New Year's Thrash.

[Big Block 88]

Ha!!! Well it's my home office so odds are it was my girl or my pops did it on accident, I know it wasn't me:baby:

 

[Kat]

After the current jobs finish, Wicked Witch is going into hiding.

Ding Dong the Witch :witch: is dead

Well until I take your laptop :hug:

 

[McRat]

Congrats Kat and Tom.

Thanks again Pat for your help. Finally got a 6900 to try out. Not doing real good at the moment but at least I know where I stand with it. About ~50:00 tpf.

50:00 TPF should yield bonus. That's 5000 minutes, or uh... ~82 hours? Preferred deadline is 4 days, or 96 hours?

 

[Mike]

50:00 TPF should yield bonus. That's 5000 minutes, or uh... ~82 hours? Preferred deadline is 4 days, or 96 hours?

Looks bad too... No time for any kind of tinkering or outage.

 

[McRat]

Looks bad too... No time for any kind of tinkering or outage.

I think I read some stuff about tweaking Opteron servers. I'll see if I can find it after dinner.

Supposedly there are some switches in the BIOS that can pick up a lot of PPD.

 

[Mike]

I think I read some stuff about tweaking Opteron servers. I'll see if I can find it after dinner.

Supposedly there are some switches in the BIOS that can pick up a lot of PPD.

Thanks Pat, I'll take a look also. Look at the latest update, looks even worse than before. My guess is the chips heating up? I have another board I can try, it's an Arima NM46(1?) I believe, it's the original board for the server. I can't look at the moment to see what board is in there for the exact model, I don't want to disturb the bigadv job. :baby: ( scared and only have so much time in case of later trouble ) I believe it is an Arima also but not sure. It's a 4 socket with socket f quad core processors. Two are bad ( I think ). Thanks again Pat, you've already helped me figure out some issues with bigadv and advmethods. I turned advmethods off the one computer and the points stayed the same but the time has decreased from 1 hour to about 33 minutes tpf. I'll change the laptop if it ever finishes it's job. :thumb:

 

[Mike]

I think I read some stuff about tweaking Opteron servers. I'll see if I can find it after dinner.

Supposedly there are some switches in the BIOS that can pick up a lot of PPD.

I couldn't find anything regarding the BIOS, however, I did disconnect the netework to it and that was good for a couple of minutes per fold. I'm not sure why, may be nothing but, I was using the network to check the status via HFM.NET I now plug it in just long enough to check the status.

 

[Mike]

Looks like the bigadv job may have a half day or so spare by the end of the job, times were as bad as 58+ minutes per fold.

Has anyone tried this? vvvvv

I disabled all the services I didn't need on my Windows 7 laptop which is doing a SMP job. I also killed all the processes and quick launch programs in the tray. Seem ( for the last half day to be a bit better on completing the job and HFM.NET is reporting a little better PPD. Or, did I just repeat what has already been posted

 

[McRat]

I couldn't find anything regarding the BIOS, however, I did disconnect the netework to it and that was good for a couple of minutes per fold. I'm not sure why, may be nothing but, I was using the network to check the status via HFM.NET I now plug it in just long enough to check the status.

Read through this thread carefully: http://foldingforum.org/viewtopic.php?f=55&t=16840

I think the 6128 is part of the same BIOS family as your Opterons.

 

[McRat]

Looks like the bigadv job may have a half day or so spare by the end of the job, times were as bad as 58+ minutes per fold.

Has anyone tried this? vvvvv

I disabled all the services I didn't need on my Windows 7 laptop which is doing a SMP job. I also killed all the processes and quick launch programs in the tray. Seem ( for the last half day to be a bit better on completing the job and HFM.NET is reporting a little better PPD. Or, did I just repeat what has already been posted

Dunno, haven't played with any of that yet.

 

[Mike]

Read through this thread carefully: http://foldingforum.org/viewtopic.php?f=55&t=16840

I think the 6128 is part of the same BIOS family as your Opterons.

Thanks Pat, I've read and re-read the post and have a sticky on the screen to check for " power now " being disabled once this WU is complete. I wish I would have done the -oneunit flag.

I wonder too if this motherboard or cpu's are bad? I read where one person was getting really poor performance and found memory bad. Hell have these parts are EBAY treasures. I could have a couple of sticks bad, I'll try swapping those also.

Thanks again for helping. PPD are already getting better. Just the correct setup has been a HUGE help. Hope this helps someone else also.

