• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

New bigadv WU: P2684

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
ihrsetrdr

ihrsetrdr

Señor Senior Member
​
Joined
May 17, 2005
Location
High Desert, Calif.
If your using the "-bigadv" flag you may pick up one of these P2684's, which you'll notice are taking approx 50% longer TPF than the typical bigadv WU. Credit is 8955.00, the bonus calculated using Kfactor 26.40.
 
Yep... I've picked one up and it also failed on several occasions. It looks like it's finally working now. This is on an i7 @ 200x19 w/two 285s running in Win7 x64.

Here's the data from HFM:

Code: 
 Project ID: 2684
 Core: GRO-A3
 Credit: 8955
 Frames: 100


 Name: Ubuntu VM SMP - i7 920
 Number of Frames Observed: 5

 Min. Time / Frame : 00:48:34 - 18,196.1 PPD
 Avg. Time / Frame : 00:50:01 - 17,410.6 PPD
 Cur. Time / Frame : 00:48:40 - 18,077.7 PPD
 R3F. Time / Frame : 00:48:54 - 17,991.5 PPD
 All  Time / Frame : 00:49:22 - 17,821.4 PPD
 Eff. Time / Frame : 00:51:28 - 17,094.2 PPD
 
Yeah... I had this same unit fail several times in a row based on the data in the queue.dat. But now it seems to be Folding. I'm up to a whopping 9% now. :rolleyes: I'm just barely doing to make the preferred deadline by ~12 hours at the current production rate... so there's no margin for error. HFM calculates the credit in the 60000 range and that's a stark contrast to the ~9800 points I'd get for it if I miss the deadline. :(
 
So far- not a "great" track record with these P2684s:

hfm.png



BTW Ryan- what method are you using to grab the Benchmarks data that you posted in the code tags? I was unable to highlight & copy/paste either with gedit or directly into the forum message editor. I also tried wine notepad, but got errors:

Code: 
err:clipboard:X11DRV_CLIPBOARD_RenderFormat Failed to cache
 clipboard data owned by another process. Format=000d
 
They sure don't appear to be very stable... hopefully just some new WU growing pains.

I just used a good ole Ctrl+C and Ctrl+V. This is probably another "Mono issue". Since I don't use HFM daily in Mono this is just something I haven't come across. Could also be related to the particular desktop environment you're using. I've entered an issue here so I'll check it out the next time I'm compatibility testing with Mono.
 
I'm going to delve into this, and learn more about mono.There is a mono clipboard, but would take some configuration; I'll read up on the "nuts & bolts" of it.
 
Anyone else still picking these up? I am not... and I'm not sad about it. :)
 
Just got one on each rig.

I restarted VMware mid WU and when I went to resume folding....bork!
some sort of error in the WU.
 
OK, these things are giving me a headache. Had one die (needed to reboot) at ~38%, which is over a day worth of Folding down the drain... installed the Windows SMP2 Client and proceeding with normal A3 work.

Hopefully we'll get bigadv A3 on Windows SMP2 and it'll be super stable. :)
 
I've got a P2683 which I think is a 'close cousin', the rest are P2681s:



screenshotpu.png



It's shocking to see the failure rate in HFM.net, as each failure represents a large portion of between 1 and almost 3 days of computer time.

I still haven't found a way to operate the mono clipboard, I'm a little surprised that I can't use the system clipboard. there's gotta be a way. ;)
 
Yeah, looks like you've skirted the bigadv A3 work... at least for now. ;) I've got one rig doing bigadv A3 and the other on WinSMP A3. Rigs are similarly setup, both i7 920 @ 3.8GHz. The Windows client is getting just ~1000ppd less than the bigadv A3. That's what makes the bigadv A3 (p2684) so disappointing... why run the risk of it failing and dedicate all those resources when one can run a much more stable work with a minimal hit in production.

Of course, I know you're not on Windows... so this option currently isn't available. When running standard A3 work on my Linux VMs, the production was not very good. I only got ~10k ppd out of it while I'm getting ~16k ppd out of WinSMP A3 work.

I'll have a look at the clipboard issue in Linux... I hope to start some development on the next version of HFM this weekend. :)
 
Last edited:
Normally if I stop and restart these WU's it will error out. But last night we had a power outage and I had ~8 min to shut down my one rig. (1500VA UPS lol)

Shut down the CPU and GPU clients and when I restarted them this morning the P2684 is still folding away... Kinda odd.
 
ihrsetrdr said:
I still haven't found a way to operate the mono clipboard, I'm a little surprised that I can't use the system clipboard. there's gotta be a way. ;)
Click to expand...

Perhaps the mono clipboard is not the way to go; I'm thinking maybe just directly reading the BenchmarkCache.dat
with....some kind of app or text editor may be what I'm needing.
 
ihrsetrdr said:
Perhaps the mono clipboard is not the way to go; I'm thinking maybe just directly reading the BenchmarkCache.dat
with....some kind of app or text editor may be what I'm needing.
Click to expand...

no no no, just used dasher..simple! ;)

P2684 benchie:

Code: 
 Project ID: 2684
 Core: GRO-A3
 Credit: 8955
 Frames: 100

 Name: asrock-2
 Path: /home/as2share/bigadv
 Number of Frames Observed: 100

 Min. Time / Frame : 00:49:24 - 17,737.7 PPD
 Avg. Time / Frame : 00:49:48 - 17,524.4 PPD

 Name: asrock3
 Path: /home/hrsetrdr/Desktop/fah
 Number of Frames Observed: 91

 Min. Time / Frame : 00:50:04 - 17,384.6 PPD
 Avg. Time / Frame : 00:50:51 - 16,984.4 PPD
 
Argh!!!! Just had one die at 94% after 3 days of folding. And guess what just picked up the same one again. Argh!!!! Very Frustrating
 
You must log in or register to reply here.

Similar threads

don256us
And so the world turns. (Now that we all rule it)*
Replies
10
Views
627
don256us
don256us
don256us
Farwalker, Mark620 and time to catch up.
Replies
2
Views
527
Farwalker2u
Farwalker2u
don256us
New year new goals.
Replies
12
Views
790
WhitehawkEQ
WhitehawkEQ
don256us
Let's see.... What's new here?... Hmmm......
Replies
5
Views
523
dfonda
dfonda
don256us
10, 4 and 3 billion. Yep. It goes like that sometimes.
Replies
6
Views
551
don256us
don256us
Back