HCC GPU Launched.

[Sparky]

You guys might want to check what clocks your cards are running at. My BOINC just started running these by itself (and I'm still on 6.12) and I noticed that even when running the tasks my GPU is still in 2D clocks at 500MHz. If we could get them in 3D clocks mode....

[OldChap]

I used afterburner to take the clocks up to 900 now: completion times around 3:20

[Sparky]

My 7950 at 2D 500MHz clocks takes 1 min 15 sec to get to 99.4% done, then that last 0.6% takes an additional 30 seconds from the CPU. lol so weird.

In other news...

HCCGPU.PNG

[kjeldoran]

You guys might want to check what clocks your cards are running at. My BOINC just started running these by itself (and I'm still on 6.12) and I noticed that even when running the tasks my GPU is still in 2D clocks at 500MHz. If we could get them in 3D clocks mode....

Hmm, that's odd as it is doing 3D clocks for me without changing anything. GPUz and CCC both report 3D clocks.

Lifted straight from WCG forums http://www.worldcommunitygrid.org/fo...3962_offset,90

<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_hcc1</name>
    <nbytes>3499008.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <executable/>
    <signature_required/>
    <file_signature>
792397d3ec34c91119da45507eb1f8e732452b9d40c1d4adc313cb4d0b6db671
0382de7f034a2dea10b75e2ab2e012b1ce9c3effa738a48f50daa615fa3e6b82
3f4e28d9986968d347552f0298571f3755a601f8834ab9f554d457ea2f0a71c4
843ef8c8b5e4b04633591321627f9a403f12f380bcb5702593f2e42ca0ea5514
.
</file_signature>
</file_info>
<file_info>
    <name>hcckernel.cl.6.56</name>
    <nbytes>63144.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <signature_required/>
    <file_signature>
a2633e96b0df14406470c5c98e49533b903776e06356cf2dfad1ee853eb15c42
190a29639bed90e5a02c426669ff46441f90f1dad3fe303f05314befeffc28e5
7433da8e2545665f9900ec714a6f9f2d83246de5c32c6c060ac017d1392e0ed6
34839df4b734c410bb5007fe6108d793be40665a4c538654240dc28055bc01fe
.
</file_signature>
</file_info>
<file_info>
    <name>hcc1_image01_6.56.tga</name>
    <nbytes>262188.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <signature_required/>
    <file_signature>
7117813faaff1bceb99d172432ff67417397b809172fc09c47a0db414106c68d
de9bf88ed64f518e78b5e7ab79637335c9b166003bd3f32a29915101c1758746
d1db7a90bc201c35b1c65fcfb2a9b091253b1979bc6f9db748feb5a6e9d5c804
cd875399b2ee4016fc210f5b76f15aa0902477ce1c0a945fb56c178a0d65de4c
.
</file_signature>
</file_info>
<file_info>
    <name>hcc1_image01_6.56.tga</name>
    <nbytes>262188.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <signature_required/>
    <file_signature>
7117813faaff1bceb99d172432ff67417397b809172fc09c47a0db414106c68d
de9bf88ed64f518e78b5e7ab79637335c9b166003bd3f32a29915101c1758746
d1db7a90bc201c35b1c65fcfb2a9b091253b1979bc6f9db748feb5a6e9d5c804
cd875399b2ee4016fc210f5b76f15aa0902477ce1c0a945fb56c178a0d65de4c
.
</file_signature>
</file_info>
<file_info>
    <name>hcc1_image02_6.56.tga</name>
    <nbytes>32812.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <signature_required/>
    <file_signature>
0d3314db1591dcf067aa6551e3582f5274ff10287fb384c11934a98245fd1631
be5249db0ae6d0138b191618ae9502e853a681f879e7fa3c94fec8effdd6685b
a325ec7538ec8abb1fd719629c691bf721da031a87cf69d20b0f501a38e16a92
dbdd7442d90b64f30deff58bc46c5096fb017f4d6410eb8e9bfacb09b2c6a8ae
.
</file_signature>
</file_info>
<file_info>
    <name>hcc1_image03_6.56.tga</name>
    <nbytes>131116.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <signature_required/>
    <file_signature>
58915b00c9fae53d18ef72f179c6ff1222c134b0c0e696c3bbfddb91c650b26f
c5856311eea2fe2128d9dbdd558b47963332cf9f9ae910689a133b03e7113715
39babe2ce7303b73e9bfc69b1f01cc4199c7dfab13322d5b40adcb8ee167832d
81df23835ba4e12430a50aeb534b8bfed9a82ce54fb0a4018104f1adb36778d9
.
</file_signature>
</file_info>
<file_info>
    <name>hcc1_image04_6.56.tga</name>
    <nbytes>262188.000000</nbytes>
    <max_nbytes>0.000000</max_nbytes>
    <status>1</status>
    <signature_required/>
    <file_signature>
59882941f8d7b844b3ac20301fba70c98ab9cedd5a41855d24b0721b6b6275e6
f1de708d8b43ae7f0198f21924f528c1c710b6d36776d881fee6ebb539cf2f4e
4c4f63997be3eb6157b9a3a5922bdbdebb26fc1cfeae4d13980682800c697ffa
54a7a64cc14694ec457132320f0c2bcc9c3fed675f71cbe388d9ec55332c1bae
.
</file_signature>
</file_info>
<app_version>
    <app_name>hcc1</app_name>
    <version_num>656</version_num>
    <platform>windows_intelx86</platform>
    <avg_ncpus>1.000000</avg_ncpus>
    <max_ncpus>0.783972</max_ncpus>
    <flops>56530729222.498611</flops>
    <plan_class>nvidia_hcc1</plan_class>
    <api_version>7.1.0</api_version>
    <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>
    <file_ref>
        <file_name>hcc1_image01_6.56.tga</file_name>
        <open_name>boinc_wcg_skin_w-logo.tga</open_name>
    </file_ref>
    <file_ref>
        <file_name>hcc1_image02_6.56.tga</file_name>
        <open_name>pbibmablk.tga</open_name>
    </file_ref>
    <file_ref>
        <file_name>hcc1_image03_6.56.tga</file_name>
        <open_name>UHN_stacked2.tga</open_name>
    </file_ref>
    <file_ref>
        <file_name>hcc1_image04_6.56.tga</file_name>
        <open_name>HCC_LOGO.tga</open_name>
    </file_ref>
    <coproc>
        <type>CUDA</type>
        <count>0.5000000</count>
    </coproc>
    <gpu_ram>266338304.000000</gpu_ram>
</app_version>
</app_info>

