That was a 5870 you inverted number person you..:D
Printable View
That was a 5870 you inverted number person you..:D
In case anyone is interested...there is a 5770 in the FS section for $50 + shipping.
I think this is the most this rig and 7970 can do
Attachment 130736
Almost exactly 110,000 Boinc PPD and the highest scoring single GPU rig only beaten by three dual 7970 rigs
Look at those numbers.,.,770K from one machine..I wish DDTUNG was here to see this!:rofl:
40% more than my 2 rigs and 4 5870's.
I think Mine will be on ebay before long to help fund 1 of those 7970's...... in fact I can see me selling a bunch of unwanted stuff to that end.
I hear Alex is funding one for you Dave :D .... you can fit 4 in the beast I'm sure even if they won't pull quite those numbers individually
I think I need to do some tinkering and get my 7950 at least doing 2 units.
You can get in on it now but the way the payments are set up they are all calculated on your work done not just as a total but with how much was done in specific years and a percentage of the total is dedicated to each of those years.
This was to make sure that the guys who were in the project before money was ever mentioned (2009) wouldn't be blown out by guys coming in with datacenters and multi gigabit lines.
I started in May 2006 and stayed with it when most couldn't because of limitations by their ISP..I was lucky, mine never said a word.
I assume this is MJ12?
Ok, So I have two 7770's coming to me this week that were going to replace my single Sapphire 7870 that I was going to return. I also have a single 7770 coming for my Work PC. Change of plans now (I'll get it figured out soon enough!)...
-- Two 7770's will be exchanged for 1 Gigabyte GV-R787OC-2GD 7870 with 1100 core out of the box and placed in my Home PC running 3 GPU WU's.
-- Single Sapphire is going back (won't OC at all & has issues at mfg specs) and replaced with the above mentioned setup.
-- Single 7770 for Work PC stays as planned.
So I will have two 7870's in my Home PC running 3 WU's each for a total of 6 WU's on that system. Should be good for 800,000+ PPD WCG easily :up:
And then 2 WU's running on the 7770 for another 150,000 PPF WCG.
So hopefully by SC Week I'll be doing closed to 1 Million PPD WCG :D
Ah, I see I have a rival here.
I came to the same conclusion that if you are able to run multiple WU's per GPU, the 7870's become more efficient than the 7770 series,
so now I am also expecting to receive 2 7870 next week and already have a 5870.
I find myself in the strange position though that I have too few CPU cores available to feed these few cards :D
Rivalry and competition is good!
Except now I think I may need a new power supply. I only have 2 PCIe connectors on my 550W plus I'm not sure it can handle 2x 7870 with everything else???
You'd be cutting it close. My 2700k @ 4.7GHz and GTX 570 @ 800MHz core is drawing between 300W-350W according to my UPS. I think the Corsair TX750 is around 80% efficient, so around 240W-280W from the PSU. I also have several HDDS, many fans, and a Corsair H80. You would "probably" be fine, but get a watt meter and be sure.
Well, I was drawing 291W at the wall on a Kill-A-Watt running the 7870 with 3 GPU + 5 GFAM. So adding a second one would throw me into the 350W range I think, but my current PS only has 2 PCIe connectors and I don't like that is has 3x 12v Rails @ 19A each either. I think one 12v high amp rail is better.
Also, my last 100 WU's I just blew threw ranged from 1:14 to 2:17 when running 3 GPU WU's only :rofl:
Is your CPU overclocked Bluestang?
My 5870 + i7 930 is pulling 210W from the wall and I expect the 7870 to use less (it should according to reviews), but I will test once I receive them. If they indeed use less, I plan to just run them on a Seasonic 430, othewise I will shuffle things around and dedicate a Seasonic 620 to them.
Oh, and apparently I also got 2 7770's coming........
Yes, OC'ed as per my sig (Home PC).
http://www.antec.com/Believe_it/PSU/index.php
Read this through regarding PSU myths and that may help clear up some confusion of multi rail/single rail PSUs. Keep in mind 75W is from the PCIe slot so that's a bit less from the PCIe connector alone.
