Just got a Project 2684

You are correct. These are Linux VMs with A3 Big Adv WUs.

The ~8 to 10K PpD drop is NOT what I am looking for.

I am sure there is going to be some tweaking, but I hope there is better efficiency.

The current WU I am working on still has 2.1 days left. It is currenly worth 61K (after bonus), but my frame times are 52:08. My other i7 is still chugging thru a regular Big Adv with ~23K PpD.
 
The other day my computer locked up for unknown reasons and I lost a 2684. Somehow I don't feel all that sad about it although I normally don't like losing WUs
 
Ok, my folding rig picked up the dreaded A3 2684 bigadv work unit in my VM last night.

*sigh*

My TPF is currently averaging between 51:50 and 52:10. I don't know if it is the VM or the A3 client, but I noticed that this A3 bigadv seems to be having problems loading up all 8 of my virtual cores. Even in Core Temp, I noticed that there were lots of dips in cpu utilization between the 4 physical cores occasionally down to 97-98%. When running the A2 bigadv stuff, there wasn't much of an issue as all 8 threads were at 100% utilization with occasional dips to 99% which I assume are just some inefficiencies of running an emulator like the VM.

This means that i'm now down to 16.4k ppd which is 6-7k ppd lower than doing the A2 bigadv in the VM. This also means that I will be cutting the 4 day deadline very close as it will take me 3.6 - 3.7 days to finish this work unit. If I was running standard A3's in the Windows SMP client, it would be netting me 12-14k ppd at my current overclock.

I just hope that they get the windows bigadv going soon............
 
amdgamer said:
My TPF is currently averaging between 51:50 and 52:10. I don't know if it is the VM or the A3 client, but I noticed that this A3 bigadv seems to be having problems loading up all 8 of my virtual cores. Even in Core Temp, I noticed that there were lots of dips in cpu utilization between the 4 physical cores occasionally down to 97-98%. When running the A2 bigadv stuff, there wasn't much of an issue as all 8 threads were at 100% utilization with occasional dips to 99% which I assume are just some inefficiencies of running an emulator like the VM.
Click to expand...
This might appear a bit strange but there could be a silver lining here. It just indicates that Stanford has not optimized A3 -bigadv yet and we should see improvements all around once a newer core and/or newer WUs are released - I hope. The low performance is definitely attributable at least in part to inefficient core use. Once that increases to a near steady 100%, TPFs should drop.

/crosses fingers
 
Update on the one I am folding - at 67%, still holding at around 48:30 per frame, just over 18K ppd. It should upload tomorrow evening at around 9 CST, which is about 14 hours ahead of the deadline.
 
Dual E5530's at 2.4 (no turbo) - 39.36 TPF, 24.5K PPD. This machine does close to 29K PPD on A2 bigadv. Better, but still not quite up to A2 bigadv levels. I'd like to see someone with a higher end dually get ahold of one of these.
 
wow.....

My i7 -bigadv folder (in a VM) just grabbed one of these. 50 min TPF. 12 hours of buffer time from the deadline.

This is going to be one hell of an upload. :eek:
 
Also consider that as someone has stated earlier A3 in linux is a lot slower than windows so when we get the A3 windows client we might still be OK....
 
It better frickin' hurry!! I'm ready to move to Windows A3, but this Linux A3 is KILLING ME!!
 
Well I just picked up a second 2684 :(

I'm down 14K PpD between the two i7 920's.
 
Still getting good old a2s
Posted via [H] Mobile Device
 
Vaulter98c said:
Still getting good old a2s
Posted via [H] Mobile Device
Click to expand...

Care to share any tips with us? :D

I cannot believe how I am on my 2nd full day of this 2684 and I am nowhere near done. This is going to be a nailbiter for me as my rig will cut it very close.
 
Looks like I'm lucky this time around. I just posted 2 A2 bigadvs and got 2 A2s in return.
 