Perhaps one of you clever b'tards would like to explain what this is and where to put it ???

I am thinking that this will mean that for me the 1m 20s that the GPU is idle out of each 3m 30 wu would be recovered and worked....It may even take the GPU to 100% utilisation

Double points???? 140% at least.....surely!

I see nvidia info in this, do you think we could edit it to work with ATI?

[my87csx481]

Ok, I think I've figured it out.

Here's how I've found out how to only get HCC GPU WU's and still be able to get CPU units from other projects.

1: Select your other projects.
2: Do not select HCC.
3: Make sure "If my computer can process work on my graphics card, then please send me work to run on my graphics card for the projects that I have selected above." is checked.
4: Make sure "If there is no work available for my computer for the projects I have selected above, please send me work from another project." is checked.
5: Tell your rigs to update.

Doing this seems to work for me.

Edit: Downside: If your other projects run out of cpu work then you may get some stuff you don't want.

Can anyone else confirm this? Doesn't seem to be working for me....yet.

[[Ion]]

Can anyone else confirm this? Doesn't seem to be working for me....yet.

Yes. I'm not getting any GPU WUs ATM (nor any CPU WUs--probably because of project maintenance), but otherwise it's working well. I'm getting only SN2S, DSFL, and GFAM WUs on my CPUs, and then HCCs on my GPUs

[my87csx481]

Does it say in your event log "Requesting tasks for GPU"?

[Otis11]

Hmm, that's odd as it is doing 3D clocks for me without changing anything. GPUz and CCC both report 3D clocks.



I see nvidia info in this, do you think we could edit it to work with ATI?

To my knowledge ATI doesn't have the issue that this fixes... Only NVIDIA cards that have the unique shader ratios will benefit from this.

Unless I missed something?

[werdwerdus]

Can anyone else confirm this? Doesn't seem to be working for me....yet.

It seems to be working for me! Running on GT 240 on 4th sig rig. Runtimes are around 14-16 minutes.

I guess I don't need to run 2 boinc clients per rig to get only HCC gpu work units and other project CPU work units.

[kjeldoran]

To my knowledge ATI doesn't have the issue that this fixes... Only NVIDIA cards that have the unique shader ratios will benefit from this.

Unless I missed something?

Well, based on what I read it allows to run more than one GPU WU. This would be nice since it allows you to run another GPU WU while the other GPU WU is doing the CPU work. Effectively doubling the amount of WUs you complete in a day. Since the code is for nvidia only, I was hoping there would be a way for AMD GPUs.

[[Ion]]

Does it say in your event log "Requesting tasks for GPU"?

Yes. It says "Requesting new tasks for CPU and NVIDIA"
And then it says "Scheduler request failed: HTTP internal server error". The weird thing is that my GTX460 rig is getting a ton of GPU WUs.

[my87csx481]

I gotta' triple check then. Mine just says "Requesting new tasks for CPU"

Thanks for the responses.

Edit: found the problem. says "no usable GPUs found"....WTH man. It's a gtx 670.
Edit again: BOINC is installed as a service. I'll try to reinstall "unprotected". This seems like deja-vu. Didn't I go through this once already?

[[Ion]]

I gotta' triple check then. Mine just says "Requesting new tasks for CPU"

Thanks for the responses.

Edit: found the problem. says "no usable GPUs found"....WTH man. It's a gtx 670.

Do you have BOINC installed as a service? Service installs don't support GPU crunching

[my87csx481]

Yup, see above 2nd edit.