Has anyone purchased and used 128 bit HD 7770's? I can get a deal on some refurbed ones. for $99. I'm wondering if there is a big difference between 128 and 256 bit on these cards. Especially since for another $20 there's a brand new 256 bit HD 6870 out there.
I hear everybody saying memory speed does not matter, only the core speed is important. I did not really test this myself, but I know that when I lowered the frequency of the memory on my 5870 and did not see a difference in ppd.
So for running WCG it should not matter if you get the 128 or 256 bit version.
Ok, got that 7770 installed in my Work PC this morning and have it running 2 WU's :up:
CPU is onyl dual core w/o HT, so no. I couldn't see running 3 or more WU's on this card...at least not while trying to use the PC that is.
Anyone knows how the HD6xxx series perform, got access to a 6990 and a 6970?
As far as WCG/HCC GPU's are concerned:
http://www.worldcommunitygrid.org/fo...d_thread,34055
Tom's Hardware's take:
http://www.tomshardware.com/reviews/...ew,3107-7.html
Looks like I finally got the kinks worked out. Running 12 GPU tasks on my GTX480 @ 800mhz and 980x @ 4.2ghz anywhere between 75-90 seconds per WU :D
My GTX 570, which is almost identical to the GTX 480, takes about 120-180 seconds per GPU task with my 2700k @ 4.7 GHz. Even with 2 or 3 tasks running, it still doesn't save me any time. What app_info are you using?Quote:
Originally Posted by =[PULSAR]=
Looks like I finally got the kinks worked out. Running 12 GPU tasks on my GTX480 @ 800mhz and 980x @ 4.2ghz anywhere between 75-90 seconds per WU
Are we talking about total WU time or GPU time per WU?
... and I haven't run a GPU WU in the past 12+ hours at least. It had a backlog of CPU units to do. Of course.
You know, I was thinking it would be possible to make a 1 mil boinc ppd rig. You would need:
5 7990
2 xeon e7-2870
a motherboard with 5 PCI-e slots
2 PSUs since this rig will probably pull 1.5kW
Water cooling setup since we are already spending like $15000+
Upgraded power socket to at least 20A
Mydog was getting 110,000 boinc ppd with 4 threads on each GPU. Using the above setup will allow you to have 40 GPU threads supported with 40 CPU threads. Possibly 1 mil per day since the CPU speed is a lot slower.
Now, who wants to go for it? lol
MM,
Not dreaming, they are taking between 15-18 mins to complete a set of 12. I'll post a screenie Wednesday when I get home.
Oneshot what is your CPU utilization while running 3 tasks. I think the key to maximize output is utilizing your CPU 100% which typically requires you to run same number of tasks as threads.
That means each takes 15-18 mins, I'm running 6 tasks now on my 7970 and I get an average time of 150 sec on each and all meaning if I start all 6 now they'll all finish in 2.5 mins.
Go into you WCG page and find the Valid WU's for that rig under "Result Status", klikk on valid on one of the results and scroll to the bottom of the new window where you'll find these numbers:
Attachment 130776
Those are the total time and CPU time for that WU.
Fixed
6 WU's made 110,000 Boinc PPD
And you couldn't do 40 GPU WU's because you'd run out of CPU cores/threads, I would say max 32 but I'd reccomend 30. Testing 8 WU's now on the 7970 and the can't seem to get out of sync so that the load evens out. The WU's act a bit like the traffic in the rush hour, as soon as one or two leaves the GPU and start on the CPU the others play catch up as the GPU load is getting easier. I also get 100% CPU usage on my 3770K when the WU's finish on the GPU.
Tried pulling up the work file and this is what it gave me, I think I may just be explaining it wrong. My dog looks like your going to pull 100k on that card :p:
Quote:
Result Log
Result Name: X0930070290850200606141622_ 0--
<core_client_version>7.0.28</core_client_version>
<![CDATA[
<stderr_txt>
</stderr_txt>
]]>
You had me worried for a minute. My 7970 is completing 18WUs in that time frame... still wondering what a GTX 680 loaded with multiple tasks would do. :shrug: Nvidia was always the dominant card with F@H GPU WUs. I about shiet my drawers when I read the ATI appeared to be the best for HCC WCG!!! :rofl:
Ati was the best for bitcoin too so I don't see it as a big deal
If you could do either:
Single 7870 running 8 WU's
or
Two 7870 running 4 WU's each
Which would you do???