Results size and time until deadline:

Code: 
[01:04:59] Finished Work Unit:
[01:04:59] - Reading up to 52629024 from "work/wudata_02.trr": Read 52629024
[01:04:59] trr file hash check passed.
[01:04:59] - Reading up to 47018516 from "work/wudata_02.xtc": Read 47018516
[01:04:59] xtc file hash check passed.
[01:04:59] edr file hash check passed.
[01:04:59] logfile size: 207149
[01:04:59] Leaving Run
[01:05:00] - Writing 100022633 bytes of core data to disk...
[01:05:01]   ... Done.
[01:05:17] - Shutting down core
[01:05:17] 
[01:05:17] Folding@home Core Shutdown: FINISHED_UNIT
[01:05:19] CoreStatus = 64 (100)
[01:05:19] Unit 2 finished with 44 percent of time to deadline remaining.
[01:05:19] Updated performance fraction: 0.684310
[01:05:19] Sending work to server
[01:05:19] Project: 2684 (Run 0, Clone 0, Gen 3)
 
Well, got my first 2684 today. Can't say I like it but it seems to be doing a little better than some people posted here. I see about a 48 min TPF. I expected worse however it's in native Linux. This is a slow system relatively speaking; no i7 here. Only running @2.86GHz with Core2 architecture and only 4GB RAM, but 8 true cores FWIW.
 
Mr. Pedantic said:
100MB upload...how does that compare to the A2 bigadvs?
Click to expand...

That is right on par with the 2681-2683 A2 bigadv stuff as those three usually vary between 95 - 101megs when uploading. I wonder whether the A3 2684 bigadv is doing something different, or the A3 core is just not optimized for bigadv as others have suggested.
 
A3 is sucky on Linux, I'm assuming that just grows exponetialy (sp is way off lol) with A3 bigadv's
Posted via [H] Mobile Device
 
I sure hope it gets optimized better for Linux soon. I really don't want to buy Windows licenses.
 
Ok, I have been out all day and come back home to see a log full. Apparantly i've been having some major problems with the 2684 A3 bigadv stuff and am convinced that this crap isn't ready yet. My percentage has likely taken a major hit as apparantly it crashed several times after it finished the 2684 it was working on. Apparantly the next several 2684's it drew all crashed during the first 1%.

Code: 
[B]13:18:37] Completed 242500 out of 250000 steps (97%) 
[14:08:48] Completed 245000 out of 250000 steps (98%) 
[14:58:58] Completed 247500 out of 250000 steps (99%) 
[15:49:09] Completed 250000 out of 250000 steps (100%) 
[15:49:19] DynamicWrapper: Finished Work Unit: sleep=10000 
[15:49:29] 
[15:49:29] Finished Work Unit: 
[15:49:29] - Reading up to 52629024 from "work/wudata_08.trr": Read 52629024 
[15:49:30] trr file hash check passed. 
[15:49:30] - Reading up to 47016272 from "work/wudata_08.xtc": Read 47016272 
[15:49:31] xtc file hash check passed. 
[15:49:31] edr file hash check passed. 
[15:49:31] logfile size: 207748 
[15:49:31] Leaving Run 
[15:49:35] - Writing 100020988 bytes of core data to disk... 
[15:49:38] ... Done. 
[15:49:58] - Shutting down core 
[15:49:58] 
[15:49:58] Folding@home Core Shutdown: FINISHED_UNIT 
[15:49:59] CoreStatus = 64 (100) 
[15:49:59] Sending work to server 
[15:49:59] Project: 2684 (Run 3, Clone 17, Gen 0) 


[15:49:59] + Attempting to send results [May 30 15:49:59 UTC] 
[16:17:13] + Results successfully sent 
[16:17:13] Thank you for your contribution to Folding@Home. 
[16:17:13] + Number of Units Completed: 8 

