Windows SMP frame times....post them here.

Thanks for the update Killer. I'll be having some new specs for you in the next few days.....hopefully what an E6420 can do at 3.6ghz.....maybe 3.8g.

Now I'll probably jinx myself to only a 3ghz oc.......shhhhhh :)
 
Xilikon said:
No on here but it is a awesome idea. I will have 2 linux SMP systems to report the details.

Click to expand...

I have two as well, just up and running this week. I think I have everything working pretty good so far.

Both are C2D E6600 running at 3.4ghz w/1gig ram under Ubuntu, no gui.
WU Average time
2609 - 12.75
2608 - 12.75
2605 - 10.35
2604 - 10.40
 
charpman said:
Sweet! thank you! :)

I have some X2 4200's running Linux SMP in VMWare I'll get that info to you as well as my WindowsSMP times.
Click to expand...

You're welcome.
On the VMware times, make sure your clocks are correct in the VM. Mine were not so I used a stopwatch to measure the frametimes.
 
I've got my E6600 up to 3.41 GHz (8*427, 4-4-4-12, 1.5V) running Windows SMP (Project 2610) and I'm getting ~11:09-11:20 per step.

 
aldamon said:
I've got my E6600 up to 3.41 GHz (8*427, 4-4-4-12, 1.5V) running Windows SMP (Project 2610) and I'm getting ~11:09-11:20 per step.

Click to expand...

That makes me feel a little better since my 6420 is stuck at 3.2ghz on stock cooling until I get it setup in my water cooling loop. At 3.2 on p2610 I average 10:31 / frame. P2652 is 11:10 / frame
 
SpoogeMonkey said:
That makes me feel a little better since my 6420 is stuck at 3.2ghz on stock cooling until I get it setup in my water cooling loop. At 3.2 on p2610 I average 10:31 / frame. P2652 is 11:10 / frame
Click to expand...

Looks like I settled into the 10:57 range over night when not using the computer with P2610 (Run 1, Clone 313, Gen 16).

[08:34:10] Completed 350000 out of 500000 steps (70 percent)
[08:45:07] Writing local files
[08:45:07] Completed 355000 out of 500000 steps (71 percent)
[08:56:05] Writing local files
[08:56:05] Completed 360000 out of 500000 steps (72 percent)
[09:07:02] Writing local files
[09:07:02] Completed 365000 out of 500000 steps (73 percent)
[09:17:58] Writing local files
[09:17:58] Completed 370000 out of 500000 steps (74 percent)
[09:28:55] Writing local files
[09:28:55] Completed 375000 out of 500000 steps (75 percent)
[09:39:53] Writing local files
[09:39:53] Completed 380000 out of 500000 steps (76 percent)
[09:50:51] Writing local files
[09:50:51] Completed 385000 out of 500000 steps (77 percent)
[10:01:48] Writing local files
[10:01:48] Completed 390000 out of 500000 steps (78 percent)
[10:12:45] Writing local files
[10:12:46] Completed 395000 out of 500000 steps (79 percent)
[10:23:43] Writing local files
[10:23:43] Completed 400000 out of 500000 steps (80 percent)
[10:34:44] Writing local files
[10:34:44] Completed 405000 out of 500000 steps (81 percent)
[10:45:48] Writing local files
[10:45:48] Completed 410000 out of 500000 steps (82 percent)

SpoogeMonkey, I'm curious how you're still 26 seconds faster per frame when I have 200 MHz on you. Was that a typo? Am I killing my times with a faster FSB (DS3 latencies) or are there differences between 2610 units?

