/8/2016 7:43:50 AM | | CUDA: NVIDIA GPU 3 (not used): GeForce GTX TITAN X (driver version 368.39, CUDA version 8.0, compute capability 5.2, 4096MB, 3066MB available, 6611 GFLOPS peak) 6/8/2016 7:43:50 AM | | OpenCL: NVIDIA GPU 0: GeForce GTX 1080 (driver version 368.39, device version OpenCL 1.2 CUDA, 8192MB, 3044MB available, 8876 GFLOPS peak) 6/8/2016 7:43:50 AM | | OpenCL: NVIDIA GPU 1 (ignored by config): GeForce GTX TITAN X (driver version 368.39, device version OpenCL 1.2 CUDA, 12288MB, 3066MB available, 6611 GFLOPS peak) 6/8/2016 7:43:50 AM | | OpenCL: NVIDIA GPU 2 (ignored by config): GeForce GTX TITAN X (driver version 368.39, device version OpenCL 1.2 CUDA, 12288MB, 3066MB available, 6611 GFLOPS peak) 6/8/2016 7:43:50 AM | | OpenCL: NVIDIA GPU 3 (ignored by config): GeForce GTX TITAN X (driver version 368.39, device version OpenCL 1.2 CUDA, 12288MB, 3066MB available, 6611 GFLOPS peak) 6/8/2016 7:43:50 AM | SETI@home | Found app_info.xml; using anonymous platform Note I don't have the cc_config in so I can see how the 1 card does by itself |
Chris Adamek Send message Joined: 15 May 99 Posts: 251 Credit: 434,772,072 RAC: 236 |
#humblebrag =) Someone wanted the start up with the 1080 info |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14653 Credit: 200,643,578 RAC: 874 |
Someone wanted the start up with the 1080 info That was me for Ray Hinchliffe. I think that's what you would have expected - implying the current formula doesn't need correction? Note I don't have the cc_config in so I can see how the 1 card does by itself Good to see the confirmation that the 'lesser' cards are detected, but simply marked 'not used'. I see you're using v7.4.42 on that host, but I think v7.6.22 (or later) would do the same. |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
#humblebrag =) ;) It does run hot 73C had to adjust the fan curve to get the temps down to 66C |
Chris Adamek Send message Joined: 15 May 99 Posts: 251 Credit: 434,772,072 RAC: 236 |
Whole lot cheaper to replace a fan than the die...lol Eagerly awaiting some iniital results from it.=) #humblebrag =) |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
First 3 results are up, but there was some issues with the kernal. Posted in Raistmer's thread 8.12 about it. Finished without any problems |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14653 Credit: 200,643,578 RAC: 874 |
First 3 results are up, but there was some issues with the kernal. I don't think that shows any problems - simply that's it's the first time the card has been run, so the source .cl code file is compiled into more efficient binary versions for re-use next time. |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
Ok, good to know. Richard I posted the co_processor.xml over at Einstein. If you want it here I can post it here as well Note, the new driver seems to have increased the efficency of the card. It now appears to several minutes faster than it was with the previous driver against my Titans. |
Chris Adamek Send message Joined: 15 May 99 Posts: 251 Credit: 434,772,072 RAC: 236 |
How many wu's are you running at a time, 3? |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
How many wu's are you running at a time, 3? Yes, temps at 68C fan curve set to max at 80C currently with RPM 3202 and 46.6% power to fans |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14653 Credit: 200,643,578 RAC: 874 |
Yes, seen it, thanks. I'll pass all that back to Ray - should be all he needs. I assume your card is a standard 8 GB memory? BOINC is only seeing 4 GB, but that's normal because it's using a 32-bit subsystem for detection. |
Zalster Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 |
Yes, it is 8 GB running at 4.51GHz P2 |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
Yes, seen it, thanks. I'll pass all that back to Ray - should be all he needs. Is there a difference between NV & Radeon GPU detection? I noticed my R9 390X has always been detected with 8GB. 08-Jun-2016 11:05:36 [---] Starting BOINC client version 7.4.42 for windows_x86_64 08-Jun-2016 11:11:22 [---] Starting BOINC client version 7.6.22 for windows_x86_64 SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Chris Adamek Send message Joined: 15 May 99 Posts: 251 Credit: 434,772,072 RAC: 236 |
On my mac's it never detects more than 2048MB. Thankfully nothing relies on what Boinc detects memory wise...lol |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14653 Credit: 200,643,578 RAC: 874 |
Is there a difference between NV & Radeon GPU detection? I noticed my R9 390X has always been detected with 8GB. Yes. Detection of each card type depends on the (limited) amount of hardware information which each manufacturer makes available through their interface API - HMODULE cudalib = LoadLibrary("nvcuda.dll"); for NVidia. I think it was Claggy who first noticed that calls to that DLL returned negative values for cards close to 4 GB... Replacing them with 32-bit calls solved the problem. AMD's equivalent is obviously better at 64-bit maths, but less good at English - BOINC has to look up the card names from a list of ID numbers. |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
Is there a difference between NV & Radeon GPU detection? I noticed my R9 390X has always been detected with 8GB. Perhaps OpenCL detection could be used universally. Similar to how it was implemented in BOINC 7.6.23 and later builds for Radeon GPU name detection. Where it reads the OpenCL BoardName value instead of hard coded names for non-CAL GPUs. But it may be a non issue until there are apps that use a few GB of GPU memory. Like there are at other projects with their CPU apps. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
shizaru Send message Joined: 14 Jun 04 Posts: 1130 Credit: 1,967,904 RAC: 0 |
Ok, first bit of "news" is a rumor of a GP102 Titan getting announced sometime "soon". Next up there's this (official teasers by AMD): http://www.anandtech.com/show/10413/amd-teases-future-radeon-rx-470-radeon-rx-460-cards Also here's a mini-roundup of 1070 options (hopefully actually available sooner rather than later): http://techreport.com/news/30265/nvidia-board-partners-release-an-avalanche-of-custom-gtx-1070s Meanwhile consoles are taking 4k and VR seriously which is good news (in a roundabout way) for FLOPS-hunters like us :) http://www.techpowerup.com/223385/microsoft-xbox-scorpio-soc-powered-by-polaris-and-zen http://www.anandtech.com/show/10418/microsoft-teases-project-scorpio-for-2017-8-cores-6-teraflops-backwards-compatible-with-xbox-zen-or-jaguar and PS Neo: http://www.fudzilla.com/news/graphics/40899-playstation-vr-launching-on-october-13th http://www.tomshardware.com/news/sony-psvr-october-13-release,32071.html |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.