[16:17:40] - Preparing to get new work unit... 
[16:17:40] Cleaning up work directory 
[16:17:40] + Attempting to get work packet 
[16:17:40] Passkey found 
[16:17:40] - Connecting to assignment server 
[16:17:41] - Successful: assigned to (171.67.108.22). 
[16:17:41] + News From Folding@Home: Welcome to Folding@Home 
[16:17:41] Loaded queue successfully. 
[16:18:43] + Closed connections 
[16:18:43] 
[16:18:43] + Processing work unit 
[16:18:43] Core required: FahCore_a3.exe 
[16:18:43] Core found. 
[16:18:43] Working on queue slot 09 [May 30 16:18:43 UTC] 
[16:18:43] + Working ... 
[16:18:43] 
[16:18:43] *------------------------------* 
[16:18:43] Folding@Home Gromacs SMP Core 
[16:18:43] Version 2.19 (March 6, 2010) 
[16:18:43] 
[16:18:43] Preparing to commence simulation 
[16:18:43] - Looking at optimizations... 
[16:18:43] - Created dyn 
[16:18:43] - Files status OK 
[16:18:45] - Expanded 24834263 -> 30791309 (decompressed 123.9 percent) 
[16:18:45] Called DecompressByteArray: compressed_data_size=24834263 data_size=30791309, decompressed_data_size=30791309 diff=0 
[16:18:45] - Digital signature verified 
[16:18:45] 
[16:18:45] Project: 2684 (Run 4, Clone 2, Gen 1) 
[16:18:45] 
[16:18:45] Assembly optimizations on if available. 
[16:18:45] Entering M.D. 
[16:18:58] Completed 0 out of 250000 steps (0%) 
[16:19:01] CoreStatus = 0 (0) 
[16:19:01] Sending work to server 
[16:19:01] Project: 2684 (Run 4, Clone 2, Gen 1) 
[16:19:01] - Error: Could not get length of results file work/wuresults_09.dat 
[16:19:01] - Error: Could not read unit 09 file. Removing from queue. 
[16:19:01] - Preparing to get new work unit... 
[16:19:01] Cleaning up work directory 
[16:19:01] + Attempting to get work packet 
[16:19:01] Passkey found 
[16:19:01] - Connecting to assignment server 
[16:19:02] - Successful: assigned to (171.67.108.22). 
[16:19:02] + News From Folding@Home: Welcome to Folding@Home 
[16:19:02] Loaded queue successfully. 
[16:19:58] + Closed connections 
[16:20:03] 
[16:20:03] + Processing work unit 
[16:20:03] Core required: FahCore_a3.exe 
[16:20:03] Core found. 
[16:20:03] Working on queue slot 00 [May 30 16:20:03 UTC] 
[16:20:03] + Working ... 
[16:20:03] 
[16:20:03] *------------------------------* 
[16:20:03] Folding@Home Gromacs SMP Core 
[16:20:03] Version 2.19 (March 6, 2010) 
[16:20:03] 
[16:20:03] Preparing to commence simulation 
[16:20:03] - Ensuring status. Please wait. 
[16:20:13] - Looking at optimizations... 
[16:20:13] - Working with standard loops on this execution. 
[16:20:13] - Created dyn 
[16:20:13] - Files status OK 
[16:20:15] - Expanded 24834263 -> 30791309 (decompressed 123.9 percent) 
[16:20:15] Called DecompressByteArray: compressed_data_size=24834263 data_size=30791309, decompressed_data_size=30791309 diff=0 
[16:20:15] - Digital signature verified 
[16:20:15] 
[16:20:15] Project: 2684 (Run 4, Clone 2, Gen 1) 
[16:20:15] 
[16:20:15] Entering M.D. 
[16:20:28] Completed 0 out of 250000 steps (0%) 
[16:20:30] CoreStatus = 0 (0) 
[16:20:30] Sending work to server 
[16:20:30] Project: 2684 (Run 4, Clone 2, Gen 1) 
[16:20:30] - Error: Could not get length of results file work/wuresults_00.dat 
[16:20:30] - Error: Could not read unit 00 file. Removing from queue. 
[16:20:30] - Preparing to get new work unit... 
[16:20:30] Cleaning up work directory 
[16:20:31] + Attempting to get work packet 
[16:20:31] Passkey found 
[16:20:31] - Connecting to assignment server 
[16:20:32] - Successful: assigned to (171.67.108.22). 
[16:20:32] + News From Folding@Home: Welcome to Folding@Home 
[16:20:32] Loaded queue successfully. 
[16:21:31] + Closed connections 
[16:21:36] 
[16:21:36] + Processing work unit 
[16:21:36] Core required: FahCore_a3.exe 
[16:21:36] Core found. 
[16:21:36] Working on queue slot 01 [May 30 16:21:36 UTC] 
[16:21:36] + Working ... 
[16:21:36] 
[16:21:36] *------------------------------* 
[16:21:36] Folding@Home Gromacs SMP Core 
[16:21:36] Version 2.19 (March 6, 2010) 
[16:21:36] 
[16:21:36] Preparing to commence simulation 
[16:21:36] - Ensuring status. Please wait. 
[16:21:46] - Looking at optimizations... 
[16:21:46] - Working with standard loops on this execution. 
[16:21:46] - Created dyn 
[16:21:46] - Files status OK 
[16:21:48] - Expanded 24834263 -> 30791309 (decompressed 123.9 percent) 
[16:21:48] Called DecompressByteArray: compressed_data_size=24834263 data_size=30791309, decompressed_data_size=30791309 diff=0 
[16:21:48] - Digital signature verified 
[16:21:48] 
[16:21:48] Project: 2684 (Run 4, Clone 2, Gen 1) 
[16:21:48] 
[16:21:48] Entering M.D. 
[16:22:00] Completed 0 out of 250000 steps (0%) 
[16:22:02] mdrun returned 255 
[16:22:02] Going to send back what have done -- stepsTotalG=250000 
[16:22:02] Work fraction=4294967296.0000 steps=250000. 
[16:22:07] logfile size=12502 infoLength=12502 edr=25 trr=1 
[16:22:07] logfile size: 12502 info=12502 bed=25 hdr=1 
[16:22:07] - Writing 13040 bytes of core data to disk... 
[16:22:08] ... Done. 
[16:22:16] 
[16:22:16] Folding@home Core Shutdown: UNSTABLE_MACHINE 
[16:22:16] CoreStatus = 7A (122) 
[16:22:16] Sending work to server 
[16:22:16] Project: 2684 (Run 4, Clone 2, Gen 1) 


