v7 Folding Client...

[PVTCaboose1337]

Will this improve performance for my GTX460?

 

[hat]

Looks like the 5870 itself was bad. Whole thread here:

http://www.techpowerup.com/forums/showthread.php?t=143797

 

[TheLaughingMan]

Folding on the GPU and CPU at the same time use this, lowers the GPU output.

With CCC open, actual use of the GPU drops from 99% to 81% just because the CPU is folding. That is just ludicrous.

 

[thebluebumblebee]

Sounds like you're using GPU2 (even with V7). You need to force V7 to use core 16.

For everyone using ATI/AMD, please look at this post: http://www.techpowerup.com/forums/showpost.php?p=2248985&postcount=1508

 

[TheLaughingMan]

Sounds like you're using GPU2 (even with V7). You need to force V7 to use core 16.

For everyone using ATI/AMD, please look at this post: http://www.techpowerup.com/forums/showpost.php?p=2248985&postcount=1508

That didn't help. It is better, but still not giving me its best with the CPU running. I am just going to stop folding with the CPU.

 

[thebluebumblebee]

Just for my info, you are seeing a load on the CPU with the new core 16?

Edit: Seems core 16 nails the CPU and RAM:

With the release of the v7 client necessary to support this core a few days ago, it was only a matter of time: the Pande Group are rolling out Core 16 (OpenCL OpenMM) to open beta.

Firstly a warning: the beta tag is there for a reason, the core is not finished and may crash more often than usual or otherwise degrade system performance.

To receive these units you must have a v7 client set up in GPU mode with advmethods, a GPU and drivers with support for OpenCL (all 11.x series drivers contain the APP package necessary for this). Under these conditions, it is possible that you may receive the test project 11293, though the number of units available is limited for the first phase of the test and so many clients will be assigned Core 11 units instead.

This project simulates a complete amyloid beta peptide at high temperature. These simulations are very similar to those already performed using the CUDA OpenMM Core 15, and serve as scientific validation for Core 16. The project has the following characteristics (with the reference card being a GTX460 as with Core 15 projects):

627 atoms (a relatively small protein)
Server 171.67.108.44
Preferred Deadline: 6 days
Final Deadline: 10 days
1835 points

We tested the core and this project with a HD5770 (800 shader processors overclocked at 960MHz), powered by an Intel Core i7 920 overclocked to 3.5GHz running Windows XP with the 11.3 drivers package. With this setup we get 6650PPD with 100% GPU usage. This gives efficiency of about 70 PPD/W, which is almost twice what we achieved running this test with a Core 11 project. This is also comparable to the efficiency of the recently released nVidia GeForce GTX590.

On our test machine, we did not notice any degrading of performance, even when playing videos.

Unfortunately not all is well with this core. Firstly there is the high CPU and RAM usage in our test machine: Core 16 uses an entire thread at 100% and requires 375MB of system memory to operate. Hopefully these figures will improve as the core is optimised better, much as happened with the Core 11 variants in the past. The second point is more problematic: some users have reported crashes when folding in parallel with certain apps (notably Flash). These problems do however seem to be limited to cards based on the Cayman chip (HD6900). The developers have been able to reproduce the problem and are working on a fix.

Please let us know how the core is working for you...

Source: Official Forum

BTW, and just me being picky, they say that they got 70ppd/watt. If you take 6650PPD and divide it by 70 to get the watts, the answer is 95 watts. They are trying to say that the card only uses 95 watts in a day, when this is most likely how much it uses in an hour! Therefore, 6650/(95*24)=2.92 points per watt.

That is good news, however, as it looks like the points per watt margin between the ATI and Nvidia cards is gone.

TheLaughingMan, you might try SMP with one less thread.

 

[TheLaughingMan]

Just for my info, you are seeing a load on the CPU with the new core 16?

Edit: Seems core 16 nails the CPU amd RAM:



BTW, and just me being picky, they say that they got 70ppd/watt. If you take 6650PPD and divide it by 70 to get the watts, the answer is 95 watts. They are trying to say that the card only uses 95 watts in a day, when this is most likely how much it uses in an hour! Therefore, 6650/(95*24)=2.92 points per watt.

That is good news, however, as it looks like the points per watt margin between the ATI and Nvidia cards is gone.

TheLaughingMan, you might try SMP with one less thread.

No need. As stated above, I was working on older units that did not offer much for points. Once I switched it to core 16 I get better points per unit, which in turn give me better PPD.