 

[McRat]

Bad memory will crash (BSOD) unless you are ECC or mirror-ram, and then it normally sends an alert when it sees errors.

Lots of folk have struggled with Linux setups. bigadv used to be Linux only, now it's not officially supported from what I gather. You can run it, but at your own risk. Could be wrong, I read so much conflicting/obsolete info on the official site.

I really wanted to go Linux for my shop, and might do it in the future. But for F&H, Windows has far better support.

 

[Mike]

Bad memory will crash (BSOD) unless you are ECC or mirror-ram, and then it normally sends an alert when it sees errors.

Lots of folk have struggled with Linux setups. bigadv used to be Linux only, now it's not officially supported from what I gather. You can run it, but at your own risk. Could be wrong, I read so much conflicting/obsolete info on the official site.

I really wanted to go Linux for my shop, and might do it in the future. But for F&H, Windows has far better support.

I do have the ECC registering and it's on. I was told to use it. That may be the problem come to think of it, I should at least try it as off and see what happens. I'd read setting it active slows total throughput. I will try it after checking the other possibilities and remove the bigadv option. I should be able to get enough result info without failing a bigger job. Good tip ( or round-about one anyway ). I don't think I'm gettting errors. Guess I could look at some log files unless they would be printed to screen. Heck, I now have the GUI de-activated. Don't think it helped.

I'm mixed on the Linux thing also, seems stable, lots of support and very easy to manipulate. Just knowing what to do, though, with f@h, as you say, made me want to pull my hair out. I'm gonna try a " Server " version instead of the " Desktop " version. They have GUI support as well. I may be reading dated material as well but, seems if f@h is run as a service even with " Wine " PPD are greater.

I'm running Ubuntu 10.4 LTS as it was finally what I was able to make fold. Again, consider the computer tech.

I'm a bit confused on the memory requirements. I have 4 gigs per processor, currently 8 gigs for the two installed processors. I'd read anywhere from a half gig per core for linux to 1 gig being a bit much for linux and Windows. Seems I remember the same for windows. I've only noticed a maximum of 1.4 gig being used in the " resource monitor " Linux both cores. Also the task manager in windows, same cpu setup, showed *32bit libraries and ~.6 gig per core. I had to get *64bit libraries for linux? The Windows reference was with a 64bit Windows 2003 Server version loaded. I have 64bit 7 professional and ultimate I could try for 30 days ( ummm at a time ). Guess I should try and report back instead of wondering. :rofl:

Just trying to squeeze every bit of processor time to folding is all. I mean, If we could take advantage of every bit of BigAdv or even SMP vantage points, we could all improve. :baby: I know the operating system allocates memory in relation to requirments from programs but heck, I look at it as extra weight. All them durn' electrons have to go through the processor. Probably talking out of my rear again though.

 

[McRat]

I put in 6 gb to serve two 6-core CPU's (24 HT threads), so I'm only about 250mb per thread, but it seems fine. I found some more ram on sale, so I'm bumping it to 12, but it was folding fine with 6. I asked a HardCore SR-2 guy, and he said 6 was fine (6x 1GB chips of DDR3-1333).

Fine?
On the easier SMP jobs (6051? not the 6701/6702), the TPF is 0:00:58.
For 6701's it's 0:02:55. It does most BigAdv in 24 hours flat.:lildevil:
It's running at 3.96ghz on all 12 cores steady as a rock. Up to 115k PPD so far.

 

[Mike]

I put in 6 gb to serve two 6-core CPU's (24 HT threads), so I'm only about 250mb per thread, but it seems fine. I found some more ram on sale, so I'm bumping it to 12, but it was folding fine with 6. I asked a HardCore SR-2 guy, and he said 6 was fine (6x 1GB chips of DDR3-1333).

Fine?
On the easier SMP jobs (6051? not the 6701/6702), the TPF is 0:00:58.
For 6701's it's 0:02:55. It does most BigAdv in 24 hours flat.:lildevil:
It's running at 3.96ghz on all 12 cores steady as a rock. Up to 115k PPD so far.

Wow, your memory speed is twice this old server has in it. DDR2-667

As you've said, that is a hell-of-a-computer for a desktop. I don't think I will be building that one unless it supports four keyboards, four pointing devices, four graphics cards and four sound cards ( with ear phones ). :rofl:

I think I will do the budget AMD and try and make these ole' turds heat the house this winter ( if I can call it that ). Crazy weather.