[16:22:16] + Attempting to send results [May 30 16:22:16 UTC] 
[16:22:17] + Results successfully sent 
[16:22:17] Thank you for your contribution to Folding@Home. 
[16:22:17] - Preparing to get new work unit... 
[16:22:17] Cleaning up work directory 
[16:22:24] + Attempting to get work packet 
[16:22:24] Passkey found 
[16:22:24] - Connecting to assignment server 
[16:22:25] - Successful: assigned to (171.67.108.22). 
[16:22:25] + News From Folding@Home: Welcome to Folding@Home 
[16:22:25] Loaded queue successfully. 
[16:23:39] + Closed connections 
[16:23:44] 
[16:23:44] + Processing work unit 
[16:23:44] Core required: FahCore_a3.exe 
[16:23:44] Core found. 
[16:23:44] Working on queue slot 02 [May 30 16:23:44 UTC] 
[16:23:44] + Working ... 
[16:23:44] 
[16:23:44] *------------------------------* 
[16:23:44] Folding@Home Gromacs SMP Core 
[16:23:44] Version 2.19 (March 6, 2010) 
[16:23:44] 
[16:23:44] Preparing to commence simulation 
[16:23:44] - Looking at optimizations... 
[16:23:44] - Created dyn 
[16:23:44] - Files status OK 
[16:23:46] - Expanded 24834263 -> 30791309 (decompressed 123.9 percent) 
[16:23:46] Called DecompressByteArray: compressed_data_size=24834263 data_size=30791309, decompressed_data_size=30791309 diff=0 
[16:23:46] - Digital signature verified 
[16:23:46] 
[16:23:46] Project: 2684 (Run 4, Clone 2, Gen 1) 
[16:23:46] 
[16:23:46] Assembly optimizations on if available. 
[16:23:46] Entering M.D. 
[16:23:59] Completed 0 out of 250000 steps (0%) 
[16:24:01] CoreStatus = 0 (0) 
[16:24:01] Sending work to server 
[16:24:01] Project: 2684 (Run 4, Clone 2, Gen 1) 
[16:24:01] - Error: Could not get length of results file work/wuresults_02.dat 
[16:24:01] - Error: Could not read unit 02 file. Removing from queue. 
[16:24:01] - Preparing to get new work unit... 
[16:24:01] Cleaning up work directory 
[16:24:08] + Attempting to get work packet 
[16:24:08] Passkey found 
[16:24:08] - Connecting to assignment server 
[16:24:08] - Successful: assigned to (171.67.108.22). 
[16:24:08] + News From Folding@Home: Welcome to Folding@Home 
[16:24:09] Loaded queue successfully. 
[16:24:10] - Attempt #1 to get work failed, and no other work to do. 
Waiting before retry. 
[16:24:18] + Attempting to get work packet 
[16:24:18] Passkey found 
[16:24:18] - Connecting to assignment server 
[16:24:19] - Successful: assigned to (171.67.108.22). 
[16:24:19] + News From Folding@Home: Welcome to Folding@Home 
[16:24:19] Loaded queue successfully. 
[16:24:20] - Attempt #2 to get work failed, and no other work to do. 
Waiting before retry. 
[16:24:45] + Attempting to get work packet 
[16:24:45] Passkey found 
[16:24:45] - Connecting to assignment server 
[16:24:45] - Successful: assigned to (171.67.108.22). 
[16:24:45] + News From Folding@Home: Welcome to Folding@Home 
[16:24:45] Loaded queue successfully. 
[16:25:56] + Closed connections 
[16:26:01] 
[16:26:01] + Processing work unit 
[16:26:01] Core required: FahCore_a3.exe 
[16:26:01] Core found. 
[16:26:01] Working on queue slot 03 [May 30 16:26:01 UTC] 
[16:26:01] + Working ... 
[16:26:01] 
[16:26:01] *------------------------------* 
[16:26:01] Folding@Home Gromacs SMP Core 
[16:26:01] Version 2.19 (March 6, 2010) 
[16:26:01] 
[16:26:01] Preparing to commence simulation 
[16:26:01] - Ensuring status. Please wait. 
[16:26:11] - Looking at optimizations... 
[16:26:11] - Working with standard loops on this execution. 
[16:26:11] - Created dyn 
[16:26:11] - Files status OK 
[16:26:13] - Expanded 24834263 -> 30791309 (decompressed 123.9 percent) 
[16:26:13] Called DecompressByteArray: compressed_data_size=24834263 data_size=30791309, decompressed_data_size=30791309 diff=0 
[16:26:13] - Digital signature verified 
[16:26:13] 