 
dual quad cores - Xeon 5345

Project 2610 8min/frame
Project 2651 11min/frame

These are the projects that it's working on right now


 
Project 2652 on my E4300 @ 3.33 GHz

Code: 
[21:58:35] Project: 2652 (Run 0, Clone 155, Gen 19)
[21:58:35] 
[21:58:35] Assembly optimizations on if available.
[21:58:35] Entering M.D.
[21:58:41] Rejecting checkpoint
[21:58:42] ing local files
[21:58:42] Extra SSE boost OK.
[21:58:42] 
[21:58:42] Extra SSE boost OK.
[21:58:42] Writing local files
[21:58:42] Completed 0 out of 1000000 steps  (0 percent)
[22:10:33] Writing local files
[22:10:33] Completed 10000 out of 1000000 steps  (1 percent)
[22:22:21] Writing local files
[22:22:21] Completed 20000 out of 1000000 steps  (2 percent)
[22:34:12] Writing local files
[22:34:12] Completed 30000 out of 1000000 steps  (3 percent)
[22:45:55] Writing local files
[22:45:55] Completed 40000 out of 1000000 steps  (4 percent)
[22:57:49] Writing local files
[22:57:49] Completed 50000 out of 1000000 steps  (5 percent)
[23:09:18] Writing local files
[23:09:18] Completed 60000 out of 1000000 steps  (6 percent)
[23:20:40] Writing local files
[23:20:40] Completed 70000 out of 1000000 steps  (7 percent)
[23:32:29] Writing local files
[23:32:29] Completed 80000 out of 1000000 steps  (8 percent)
[23:44:20] Writing local files
[23:44:20] Completed 90000 out of 1000000 steps  (9 percent)
[23:56:02] Writing local files
[23:56:02] Completed 100000 out of 1000000 steps  (10 percent)

Project 2610 on the same setup (slow times, thanks to my crappy RAM while the FireStix is now in RMA) :

Code: 
[16:59:59] Project: 2610 (Run 0, Clone 52, Gen 22)
[16:59:59] 
[16:59:59] Assembly optimizations on if available.
[16:59:59] Entering M.D.
[17:00:16] 7 percent)
[17:00:16] - Starting from initial work packet
[17:00:16] 
[17:00:16] Project: 2Entering M.D.
[17:00:16] one 52, Gen 22)
[17:00:16] 
[17:00:17] Entering M.D.
[17:00:23] Rejecting checkpoint
[17:00:24] Protein: Protein
[17:00:24] Writing local files
[17:00:25] Extra SSE boost OK.
[17:00:26] Writing local files
[17:00:26] Completed 0 out of 500000 steps  (0 percent)
[17:15:26] Timered checkpoint triggered.
[17:22:28] Writing local files
[17:22:29] Completed 5000 out of 500000 steps  (1 percent)
[17:37:28] Timered checkpoint triggered.
[17:44:28] Writing local files
[17:44:28] Completed 10000 out of 500000 steps  (2 percent)
[17:59:28] Timered checkpoint triggered.
[18:06:28] Writing local files
[18:06:28] Completed 15000 out of 500000 steps  (3 percent)
[18:21:29] Timered checkpoint triggered.
[18:28:29] Writing local files
[18:28:29] Completed 20000 out of 500000 steps  (4 percent)
[18:43:29] Timered checkpoint triggered.
[18:50:29] Writing local files
[18:50:29] Completed 25000 out of 500000 steps  (5 percent)
[19:05:29] Timered checkpoint triggered.
[19:12:29] Writing local files
[19:12:29] Completed 30000 out of 500000 steps  (6 percent)
[19:27:29] Timered checkpoint triggered.
[19:34:29] Writing local files
[19:34:29] Completed 35000 out of 500000 steps  (7 percent)
[19:49:29] Timered checkpoint triggered.
[19:56:29] Writing local files
[19:56:29] Completed 40000 out of 500000 steps  (8 percent)
[20:11:29] Timered checkpoint triggered.
[20:18:28] Writing local files
[20:18:29] Completed 45000 out of 500000 steps  (9 percent)
[20:33:28] Timered checkpoint triggered.
[20:40:26] Writing local files
[20:40:26] Completed 50000 out of 500000 steps  (10 percent)

aldamon, here's the proof of my claim :D
 
Xilikon said:
aldamon, here's the proof of my claim :D
Click to expand...