I'm hoping resolving the cpu errors and possibly the BIOS configuration can put that server up at least to a couple of graphics cards. If so, dollar for dollar, already having the dinosaur and buying the upgrades will be about half the cost of two 470's and required upgrades to support. Anyway, time to work on the xeon dual core 2 socket, it's older also.

 

[Mike]

Looks like the cpu trouble may be worked out, 2356's and the same group only support 2p configurations. I guess I'll be getting 4 83** series opterons to use this board. :rofl: better find them cheap or I'll make a clock.

 

[Mike]

The bigadv and SMP units have a huge bonus if returned fast.

bonus factor = sqrt(deadline_time * k / WU_time)

So in the case of a 2692 WU:

sqrt( 6 days * 26.4 "K factor" / 1 day) = sqrt of ~158? or 12.5

Take the real Points, 8955 and multiply it by the 12.5 bonus factor: 112,000 points in one day from a single 8955 point job.

The K Factor, Deadline, and Points are from http://fah-web.stanford.edu/psummary.html

So in other words, the bonus can be many times the "real" points.

So far, Bonus points are only for SMP and BigAdv. But they are thinking about adding them to other types of jobs.

The reasoning for doing this with BigAdv is that they cannot do huge proteins without either a real supercomputer, or clusters of high powered servers. These jobs have over 1,000,000 atoms in them, and can't be broken down. Lots of ram and processing power is required.

Here is my log for the 6900 WU. Long and yet to finish sending. Hoping to calculate Points received [ by the formula and adding the ~32 minutes to upload, 58,405 total points for WU { hopefuly } ] FHM.NET shows 8955 points, Confused for sure. I should only have one 1348 point WU to subract from the 3 am update on oc's forum.

Just patiently ( ok NOT ) waiting on the 100mb file to send. Slow connection of 512k connection. What is that, maybe 56k/s upload time.