[16:26:13] Project: 2684 (Run 4, Clone 2, Gen 1) 
[16:26:13] 
[16:26:13] Entering M.D. 
[16:26:25] Completed 0 out of 250000 steps (0%) 
[16:26:27] CoreStatus = 0 (0) 
[16:26:27] Sending work to server 
[16:26:27] Project: 2684 (Run 4, Clone 2, Gen 1) 
[16:26:27] - Error: Could not get length of results file work/wuresults_03.dat 
[16:26:27] - Error: Could not read unit 03 file. Removing from queue. 
[16:26:27] - Preparing to get new work unit... 
[16:26:27] Cleaning up work directory 
[16:26:28] + Attempting to get work packet 
[16:26:28] Passkey found 
[16:26:28] - Connecting to assignment server 
[16:26:29] - Successful: assigned to (171.67.108.22). 
[16:26:29] + News From Folding@Home: Welcome to Folding@Home 
[16:26:29] Loaded queue successfully. 
[16:26:30] - Attempt #1 to get work failed, and no other work to do. 
Waiting before retry. 
[16:26:45] + Attempting to get work packet 
[16:26:45] Passkey found 
[16:26:45] - Connecting to assignment server 
[16:26:45] - Successful: assigned to (171.67.108.22). 
[16:26:45] + News From Folding@Home: Welcome to Folding@Home 
[16:26:45] Loaded queue successfully. 
[16:26:47] - Attempt #2 to get work failed, and no other work to do. 
Waiting before retry. 
[16:27:11] + Attempting to get work packet 
[16:27:11] Passkey found 
[16:27:11] - Connecting to assignment server 
[16:27:12] - Successful: assigned to (171.67.108.22). 
[16:27:12] + News From Folding@Home: Welcome to Folding@Home 
[16:27:12] Loaded queue successfully. 
[16:27:13] - Attempt #3 to get work failed, and no other work to do. 
Waiting before retry. 
[16:27:45] + Attempting to get work packet 
[16:27:45] Passkey found 
[16:27:45] - Connecting to assignment server 
[16:27:46] - Successful: assigned to (171.67.108.22). 
[16:27:46] + News From Folding@Home: Welcome to Folding@Home 
[16:27:46] Loaded queue successfully. 
[16:27:47] - Attempt #4 to get work failed, and no other work to do. 
Waiting before retry. 
[16:28:31] + Attempting to get work packet 
[16:28:31] Passkey found 
[16:28:31] - Connecting to assignment server 
[16:28:32] - Successful: assigned to (171.67.108.22). 
[16:28:32] + News From Folding@Home: Welcome to Folding@Home 
[16:28:32] Loaded queue successfully. 
[16:29:27] + Closed connections 
[16:29:32] 
[16:29:32] + Processing work unit 
[16:29:32] Core required: FahCore_a3.exe 
[16:29:32] Core found. 
[16:29:32] Working on queue slot 04 [May 30 16:29:32 UTC] 
[16:29:32] + Working ... 
[16:29:32] 
[16:29:32] *------------------------------* 
[16:29:32] Folding@Home Gromacs SMP Core 
[16:29:32] Version 2.19 (March 6, 2010) 
[16:29:32] 
[16:29:32] Preparing to commence simulation 
[16:29:32] - Ensuring status. Please wait. 
[16:29:42] - Looking at optimizations... 
[16:29:42] - Working with standard loops on this execution. 
[16:29:42] - Created dyn 
[16:29:42] - Files status OK 
[16:29:43] - Expanded 24834263 -> 30791309 (decompressed 123.9 percent) 
[16:29:43] Called DecompressByteArray: compressed_data_size=24834263 data_size=30791309, decompressed_data_size=30791309 diff=0 
[16:29:44] - Digital signature verified 
[16:29:44] 
[16:29:44] Project: 2684 (Run 4, Clone 2, Gen 1) 
[16:29:44] 
[16:29:44] Entering M.D. 
[16:29:56] Completed 0 out of 250000 steps (0%) 
[17:22:17] Completed 2500 out of 250000 steps (1%) 
[18:13:59] Completed 5000 out of 250000 steps (2%) 
[19:05:08] Completed 7500 out of 250000 steps (3%) 
[19:55:43] Completed 10000 out of 250000 steps (4%) 
[20:46:09] Completed 12500 out of 250000 steps (5%) 
[21:36:32] Completed 15000 out of 250000 steps (6%) 
[22:26:49] Completed 17500 out of 250000 steps (7%) 
[23:17:03] Completed 20000 out of 250000 steps (8%) 
[00:07:13] Completed 22500 out of 250000 steps (9%) 
[00:57:24] Completed 25000 out of 250000 steps (10%) 
[/B]
 