Running ATI HD 5850 @ 850 Mhz core /1200 Mhz memory (4800 effective)

Before v7

511 points per unit
1.47 minutes per % complete
2.972 hours per unit
8.075 units per 24 hours = 4126.21 points/day

After v7 and core 16 (per the software)

3 minutes 7 seconds per unit (with CPU folding)
1835 PPU
8663.61 PPD

The CPU now only drags the GPU down to 90% every now and then. No where near as bad a performance lose as the constant -20% with the old core.

 

[TheoneandonlyMrK]

i have only got 3 tabs in v7 and none of them are slots ive a 5870 and gt240 plus quad core i want to fold on and at min its only using the 5870?? i just have connection identity and proxy tabs??

 

[TheLaughingMan]

i have only got 3 tabs in v7 and none of them are slots ive a 5870 and gt240 plus quad core i want to fold on and at min its only using the 5870?? i just have connection identity and proxy tabs??

use the drop down in the upper right corner to change to advanced or Expert mode. Then double click on the host in the Client box. Once it is open you will see the new tabs.

Add your CPU and other GPU on the slots page.

 

[TheoneandonlyMrK]

cant see it m8, cheers for reply pic posted, is it me??

err yeh it was me i just expanded its window n heypresto sorry lol its l8 ere tho..

View attachment 41660

 

[hat]

A bit of an update on the 5870 situation:

I packed it up and sent it off to my friend a few days ago. He sent it off to HIS's RMA dept. For now it's looking good but I suppose they can still turn me down at any time for any reason... but right now it's looking good for me. It's on its way to their RMA dept now via USPS Priority...

 

[Sharada]

Card ATI HD5770 @850MHz core.
I got about 6000 ppd with core_16, and 2500 ppd with core_11.
CPU load is the same in both configs ~50% constantly. 400-500Mb memory used.
Viewer is working good in v7.

 

[hat]

Got my 5870 running on it... it reports a little over 8000PPD. That's more than my GTX260 is doing, since it's not overclocked anymore.

 

[thebluebumblebee]

To revive this thread and announce a new version, 7.1.38, look here.

 

[hat]

Still running my 5870 at stock speeds. Is this a glitch of some sort?

It seems to be dropping steadily... it used to show almost 35k, now it shows 28k

 

[theonedub]

Im guessing it will level out. I had a WU my 580 said it was getting ~38k on and it quickly sorted itself out.

 

[hat]

Down to 18k now.

 

[TheoneandonlyMrK]

i get 9 Kppd on a 5870 ocd to 1gig , what client settings have you put in?
i get about the same on my 5850 , i think its just over estimateing after a change in WU type ,other wise hook a bruva up man what settings you running

this thread just got much better

Quote:
Originally Posted by thebluebumblebee
To revive this thread and announce a new version, 7.1.38, look here.

could a mod please update the OP with this link as it seems to auto update to the latest beta v7
it would be helpfull

 

[InnocentCriminal]

Apologies on not updating the thread. Unfortunately I can't participate in Folding as it currently stands so I rarely check for updates. I've amended the first post to link to the new client download page.

this threads a bit poor on V7 info for newbies looking to start isnt it the Op is poor in updatedness

I'm a noob when it comes to folding, thought I explained that in the first post. The thread wasn't aimed at people starting out in Folding, it was aimed at people with ATi cards primarily. Feel free to make you're own thread for beginners, I know I'd benefit from it.

 

[TheoneandonlyMrK]

I'm a noob when it comes to folding, thought I explained that in the first post.

sorry mate it was in one eyeball and out the other with that minor detail
tooo damn much shamone

ill look into something this weekend

that weekend finally came

thanks innocent criminal , your efforts appreciated

since this is the final post ill re itterate something about the V7 client ,it needs NO flags with most amd cards and nvidia for that matter as its auto detection is better in the latest beta which is linked in the OP and amd cards pull x16 work units from the get go. also latest beta plus cat 12,3pre with opencl 1.2 beta 12.4 ,12.5 beta all tested working and have have increased my ppd(poss opencl 1.2driver) a bit yay currently 9600ppd(5870 9400(5850) ,bit more to give

 

[TheoneandonlyMrK]

oh here , bump

 

[TheoneandonlyMrK]

Bumpety bump (my first) uppcomeing CC gives its extremely usefull OP a bit of importance

 

[InnocentCriminal]

Bumpety bump (my first) uppcomeing CC gives its extremely usefull OP a bit of importance

Um, what?

 

[TheoneandonlyMrK]

Chimp challenge is two days away so im bumping again 2nd ever bump

 

[InnocentCriminal]

Arrh I see.