7870 with 8 threads will do 500k WCG per day as long as the CPU is a fast one ie 3.5GHz+. 2x7870 with 4 threads each will get more points but not nearly double. They'll be sitting idle a good chink of the time waiting on the CPU.
Hmmm, thanks fallwind. Well, maybe I'll keep the "free" 7870 and return the other 7870 and call it a day. Time will tell as I have a week to figure it out, but this is XS, so I'm not sure.
depends on how many cpu cores you have available. remember each GPU task needs a dedicated core!
I could test this. You guys just tell me what to do. I read in this thread to not run multiple gpu threads if your usage is ~95%. On my GTX 680, it is at 95% with one thread, so I never bothered screwing around with multiple threads. It currently runs 1 gpu thread HCC, another gpu thread HCC on a GTX 460, and 6 threads of GFAM on the cpu (3770k at 4.7).
I'm gonna' pull the 460 and put it into something else, probably this evening, depending on how I feel (feel like :banana::banana::banana::banana::banana: today, shouldn't be at work).....so, let me know what to do/try.
Thanks.
Edit: I mis-typed....mine is a 670, NOT a 680.
Well my GTX670 produces roughly half what a 7850 does. I don't imagine a GTX680 would do much better. My GTX670 does gain a bit from 2 threads vs 1. About a 4:3 ratio.
My GTX480 is about 30% faster than my GTX680L from what I can see here running 5 WU's on each in the same rig.
I just stopped networking on that rig and checked the times on the WU's, approx 10 mins(GTX480) vs 14 mins(GTX680L) You can also see it if you go in to your WCG page and look at the result details.
app_info.xml for Nvidia that I use, GPU WU's only on that rig.
Quote:
<app_info>
<app>
<name>hcc1</name>
<user_friendly_name>Help Conquer Cancer</user_friendly_name>
</app>
<file_info>
<name>wcg_hcc1_img_6.56_windows_intelx86__nvidia_h cc1</name>
<executable/>
</file_info>
<file_info>
<name>hcckernel.cl.6.56</name>
<executable/>
</file_info>
<app_version>
<app_name>hcc1</app_name>
<version_num>656</version_num>
<platform>windows_intelx86</platform>
<plan_class>nvidia_hcc1</plan_class>
<avg_ncpus>1.0</avg_ncpus>
<max_ncpus>0.783972</max_ncpus>
<coproc>
<type>CUDA</type>
<count>.5</count>
</coproc>
<file_ref>
<file_name>wcg_hcc1_img_6.56_windows_intelx86__nvi dia_hcc1</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>hcckernel.cl.6.56</file_name>
<open_name>hcckernel.cl</open_name>
</file_ref>
</app_version>
</app_info>
[s]Is the space intentional?
line 26
<file_name>wcg_hcc1_img_6.56_windows_intelx86__n vi dia_hcc1</file_name>[/s]
Nevermind. It added another space when I copy and pasted. Is the space in the actual code?
No spaces.
Do you have this in your cc_config to run on both gpus?:
<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
</options>
</cc_config>
Also, I have to modify the count right? To get 5 per gpu, I need to have .2 as the count, right?
Thanks for the clarification.
I'll play with it when I get home.
I don't know....event log says "not reporting or requesting new tasks"
I'm trying to run 4 on one gpu, and 4 on another gpu.
I read something in Snow Crash's thread about telling BOINC where to get the files or something....but didn't see any further explanation about that.