It's at least trying to get GPU work now. Thanks man.

[0ne.shot]

A bit rough but from comments seen here so far:

card	|	mins	|	secs	|	wu/hr	|	wu/day	|	BOINC	|	WCG

HD5870	|	3	|	30	|	17.1	|	411	|	13659	|	95616

hd7870	|	2	|	9	|	27.9	|	670	|	22236	|	155654

HD7970	|	1	|	30	|	40.0	|	960	|	31872	|	223104

HD7770	|	2	|	14	|	26.9	|	645	|	21407	|	149846

HD7850	|	2	|	6	|	28.6	|	686	|	22766	|	159360

Something seems a bit off with some of those numbers. The 7870 has effectively double the SPs as the 7770 and 25% more than the 7850 and half of the 7970, yet the numbers don't appear to be reflect this, or there are variables we are unaware of.

7770s are also dropping in price, so waiting a week may save a few bucks. They should be settings at ~$120 soon.

[werdwerdus]

It seems to be working for me! Running on GT 240 on 4th sig rig. Runtimes are around 14-16 minutes.

I guess I don't need to run 2 boinc clients per rig to get only HCC gpu work units and other project CPU work units.

something isn't working.. in my 2 gpu rigs I only have 1 gpu running on GPUgrid and the other is sitting idle... boinc isn't requesting more gpu tasks for WCG Might have to revert back to dual clients. Wow this is getting messy, can't wait until I can get a couple dedicated AMD gpu rigs

[fallwind]

I can't wait to see some more data points on various GPU's. My farm has zero GPU's so there's alot of empty PCI-E slots waiting to be filled!

[yojimbo197]

Something seems a bit off with some of those numbers. The 7870 has effectively double the SPs as the 7770 and 25% more than the 7850 and half of the 7970, yet the numbers don't appear to be reflect this, or there are variables we are unaware of.

I have a 1Gb GDDR5 XFX HD6670 card running HCC GPU WU's. The time per WU seems to vary from batch to batch of jobs. Some batches I"ll average in the 5-7 minute range. Other batches seem to take somewhat longer 11-14 minutes. And I"ve seen some jobs take as long as 18-20 minutes.

What complicates all this is that I've had at least three HCC GPU jobs that hung up at 94% for as long as 50 minutes to 5 hours. Two I got to re-run by suspending and then going back to them later. The third is sitting in my queue right now.......

[slaveondope]

18 soon to be idling 5870 cores watching

[Conquistador SW]

Very nice OldChap now we need some NV figures too

I calculated the average CPU time for all the WU's that have been validated by now (46 WU's) for my GTX 260 (192SP) + 3770K at 4.8GHz.



Average CPU time was 0.1208 which would mean a WU on average takes 7 min and 40 sec.
It can do slightly more than 8 WU's per hour which means it would do ~45000 WCG points per day.

Oldchap, how did you calculate from time per WU to WU's per hour? I seem to be getting slightly different numbers and probably forgetting something stupid

[fallwind]

Just for another data point, my GTX 460's are averaging about 6 minutes per workunit and claiming 33 credits each. That calculates out to around 55000 WCG/day. I recall during the beta I was getting 55 credits/WU so something has changed there.

If the Radeon 7770 really can do 150k/day then that's likely the card to go with. I'm seeing them on sale here in Canada for as low as $99!

[kik]

This is the result from my 7970 but it's only 60% loaded for some reason.

CPU Time (hours): 0.02
Claimed/Granted Credit: 33.1

[Conquistador SW]

Just for another data point, my GTX 460's are averaging about 6 minutes per workunit and claiming 33 credits each. That calculates out to around 55000 WCG/day. I recall during the beta I was getting 55 credits/WU so something has changed there.

That seems low compared to my GTX 260, or is the difference in computing power (192 cuda cores vs 335 cuda cores) not that big of a deal?

[Rob_B]

Taken from the WCG forum.

app_info.xml info for ATI/AMD would be nice.

<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__ati_hcc1</name>
 <executable/>
</file_info>
<file_info>
 <name>hcckernel.cl.6.56</name>
</file_info>
<app_version>
 <app_name>hcc1</app_name>
 <version_num>656</version_num>
 <platform>windows_intelx86</platform>
 <avg_ncpus>1.000000</avg_ncpus>
 <max_ncpus>1.000000</max_ncpus>
 <flops>32282641180.000000</flops>
 <plan_class>ati_hcc1</plan_class>
 <api_version>7.1.0</api_version>
 <file_ref>
  <file_name>wcg_hcc1_img_6.56_windows_intelx86__ati_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>
 <coproc>
  <type>ATI</type>
  <count>0.500000</count>
 </coproc>
 <gpu_ram>262144000.000000</gpu_ram>
</app_version>
</app_info>

Use at your own risk. app_info only usable for HCC GPU only.

It works for my 5850, but now it only uses the GPU. No work for the CPU. And it went up from 3 minutes per WU to over 5 minutes.

[v0dka]

Oh man I can't wait to try this out. Unfortunately I won't be home till Sunday... AARGH