Maybe 2652 likes less cache? Go figure. I do have an E4300 sitting here. Hmmmmmmmm :p

E6600 @ 3.4 GHz (425 x 8 | 4-4-4-12):

[21:02:43] Project: 2652 (Run 0, Clone 199, Gen 16)
[21:02:43]
[21:02:43] Assembly optimizations on if available.
[21:02:43] Entering M.D.
[21:03:00] percent)
[21:03:01] - Starting from initial work packet
[21:03:01]
[21:03:01] Project: 26Entering M.D.
[21:03:01] ne 199, Gen 16)
[21:03:01]
[21:03:01] Entering M.D.
[21:03:07] Rejecting checkpoint
[21:03:07] OK.
[21:03:07] in: ProteinExtra SSE boost OK.
[21:03:07]
[21:03:07] Extra SSE boost OK.
[21:03:07] Writing local files
[21:03:08] Completed 0 out of 1000000 steps (0 percent)
[21:16:12] Writing local files
[21:16:12] Completed 10000 out of 1000000 steps (1 percent)
[21:29:16] Writing local files
[21:29:16] Completed 20000 out of 1000000 steps (2 percent)
[21:42:21] Writing local files
[21:42:21] Completed 30000 out of 1000000 steps (3 percent)
[21:55:25] Writing local files
[21:55:25] Completed 40000 out of 1000000 steps (4 percent)
[22:08:28] Writing local files
[22:08:28] Completed 50000 out of 1000000 steps (5 percent)

[17:55:39] Project: 2610 (Run 1, Clone 214, Gen 22)
[17:55:39]
[17:55:40] Entering M.D.
[17:55:46] Calling FAH init
[17:55:47] Writing local files
[17:55:47] ing from checkpoint)
[17:55:47] Read checkpoint
[17:55:47] Protein: Protein
[17:55:47] Writing local files
[17:55:48] Extra SSE boost OK.
[17:55:48] Writing local files
[17:55:49] Completed 0 out of 500000 steps (0 percent)
[18:06:42] Writing local files
[18:06:42] Completed 5000 out of 500000 steps (1 percent)
[18:17:36] Writing local files
[18:17:36] Completed 10000 out of 500000 steps (2 percent)
[18:28:29] Writing local files
[18:28:29] Completed 15000 out of 500000 steps (3 percent)
[18:39:21] Writing local files
[18:39:21] Completed 20000 out of 500000 steps (4 percent)
[18:50:19] Writing local files
[18:50:19] Completed 25000 out of 500000 steps (5 percent)

[12:28:00] Project: 2604 (Run 0, Clone 513, Gen 24)
[12:28:00]
[12:28:00] 04 (Run 0, Clone 513, Gen 24)
[12:28:00]
[12:28:00] Entering M.D.
[12:28:07] rotein in POPC
[12:28:07] Writing local files
[12:28:07] cal files
[12:28:07] Extra SSE boost OK.
[12:28:08] cal files
[12:28:08] Completed 0 out of 500000 steps (0 percent)
[12:39:34] Writing local files
[12:39:34] Completed 5000 out of 500000 steps (1 percent)
[12:51:00] Writing local files
[12:51:00] Completed 10000 out of 500000 steps (2 percent)
[13:02:26] Writing local files
[13:02:26] Completed 15000 out of 500000 steps (3 percent)
[13:13:52] Writing local files
[13:13:52] Completed 20000 out of 500000 steps (4 percent)
[13:25:17] Writing local files
[13:25:17] Completed 25000 out of 500000 steps (5 percent)

 
IBM Server - Dual Xeon 2.4Ghz 512K (older processors) HT is disabled in BIOS. 4Gb RAM, Windows Server 2003 R2.

Looks like it's going to get around 49 minutes per frame. Any ideas on anything I can do to get a little more speed? It's still going to net more PPD than the 186 pointers it's been running with the old standard client and at 49 minutes per frame it should make the deadlines, close but it should be fine.