Here's my app_info:
Code:<app_info>
<app>
<name>hcc1</name>
<user_friendly_name>XtremeSystems Crunches Cancer</user_friendly_name>
</app>
<app_version>
<app_name>hcc1</app_name>
<version_num>656</version_num>
<platform>windows_intelx86</platform>
<plan_class>nvidia_hcc1</plan_class>
<avg_ncpus>1.0</avg_ncpus>
<max_ncpus>1.0</max_ncpus>
<coproc>
<type>CUDA</type>
<count>.25</count>
</coproc>
<file_ref>
<file_name>wcg_hcc1_img_6.56_windows_intelx86__nvidia_hcc1</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>hcckernel.cl.6.56</file_name>
<open_name>hcckernel.cl</open_name>
</file_ref>
</app_version>
<file_info>
<name>wcg_hcc1_img_6.56_windows_intelx86__nvidia_hcc1</name>
<executable/>
</file_info>
<file_info>
<name>hcckernel.cl.6.56</name>
<executable/>
</file_info>
</app_info>
BOINC defaults to not doing GPU work while computer is in use. Manually set activity to use GPU always and it will start downloading WU's.
Did that already....I don't think that got wiped out when I added the app_info.xml....but I double checked, and it's ticked now.
Edit: Maybe this is the problem?
Attachment 130818
In order to get the application downloaded, I think you need to download at least one WU without the app_info.
I got the same thin when I copied from here but when I copied it from the WCG forum it worked.
try this one
Attachment 130819
Just rename it
OK now I'm confused, I haven't changed a thing but now my log file says "tasks are available for ATI GPU but your preferences are set to not accept them." WTH!
*edit* found a check box on the WCG page to allow GPU units, and it was unchecked. I didn't do that. Set it back to checked. Wacky. Now to figure out how to load down my card better.
*edit2* My card temp barely wavers a couple degrees, but even at 84% load it reports the clocks are still at 2D 500MHz?
*edit3* dang, I'm still on 12.4 drivers :p: That might be part of it. Downloading the latest drivers now.
This did it....sort of. I stopped everything, removed the app_info, restarted everything, BOINC took a minute but d/l'd some new work units (GPU only now), cards only worked on 1 each, I suspended and stopped everything again, replaced the app_info, restarted everything again....all is working now.
Thanks all for your help.
OK I just dropped in a 7850 but after installing latest driver from AMD for WinXP 32 (12.6?) GPUz says I don't have OpenCL. What do I need to do now.
Quick google search
Quote:
try older drivers. IIRC AMD dropped WinXP support in OpenCL support with AMD APP 2.7. catalyst 11.2-11.4 should be last one which have proper opencl support at WinXP
There is a solution in this thread....I think....I'm looking for it.
See post #393 and #395 in this thread.
Thanks, I'm brain burnt today forgot those posts. Guess I'll switch my Nvidia cards to the XP machine and put the 7850 on Win7
I didn't realize this was necessary. I have "report results immediately" in mine. Sooo now cc_config should look like this? :shrug:
<cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
<report_results_immediately>1</report_results_immediately>
</options>
</cc_config>
Finally after almost a full day of trouble free crunching 4 WUs on my 7870...
10/24/2012 ---- 0:004:10:31:41 ---- 493,673 ---- 2,166 :up:
Can't wait to see what a true full day gets me.
What was your final solution?
Uninstalled all AMD drivers/software, rebooted, ran driversweeper till nothing was there, rebooted, and installed 12.11 beta.
While i was at it I increased power limit to 50%, fan to 40%, and GPU Core to 1250. :up:
Of course you probably just jinxed me and it will stall again shortly after I write this. :p:
Regarding the WUs that get "stuck," I think I may have fixed it on my system. I changed my 2d clocks to match my 3d clocks on my 7850 and I haven't had a stuck WU in about 36 hours. Your power consumption will go up in a 24 hour period, but it shouldn't be too much more.
By modifying the profiles XML file?
Something of a step change that I guess a whole bunch of folks are seeing.... Mine are in millions but SAM's....... will be in 10mill steps soon
My 7950 rarely goes to 0%. Jumps between 85%-99% most of the time.
I see they changed results time to "CPU Time / Elapsed Time (hours)" on the WGC Results Status page.
I just replaced an i3-3220 with an i7-3770k (6th sig rig). I also removed two GT 240 gpus that were crunching HCC GPU. Will be interesting to see how the 7 threads of WCG on the cpu compare to the points output of the 2 GT 240 cpus!