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

Reference Build: Z8NA-D6 "Computing" Rig

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
I had to do something... I have to slow ihrsetrdr's roll as best I can. :D

Me? Naw, the one we all have zoomin' up in our rear view mirrors is this guy. ;)

foldon.gif
 
Yeah, but I can't compete with that... have to stay in "my own league" as it were. ;)

/sorry, off topic... I'd really like to see some more data from Jmtyra on this build.
 
I'd really like to see some more data from Jmtyra on this build.
/rant
Greetings from 30k feet. Currently flying home from the "work project from hell" (80hrs+ per week), which starts my two weeks 'off'. Off, as in, I'm moving. Fun.
/end rant

Any-who, please let me know what data you'd like to see. I'll be happy to pull anything you'd like. :)

Having said that, this board will end up in the classifieds within the next few days, once said data has been obtained. I could RMA it, sure, but I've gone to the trouble of getting the updated BIOS (via eBay, $20, mind you) and it's gone through a few weeks of stability testing @ 100% CPU 24x7, so I figure it'll be better served by another folder/OC'er.

Any-who, here is the official announcement:
Please let me know of any info/data you'd like to see from this board!!

I won't bother the other DC teams, as I think they'll be just as disappointed with the overall processing power. IMHO, of course.

I'll check back here tomorrow (10/2) sometime to see and fulfill any data/info requests. Unfortunately, I'll probably have to do a hard cutoff of 10/5-ish so that I can finish packing everything. Sorry folks. :(

Until then...
 
I just want to see some actual benchmarks from HFM. That'll give us some firm numbers.

As many projects as you can post please. :)
 
Can-do, no problem! :D
Is this what you're looking for, just a screen shot?
 

Attachments

  • p6062 r0 c168 g44.png
    p6062 r0 c168 g44.png
    25.6 KB · Views: 321
Also posting what logs I have, for those that want them, etc.
 

Attachments

  • FAHlog.txt
    6.2 KB · Views: 35
  • FAHlog-Prev.txt
    82.9 KB · Views: 36
  • unitinfo.txt
    168 bytes · Views: 40
Dropping the WU and downloading another, but it gave me the same one. Not sure if it's because I'm still signed in as me, so I'll "sign-out" and try for another WU. Below is the WU before I dropped it.
 

Attachments

  • p6062 r0 c168 g44 #2.png
    p6062 r0 c168 g44 #2.png
    30 KB · Views: 321
Additional info: All at default specs/speeds. Using SMP 16; tried 17+, EUE occurs. 15 or less, not using max CPU. Dumping and re-downloading a WU as 'anonymous' still gives me the same project/run/clone/gen.

I'll let this one finish (after re-updating to my info, tee-hee) and see if I get something different.
 
Last edited:
Can-do, no problem! :D
Is this what you're looking for, just a screen shot?

Open up the benchmarks viewer (Ctrl+B).

You'll see information that looks like this...

Code:
 Project ID: 6062
 Core: GRO-A3
 Credit: 481
 Frames: 100


 Name: Windows SMP - DS4
 Path: \\Q6600-ds4\FAHSMP\
 Number of Frames Observed: 300

 Min. Time / Frame : 00:05:25 - 7,400.8 PPD
 Avg. Time / Frame : 00:05:35 - 7,071.9 PPD
 
I think that you could squeeze more production out of that baby, if you were up to a little experimentation, maybe try 2 folding instances(set up different directories, use the console version of f@h) and use the following flags with each:

Code:
-smp 8 -verbosity 9 -local

and be sure to use different machineIDs for each folding client.
 
Open up the benchmarks viewer (Ctrl+B).

You'll see information that looks like this...
Ahhh, OK, gotcha. Can do! :thup:

I think that you could squeeze more production out of that baby, if you were up to a little experimentation, maybe try 2 folding instances
Hmmm...good idea. I didn't think about that. Will try that shortly and post back. :)
 
Just got'er going, should have some numbers here in a fold or three. :)
 

Attachments

  • duality test.png
    duality test.png
    41.1 KB · Views: 315
Er...should I be concerned about these temps? It looks like, oddly enough, one of my procs aren't seated properly or something.

Going to investigate more...
 

Attachments

  • temp.png
    temp.png
    14.5 KB · Views: 308
Er...should I be concerned about these temps? It looks like, oddly enough, one of my procs aren't seated properly or something.
So, verified both CPUs are seated properly, heatsinks (2x H50) are secure and such. The 2nd CPU, which I removed earlier to test OC'ing in a different system, might have been the culprit. So, I removed all thermal paste, re-tinted and re-applied fresh paste. After re-securing said CPU/heatsink/etc, it's still sitting at the same temperature range, roughly. Even at POST, within the BIOS, the hw monitor shows an 8c difference in idle temp.

Something is wrong, but I can't identify it. :shrug:

Plus, from my earlier OC test, even at 4Ghz+ the E5620 can barely finish bigadv units. Which is weird for a 12mb cache, 4c/8t processor.

Feeling rather irritated at the whole 'energy efficient' ppd/kWh folding farm idea. Ugh. :bang head

Edit: Total side-note, but google is fast or they're cheating. Posted this less than 10 minutes ago, right? Was doing a google search for 'energy efficient folding farm' and guess what the third result was? Yup, you guessed it, this very thread with keywords for this exact post that I'm editing. Weird + scary.
 
Last edited:
Er...should I be concerned about these temps? It looks like, oddly enough, one of my procs aren't seated properly or something.

Going to investigate more...

I missed your mention of the model of Xeons you're running, I'm wondering if TJ Max is set correctly or not; the value of distance to TJ Max should be listed on the Intel site's spec sheet for that model processor.

Just for reference, maybe just go into the BIOS and let the rig sit at idle for a while, and observe the temps in the pc 'health' or 'hardware monitor' screen, and see if the cores are different, and if that difference is consistent with realtemp.
 
I'm using 2x E5620 Xeons; I think the re-seating did the trick, as they balanced back down after about 12hrs. Still not consistent between the two, but close enough.
 
the value of distance to TJ Max should be listed on the Intel site's spec sheet for that model processor.


Nope, couda sworn TJ Max was listed for Intel procs; musta glanced at Max TDP and saw TJ Max... :shrug:

Anyway, glad re-seating helped, although anything under 80C @100% makes me happy. ;)
 
Back