Last edited by a moderator:
Well, I have officialy given up on 2684 and Linux A3 Big Adv. A secondary i7 rig has been completely unstable for 2 days in a row and has lost countless WUs.

I've switched the rig to Windows SMP2 A3. It IS stable and frankly the reduced PpD, I can suffer with because it's about the same as I was getting with 2684 WHEN IT FINISHED!!!

If my sig rig has issues, it is SMP2 A3 for it as well. PpD at times is LOWER than I was getting with SMP2 A3.

 
As a side note, it seemed on my rigs that I could push an OC a tad farther onA3 windows than I could on bigadv's, might be able to make up a few more points that way
Posted via [H] Mobile Device
 
I'm reading many folders having issues with A3 -bigadv and i7 machines. I hope it's just the 'newfangledness' of the WUs.
 
Im thinking Stanford needs more beta guys, it seems a lot of less than perfect stuffs have come out from then
Posted via [H] Mobile Device
 
use the [.code][./code] tags amdgamer, damnit! (without the . in them)
 
Yea, I'm on my phone and you just raped it lol
Posted via [H] Mobile Device
 
Sorry about that, I had no idea how to put the log file in a window as i've seen some of you guys do. So far the current 2684 it is working on hasn't had a single problem yet, but it butchered 6 2684's as you can see in that log file.

The problem is that i'm afraid this might take me under the 80% as I havn't folded long, and have folded mostly bigadv. This is a shame as A2 bigadv was so stable for most of us.
 
I have probably been about 50/50 on whether the 2684 units process or crash out immediately. These are mainly on a stock speed dually, so an unstable overclock is not the issue.
 
musky said:
I have probably been about 50/50 on whether the 2684 units process or crash out immediately. These are mainly on a stock speed dually, so an unstable overclock is not the issue.
Click to expand...

People are definatley having problems as there is a thread over at the stanford forums. After I finish my current 2684, I think i'm going to just stick with Windows A3 SMP for now as the points difference isn't really big anyway now. I wonder whether the VM has something to do with these stability problems with these WU's, although A2's were 100% stable.
 
After my sig rig finishes the Big Adv WU it is working on, I may switch it to A3. I am leaving for the week and don't want to have to deal with instability while I am gone.

The PpD between A3 Big Adv and A3 SMP2 is minimal, but EXTREMELY stable. At this point, I'd rather be stable then lose a WU I've been working 3 days on.
 
My i7 930 @3.8 is running quite well with these. It is however getting less ppd: 17.6k ppd inside a Ubuntu 10.04 VM ( should be getting ~25k ppd with old -bigadv). I should be able to get 14k ppd with a normal SMP A3. As others have said, less RAM better stability... so on and so forth.

/waits for Win -bigadv
 
amdgamer said:
Sorry about that, I had no idea how to put the log file in a window as i've seen some of you guys do. So far the current 2684 it is working on hasn't had a single problem yet, but it butchered 6 2684's as you can see in that log file.