13:52:32] - Preparing to get new work unit...
[13:52:32] Cleaning up work directory
[13:52:32] + Attempting to get work packet
[13:52:32] Passkey found
[13:52:32] - Will indicate memory of 3905 MB
[13:52:32] - Detect CPU. Vendor: AuthenticAMD, Family: 15, Model: 2, Stepping: 2
[13:52:32] - Connecting to assignment server
[13:52:32] Connecting to http://assign.stanford.edu:8080/
[13:52:33] Posted data.
[13:52:33] Initial: ED82; - Successful: assigned to (130.237.232.141).
[13:52:33] + News From Folding@Home: Welcome to Folding@Home
[13:52:33] Loaded queue successfully.
[13:52:33] Sent data
[13:52:33] Connecting to http://130.237.232.141:8080/
[13:52:40] Posted data.
[13:52:40] Initial: 0000; - Receiving payload (expected size: 24866755)
[13:54:18] - Downloaded at ~247 kB/s
[13:54:18] - Averaged speed for that direction ~192 kB/s
[13:54:18] + Received work.
[13:54:18] Trying to send all finished work units
[13:54:18] + No unsent completed units remaining.
[13:54:18] + Closed connections
[13:54:18]
[13:54:18] + Processing work unit
[13:54:18] Core required: FahCore_a3.exe
[13:54:18] Core found.
[13:54:18] Working on queue slot 05 [January 2 13:54:18 UTC]
[13:54:18] + Working ...
[13:54:18] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 05 -np 8 -priority 96 -checkpoint 15 -verbose -lifeline 8 -version 630'
[13:54:19]
[13:54:19] *------------------------------*
[13:54:19] Folding@Home Gromacs SMP Core
[13:54:19] Version 2.22 (Mar 12, 2010)
[13:54:19]
[13:54:19] Preparing to commence simulation
[13:54:19] - Looking at optimizations...
[13:54:19] - Created dyn
[13:54:19] - Files status OK
[13:54:31] - Expanded 24866243 -> 30796293 (decompressed 123.8 percent)
[13:54:31] Called DecompressByteArray: compressed_data_size=24866243 data_size=30796293, decompressed_data_size=30796293 diff=0
[13:54:31] - Digital signature verified
[13:54:31]
[13:54:31] Project: 6900 (Run 33, Clone 3, Gen 11)
[13:54:31]
[13:54:32] Assembly optimizations on if available.
[13:54:32] Entering M.D.
[13:54:44] Completed 0 out of 250000 steps (0%)
[14:47:44] Completed 2500 out of 250000 steps (1%)
[15:39:37] Completed 5000 out of 250000 steps (2%)
[16:31:33] Completed 7500 out of 250000 steps (3%)
[16:38:31] - Autosending finished units... [January 2 16:38:31 UTC]
[16:38:31] Trying to send all finished work units
[16:38:31] + No unsent completed units remaining.
[16:38:31] - Autosend completed
[17:25:22] Completed 10000 out of 250000 steps (4%)
[18:20:54] Completed 12500 out of 250000 steps (5%)
[19:13:53] Completed 15000 out of 250000 steps (6%)
[20:08:04] Completed 17500 out of 250000 steps (7%)
[21:01:23] Completed 20000 out of 250000 steps (8%)
[21:56:02] Completed 22500 out of 250000 steps (9%)
[22:38:31] - Autosending finished units... [January 2 22:38:31 UTC]
[22:38:31] Trying to send all finished work units
[22:38:31] + No unsent completed units remaining.
[22:38:31] - Autosend completed
[22:49:55] Completed 25000 out of 250000 steps (10%)
[23:43:53] Completed 27500 out of 250000 steps (11%)
[00:36:23] Completed 30000 out of 250000 steps (12%)
[01:29:32] Completed 32500 out of 250000 steps (13%)
[02:23:03] Completed 35000 out of 250000 steps (14%)
[03:18:58] Completed 37500 out of 250000 steps (15%)
[04:14:58] Completed 40000 out of 250000 steps (16%)
[04:38:31] - Autosending finished units... [January 3 04:38:31 UTC]
[04:38:31] Trying to send all finished work units
[04:38:31] + No unsent completed units remaining.
[04:38:31] - Autosend completed
[05:11:07] Completed 42500 out of 250000 steps (17%)
[06:07:13] Completed 45000 out of 250000 steps (18%)
[07:03:19] Completed 47500 out of 250000 steps (19%)
[07:56:21] Completed 50000 out of 250000 steps (20%)
[08:47:47] Completed 52500 out of 250000 steps (21%)
[09:39:30] Completed 55000 out of 250000 steps (22%)
[10:32:06] Completed 57500 out of 250000 steps (23%)
[10:38:32] - Autosending finished units... [January 3 10:38:32 UTC]
[10:38:32] Trying to send all finished work units
[10:38:32] + No unsent completed units remaining.
[10:38:32] - Autosend completed
[11:25:01] Completed 60000 out of 250000 steps (24%)
[12:18:13] Completed 62500 out of 250000 steps (25%)
[13:10:35] Completed 65000 out of 250000 steps (26%)
[14:04:36] Completed 67500 out of 250000 steps (27%)
[15:00:49] Completed 70000 out of 250000 steps (28%)
[15:58:28] Completed 72500 out of 250000 steps (29%)
[16:38:08] - Autosending finished units... [January 3 16:38:08 UTC]
[16:38:08] Trying to send all finished work units
[16:38:08] + No unsent completed units remaining.
[16:38:08] - Autosend completed
[16:53:55] Completed 75000 out of 250000 steps (30%)
[17:47:21] Completed 77500 out of 250000 steps (31%)
[18:39:54] Completed 80000 out of 250000 steps (32%)
[19:34:52] Completed 82500 out of 250000 steps (33%)
[20:30:19] Completed 85000 out of 250000 steps (34%)
[21:25:43] Completed 87500 out of 250000 steps (35%)
[22:20:17] Completed 90000 out of 250000 steps (36%)
[22:38:06] - Autosending finished units... [January 3 22:38:06 UTC]
[22:38:06] Trying to send all finished work units
[22:38:06] + No unsent completed units remaining.
[22:38:06] - Autosend completed
[23:14:02] Completed 92500 out of 250000 steps (37%)
[00:08:34] Completed 95000 out of 250000 steps (38%)
[01:02:53] Completed 97500 out of 250000 steps (39%)
[01:55:57] Completed 100000 out of 250000 steps (40%)
[02:51:54] Completed 102500 out of 250000 steps (41%)
[03:47:47] Completed 105000 out of 250000 steps (42%)
[04:38:06] - Autosending finished units... [January 4 04:38:06 UTC]
[04:38:06] Trying to send all finished work units
[04:38:06] + No unsent completed units remaining.