[14:00:38] Project: 2651 (Run 0, Clone 592, Gen 5)
[14:00:54] Completed 0 out of 500000 steps (0 percent)
[14:49:46] Writing local files
[14:49:46] Completed 5000 out of 500000 steps (1 percent)
[15:38:41] Writing local files
[15:38:41] Completed 10000 out of 500000 steps (2 percent)
[16:27:39] Writing local files
[16:27:39] Completed 15000 out of 500000 steps (3 percent)
 
Project: 2610 (Run 0, Clone 825, Gen 14)
[20:31:16] Completed 130000 out of 500000 steps (26 percent)
[20:42:31] Writing local files
[20:42:31] Completed 135000 out of 500000 steps (27 percent)
[20:53:48] Writing local files
[20:53:48] Completed 140000 out of 500000 steps (28 percent)
[21:05:04] Writing local files
[21:05:04] Completed 145000 out of 500000 steps (29 percent)
[21:16:36] Writing local files
[21:16:36] Completed 150000 out of 500000 steps (30 percent)

Intel C2D E6320 @ 3.108 Ghz 2GB 4MB L2 Cache
 
E4300 @ 3150 MHz
Protein 2608 (0/455/32)
17:50/frame

Code: 
[00:04:03] Project: 2608 (Run 0, Clone 455, Gen 32)
[00:04:03] 
[00:04:06] Entering M.D.
[00:04:12] Calling FAH init
[00:04:13] Read topology
[00:04:13] files
[00:04:13] Completed 145000 out of 500000 steps  (29 percent)
[00:04:13] in
[00:04:13] Writing local files
[00:04:13] Completed 145000 out of 500000 steps  (29 percent)
[00:04:14] Extra SSE boost OK.
[00:22:02] Writing local files
[00:22:03] Completed 150000 out of 500000 steps  (30 percent)
[00:39:50] Writing local files
[00:39:50] Completed 155000 out of 500000 steps  (31 percent)
[00:57:37] Writing local files
[00:57:37] Completed 160000 out of 500000 steps  (32 percent)
[01:15:24] Writing local files
[01:15:25] Completed 165000 out of 500000 steps  (33 percent)
[01:33:12] Writing local files
[01:33:12] Completed 170000 out of 500000 steps  (34 percent)
[01:51:01] Writing local files
[01:51:01] Completed 175000 out of 500000 steps  (35 percent)
[02:08:49] Writing local files
[02:08:49] Completed 180000 out of 500000 steps  (36 percent)

Protein 2652 (0/143/25)
12:44/frame
Code: 
[05:41:02] Project: 2652 (Run 0, Clone 143, Gen 25)
[05:41:02] 
[05:41:03] Entering M.D.
[05:41:09] Calling FAH init
[05:41:09] Writing local files
[05:41:09] Completed 223989 out of 1000000 steps  (22 percent)
[05:41:09] Extra SSE boost OK.
[05:41:09] s
[05:41:09] Completed 223989 out of 1000000 steps  (22 percent)
[05:41:09] Extra SSE boost OK.
[05:48:49] Writing local files
[05:48:49] Completed 230000 out of 1000000 steps  (23 percent)
[06:01:33] Writing local files
[06:01:33] Completed 240000 out of 1000000 steps  (24 percent)
[06:14:18] Writing local files
[06:14:18] Completed 250000 out of 1000000 steps  (25 percent)
[06:27:04] Writing local files
[06:27:04] Completed 260000 out of 1000000 steps  (26 percent)
[06:39:48] Writing local files
[06:39:48] Completed 270000 out of 1000000 steps  (27 percent)
[06:52:31] Writing local files
[06:52:31] Completed 280000 out of 1000000 steps  (28 percent)
[07:05:16] Writing local files
[07:05:16] Completed 290000 out of 1000000 steps  (29 percent)
[07:18:01] Writing local files
[07:18:01] Completed 300000 out of 1000000 steps  (30 percent)
 
MixManSC said:
IBM Server - Dual Xeon 2.4Ghz 512K (older processors) HT is disabled in BIOS. 4Gb RAM, Windows Server 2003 R2.