The problem is that i'm afraid this might take me under the 80% as I havn't folded long, and have folded mostly bigadv. This is a shame as A2 bigadv was so stable for most of us.
Click to expand...

I would take out the -bigadv flag on your VM, OR run the WinSMP version with the -bigadv and -advmethods flags, and allocate a fairly large sum of RAM to it, something like 2560mb to ensure you don't have an issue if/when BigAdv A3 for WinSMP comes.

All my rigs are 24 hour Prime95 and Linx stable and I've gotten multiple failures with BigAdv A3 units in my Linux VM. I'd rather take the hit and run WinSMP because slightly lower points is better than no points.

I do believe A3 for Linux is definitely not ready for prime time yet. Many have failed at the 0% mark.
 
Well my primary w/VM just grabbed one of these, hopefully it doesn't flush.

Code: 
[16:38:07] + Processing work unit
[16:38:07] Core required: FahCore_a3.exe
[16:38:07] Core found.
[16:38:07] Working on queue slot 03 [June 1 16:38:07 UTC]
[16:38:07] + Working ...
[16:38:08] 
[16:38:08] *------------------------------*
[16:38:08] Folding@Home Gromacs SMP Core
[16:38:08] Version 2.19 (March 6, 2010)
[16:38:08] 
[16:38:08] Preparing to commence simulation
[16:38:08] - Looking at optimizations...
[16:38:08] - Created dyn
[16:38:08] - Files status OK
[16:38:10] - Expanded 24834670 -> 30791309 (decompressed 123.9 percent)
[16:38:10] Called DecompressByteArray: compressed_data_size=24834670 data_size=30791309, decompressed_data_size=30791309 diff=0
[16:38:10] - Digital signature verified
[16:38:10] 
[16:38:10] Project: 2684 (Run 4, Clone 16, Gen 2)
[16:38:10] 
[16:38:10] Assembly optimizations on if available.
[16:38:10] Entering M.D.
[16:38:30] Completed 0 out of 250000 steps  (0%)
[17:27:04] Completed 2500 out of 250000 steps  (1%)

2684a3.png

I will say I'm pleased with the credit (pts+bonus I assume), TPF and ETA are disappointing but oh well...not too bad I suppose from a pts comparison perspective (A2 bigadv - A3 bigadv) especially since this is in a VM.

What is the recourse now...what should I do? (what should we do?)
Switch over all *nix VM to Win SMP2?
My dedicated boxen will be grabbing another WU in about 7hr 28min once it completes the current A2 -bigadv.

Is there any word on how much longer A2 -bigadv's will be handed out? I mean if there's WU's available I'll stay put but then again if there's only a handful left I don't want my stuff sitting and crunching A3 in *nix if the stability is not there (which it sounds is the case).
 
Last edited:
10e said:
I would take out the -bigadv flag on your VM, OR run the WinSMP version with the -bigadv and -advmethods flags, and allocate a fairly large sum of RAM to it, something like 2560mb to ensure you don't have an issue if/when BigAdv A3 for WinSMP comes.

All my rigs are 24 hour Prime95 and Linx stable and I've gotten multiple failures with BigAdv A3 units in my Linux VM. I'd rather take the hit and run WinSMP because slightly lower points is better than no points.

I do believe A3 for Linux is definitely not ready for prime time yet. Many have failed at the 0% mark.
Click to expand...

APOLLO said:
Yup, count me in with 0% EUE group. :(
Click to expand...

Yup, all of my 6 failures with A3 bigadv happened within the first 1% :(

Welcome to the club guys! We are know all honorary members of the A3 bigadv 0% EUE club!
 
amdgamer said:
Yup, all of my 6 failures with A3 bigadv happened within the first 1% :(

Welcome to the club guys! We are know all honorary members of the A3 bigadv 0% EUE club!
Click to expand...

Thanks. I feel special ! :p

Zero2Dash: I'm also noticing that your 2684 is starting with Core 2.19 as did AMDGamer's, so if it fails, you may want to stop the client and delete the queue.dat file, work directory and FAHCore_xx.exe files in the folding directory, it accessible.

Mine has been on 2.21. Not that it's made a difference.

I now have 2x2684s crunching away. If they fail, I'm going to 100% switch all boxes to WinSMP and hope for the best.

This is why I never asked to be a beta tester :rolleyes:
 
You must log in or register to reply here.
Back
Top