[04:38:06] - Autosend completed
[04:43:39] Completed 107500 out of 250000 steps (43%)
[05:39:32] Completed 110000 out of 250000 steps (44%)
[06:35:31] Completed 112500 out of 250000 steps (45%)
[07:30:19] Completed 115000 out of 250000 steps (46%)
[08:25:31] Completed 117500 out of 250000 steps (47%)
[09:18:45] Completed 120000 out of 250000 steps (48%)
[10:11:55] Completed 122500 out of 250000 steps (49%)
[10:38:06] - Autosending finished units... [January 4 10:38:06 UTC]
[10:38:06] Trying to send all finished work units
[10:38:06] + No unsent completed units remaining.
[10:38:06] - Autosend completed
[11:04:42] Completed 125000 out of 250000 steps (50%)
[11:57:36] Completed 127500 out of 250000 steps (51%)
[12:50:24] Completed 130000 out of 250000 steps (52%)
[13:43:49] Completed 132500 out of 250000 steps (53%)
[14:36:12] Completed 135000 out of 250000 steps (54%)
[15:28:00] Completed 137500 out of 250000 steps (55%)
[16:20:38] Completed 140000 out of 250000 steps (56%)
[16:38:06] - Autosending finished units... [January 4 16:38:06 UTC]
[16:38:06] Trying to send all finished work units
[16:38:06] + No unsent completed units remaining.
[16:38:06] - Autosend completed
[17:13:34] Completed 142500 out of 250000 steps (57%)
[18:06:30] Completed 145000 out of 250000 steps (58%)
[18:59:27] Completed 147500 out of 250000 steps (59%)
[19:51:53] Completed 150000 out of 250000 steps (60%)
[20:43:42] Completed 152500 out of 250000 steps (61%)
[21:36:31] Completed 155000 out of 250000 steps (62%)
[22:29:24] Completed 157500 out of 250000 steps (63%)
[22:38:06] - Autosending finished units... [January 4 22:38:06 UTC]
[22:38:06] Trying to send all finished work units
[22:38:06] + No unsent completed units remaining.
[22:38:06] - Autosend completed
[23:22:17] Completed 160000 out of 250000 steps (64%)
[00:14:22] Completed 162500 out of 250000 steps (65%)
[01:06:25] Completed 165000 out of 250000 steps (66%)
[01:58:46] Completed 167500 out of 250000 steps (67%)
[02:51:38] Completed 170000 out of 250000 steps (68%)
[03:44:56] Completed 172500 out of 250000 steps (69%)
[04:38:06] - Autosending finished units... [January 5 04:38:06 UTC]
[04:38:06] Trying to send all finished work units
[04:38:06] + No unsent completed units remaining.
[04:38:06] - Autosend completed
[04:38:11] Completed 175000 out of 250000 steps (70%)
[05:31:27] Completed 177500 out of 250000 steps (71%)
[06:24:00] Completed 180000 out of 250000 steps (72%)
[07:16:04] Completed 182500 out of 250000 steps (73%)
[08:08:35] Completed 185000 out of 250000 steps (74%)
[09:00:53] Completed 187500 out of 250000 steps (75%)
[09:53:55] Completed 190000 out of 250000 steps (76%)
[10:38:06] - Autosending finished units... [January 5 10:38:06 UTC]
[10:38:06] Trying to send all finished work units
[10:38:06] + No unsent completed units remaining.
[10:38:06] - Autosend completed
[10:47:00] Completed 192500 out of 250000 steps (77%)
[11:40:06] Completed 195000 out of 250000 steps (78%)
[12:33:10] Completed 197500 out of 250000 steps (79%)
[13:26:20] Completed 200000 out of 250000 steps (80%)
[14:21:29] Completed 202500 out of 250000 steps (81%)
[15:17:36] Completed 205000 out of 250000 steps (82%)
[16:12:56] Completed 207500 out of 250000 steps (83%)
[16:38:06] - Autosending finished units... [January 5 16:38:06 UTC]
[16:38:06] Trying to send all finished work units
[16:38:06] + No unsent completed units remaining.
[16:38:06] - Autosend completed
[17:05:51] Completed 210000 out of 250000 steps (84%)
[17:58:50] Completed 212500 out of 250000 steps (85%)
[18:52:43] Completed 215000 out of 250000 steps (86%)
[19:48:14] Completed 217500 out of 250000 steps (87%)
[20:42:08] Completed 220000 out of 250000 steps (88%)
[21:34:37] Completed 222500 out of 250000 steps (89%)
[22:27:20] Completed 225000 out of 250000 steps (90%)
[22:38:07] - Autosending finished units... [January 5 22:38:07 UTC]
[22:38:07] Trying to send all finished work units
[22:38:07] + No unsent completed units remaining.
[22:38:07] - Autosend completed
[23:19:57] Completed 227500 out of 250000 steps (91%)
[00:14:59] Completed 230000 out of 250000 steps (92%)
[01:11:00] Completed 232500 out of 250000 steps (93%)
[02:05:34] Completed 235000 out of 250000 steps (94%)
[02:59:13] Completed 237500 out of 250000 steps (95%)
[03:52:49] Completed 240000 out of 250000 steps (96%)
[04:38:07] - Autosending finished units... [January 6 04:38:07 UTC]
[04:38:07] Trying to send all finished work units
[04:38:07] + No unsent completed units remaining.
[04:38:07] - Autosend completed
[04:46:26] Completed 242500 out of 250000 steps (97%)
[05:40:03] Completed 245000 out of 250000 steps (98%)
[06:35:25] Completed 247500 out of 250000 steps (99%)
[07:29:08] Completed 250000 out of 250000 steps (100%)
[07:29:23] DynamicWrapper: Finished Work Unit: sleep=10000
[07:29:33]
[07:29:33] Finished Work Unit:
[07:29:33] - Reading up to 52713120 from "work/wudata_05.trr": Read 52713120
[07:29:33] trr file hash check passed.
[07:29:33] - Reading up to 46992080 from "work/wudata_05.xtc": Read 46992080
[07:29:34] xtc file hash check passed.
[07:29:34] edr file hash check passed.
[07:29:34] logfile size: 211547
[07:29:34] Leaving Run
[07:29:34] - Writing 100084687 bytes of core data to disk...
[07:29:35] ... Done.
[07:29:36] - Shutting down core
[07:29:36]
[07:29:36] Folding@home Core Shutdown: FINISHED_UNIT
[07:29:38] CoreStatus = 64 (100)
[07:29:38] Unit 5 finished with 38 percent of time to deadline remaining.
[07:29:38] Updated performance fraction: 0.798553
[07:29:38] Sending work to server
[07:29:38] Project: 6900 (Run 33, Clone 3, Gen 11)