Looks like it's going to get around 49 minutes per frame. Any ideas on anything I can do to get a little more speed? It's still going to net more PPD than the 186 pointers it's been running with the old standard client and at 49 minutes per frame it should make the deadlines, close but it should be fine.
Click to expand...
It will make the deadline if you have no down time. I run a dual AMD socket-A system that makes the deadline with smaller WUs, even if it takes the machine 52 min/frame.

The only things I can suggest is OC (if it's possible) and running a Linux distro instead of MS. Just changing OS will slightly improve the frame times from what others have noted in the past. It could make a difference of a couple of min/frame. Although a small improvement won't net you a significant increase in points, it will help you gain some leeway in case there is down time. Every little bit helps, also check to see if there are any unnecessary apps or services always running in the background consuming CPU cycles.
 
I am noticing a *very* large difference with 2MB versus 4MB core2 chips. Has anyone else noticed this? (at least on the 260X, 261X proteins)
 
E6420 @ 3.4 Ghz ( I dropped the speed/voltage in my sig down due to hot summer temps here)
FAHSpy is saying 12:34 per frame with 2015 PPD.

[22:47:37] Project: 2653 (Run 28, Clone 11, Gen 14)
[22:47:37]
[22:47:39] Entering M.D.
[22:47:45] Calling FAH init
[22:47:46] in POPC
[22:47:46] Writing local files
[22:47:46] Extra SSE boost OK.
[22:47:46] eckpoint
[22:47:46] Protein: Protein in POPC
[22:47:46] Writing local files
[22:47:47] Extra SSE boost OK.
[22:47:47] Writing local files
[22:47:47] Completed 0 out of 500000 steps (0 percent)
[23:00:19] Writing local files
[23:00:19] Completed 5000 out of 500000 steps (1 percent)
[23:12:54] Writing local files
[23:12:54] Completed 10000 out of 500000 steps (2 percent)
[23:25:27] Writing local files
[23:25:27] Completed 15000 out of 500000 steps (3 percent)
[23:37:59] Writing local files
[23:37:59] Completed 20000 out of 500000 steps (4 percent)
[23:50:38] Writing local files
[23:50:38] Completed 25000 out of 500000 steps (5 percent)
[00:03:06] Writing local files
[00:03:06] Completed 30000 out of 500000 steps (6 percent)
[00:15:37] Writing local files
[00:15:37] Completed 35000 out of 500000 steps (7 percent)
[00:28:54] Writing local files
[00:28:54] Completed 40000 out of 500000 steps (8 percent)
[00:42:02] Writing local files
[00:42:02] Completed 45000 out of 500000 steps (9 percent)
[00:55:03] Writing local files
[00:55:03] Completed 50000 out of 500000 steps (10 percent)

And just started:
11:25 per frame 1919 PPD
[20:09:34] Project: 2610 (Run 0, Clone 617, Gen 45)
[20:09:34]
[20:09:36] Entering M.D.
[20:09:43] Rejecting checkpoint
[20:09:44] Protein: Protein
[20:09:44] Writing local files
[20:09:45] Extra SSE boost OK.
[20:09:45] Writing local files
[20:09:45] Completed 0 out of 500000 steps (0 percent)
[20:21:07] Writing local files
[20:21:07] Completed 5000 out of 500000 steps (1 percent)
[20:32:30] Writing local files
[20:32:30] Completed 10000 out of 500000 steps (2 percent)
[20:43:52] Writing local files
[20:43:53] Completed 15000 out of 500000 steps (3 percent)
[20:55:15] Writing local files
[20:55:15] Completed 20000 out of 500000 steps (4 percent)
[21:06:37] Writing local files
[21:06:38] Completed 25000 out of 500000 steps (5 percent)
[21:17:59] Writing local files
[21:18:00] Completed 30000 out of 500000 steps (6 percent)
[21:29:22] Writing local files
[21:29:22] Completed 35000 out of 500000 steps (7 percent)
[21:40:45] Writing local files
[21:40:45] Completed 40000 out of 500000 steps (8 percent)
[21:52:16] Writing local files
[21:52:17] Completed 45000 out of 500000 steps (9 percent)
 
You must log in or register to reply here.
Back
Top