[07:29:38] + Attempting to send results [January 6 07:29:38 UTC]
[07:29:38] - Reading file work/wuresults_05.dat from core
[07:29:39] (Read 100084687 bytes from disk)
[07:29:39] Connecting to http://130.237.232.141:8080/

07:29:39] Connecting to http://130.237.232.141:8080/
[07:56:19] Posted data.
[07:56:19] Initial: 0000; - Uploaded at ~60 kB/s
[07:56:23] - Averaged speed for that direction ~59 kB/s
[07:56:23] + Results successfully sent

 

[McRat]

It should make roughly 6 * 9, or 54k, or about 15k PPD.

However, that assumes that user has over 80% success with passkey on SMP.

There is a problem though. The Preferred Deadline is 4 days, and after that they reassign the WU to another computer. While a 6900 WU (the fastest units) did make it, a 2684 (the slowest) probably won't. You will make 8955 points at completion if you exceed 4 days but less than 6, however it makes the WU get run twice.

Personally, I would go back to SMP for now, but it's your call. You should perhaps hit the 6900's in 3 days or less in order to insure that a 2684 will complete in less than 4.

 

[Mike]

Uh oh, thanks Pat.

I did not put the -oneunit argument in as I thought and it now has a 2686 WU. I will stop it with ctrl-c and set that argument and see what happens. Nothing to loose now, just may set those folks back. Ouch...

I will also wait to check the BIOS and get rid of " Wine ".



* decided not to risk a chance on killing two by waiting on the possibility of this one not making time and changing the BIOS settings with no difference to time on the next.

Anyway, Power Now was enabled in the BIOS, set it to disabled and changed the rest to what was suggested in that thread you posted Pat. We'll see. Thanks again.

 

[Mike]

It should make roughly 6 * 9, or 54k, or about 15k PPD.

However, that assumes that user has over 80% success with passkey on SMP.

There is a problem though. The Preferred Deadline is 4 days, and after that they reassign the WU to another computer. While a 6900 WU (the fastest units) did make it, a 2684 (the slowest) probably won't. You will make 8955 points at completion if you exceed 4 days but less than 6, however it makes the WU get run twice.

Personally, I would go back to SMP for now, but it's your call. You should perhaps hit the 6900's in 3 days or less in order to insure that a 2684 will complete in less than 4.

Looks like I got 58181 total points for the WU. We'll see how the 2686 WU goes though. :baby: