Project: 6013 (Run 0, Clone 50, Gen 140)

[AtwaterFS]

Made 4 or so pop-ups about Visual Basic FahCore_a3 errors.

I stopped the service when I realized it was sitting there doing nothing:

Code: Select all

[22:32:55] Project: 6013 (Run 0, Clone 50, Gen 140)
[22:32:55] 
[22:32:55] Assembly optimizations on if available.
[22:32:55] Entering M.D.
[22:33:19] Completed 0 out of 250000 steps  (0%)
[14:02:14] CoreStatus = 3 (3)
[14:02:14] Client-core communications error: ERROR 0x3
[14:02:14] Deleting current work unit & continuing...
[14:02:29] - Preparing to get new work unit...
[14:02:29] Cleaning up work directory
[14:02:29] + Attempting to get work packet
[14:02:29] Passkey found
[14:02:29] - Connecting to assignment server
[14:02:29] - Successful: assigned to (130.237.232.140).
[14:02:29] + News From Folding@Home: Welcome to Folding@Home
[14:02:30] Loaded queue successfully.
[14:02:37] + Closed connections
[14:02:42] 
[14:02:42] + Processing work unit
[14:02:42] Core required: FahCore_a3.exe
[14:02:42] Core found.
[14:02:42] Working on queue slot 04 [June 23 14:02:42 UTC]
[14:02:42] + Working ...
[14:02:42] 
[14:02:42] *------------------------------*
[14:02:42] Folding@Home Gromacs SMP Core
[14:02:42] Version 2.19 (Mar 12, 2010)
[14:02:42] 
[14:02:42] Preparing to commence simulation
[14:02:42] - Looking at optimizations...
[14:02:42] - Created dyn
[14:02:42] - Files status OK
[14:02:44] - Expanded 977970 -> 10427873 (decompressed 1066.2 percent)
[14:02:44] Called DecompressByteArray: compressed_data_size=977970 data_size=10427873, decompressed_data_size=10427873 diff=0
[14:02:44] - Digital signature verified
[14:02:44] 
[14:02:44] Project: 6013 (Run 0, Clone 50, Gen 140)
[14:02:44] 
[14:02:44] Assembly optimizations on if available.
[14:02:44] Entering M.D.
[14:03:08] Completed 0 out of 250000 steps  (0%)
[14:03:26] Service stop request received.

Folding@Home Client Shutdown.

[PantherX]

[22:33:19] Completed 0 out of 250000 steps (0%)
[14:02:14] CoreStatus = 3 (3)
Did you edit the FAHLog?

[jima13]

No reason it should take over an hour per frame...I've deleted it 3 times on two different quad systems.....should be pulled, IMO

[bruce]

This WU has successfully been completed by others. (By deleting it and letting it expire, you've caused extra copies to be distributed so it was completed twice, not once.)

[jima13]

This WU has successfully been completed by others. (By deleting it and letting it expire, you've caused extra copies to be distributed so it was completed twice, not once.)

To bad, because here's another that's about to get dumped>

Code: Select all

[00:58:48] Folding@Home Gromacs SMP Core
[00:58:48] Version 2.22 (Mar 12, 2010)
[00:58:48] 
[00:58:48] Preparing to commence simulation
[00:58:48] - Looking at optimizations...
[00:58:48] - Created dyn
[00:58:48] - Files status OK
[00:58:49] - Expanded 977970 -> 10427873 (decompressed 1066.2 percent)
[00:58:49] Called DecompressByteArray: compressed_data_size=977970 data_size=10427873, decompressed_data_size=10427873 diff=0
[00:58:49] - Digital signature verified
[00:58:49] 
[00:58:49] Project: 6013 (Run 0, Clone 50, Gen 140)
[00:58:49] 
[00:58:49] Assembly optimizations on if available.
[00:58:49] Entering M.D.
[00:59:14] Completed 0 out of 250000 steps  (0%)
[03:03:56] Completed 2500 out of 250000 steps  (1%)
[04:46:42] - Autosending finished units... [June 27 04:46:42 UTC]
[04:46:42] Trying to send all finished work units
[04:46:42] + No unsent completed units remaining.
[04:46:42] - Autosend completed
[04:57:46] Completed 5000 out of 250000 steps  (2%)
[07:00:46] Completed 7500 out of 250000 steps  (3%)
[09:03:25] Completed 10000 out of 250000 steps  (4%)
[10:46:48] - Autosending finished units... [June 27 10:46:48 UTC]
[10:46:48] Trying to send all finished work units
[10:46:48] + No unsent completed units remaining.
[10:46:48] - Autosend completed
[11:01:12] Completed 12500 out of 250000 steps  (5%)
[12:55:10] Completed 15000 out of 250000 steps  (6%)
[14:49:09] Completed 17500 out of 250000 steps  (7%)
[16:46:48] - Autosending finished units... [June 27 16:46:48 UTC]
[16:46:48] Trying to send all finished work units
[16:46:48] + No unsent completed units remaining.
[16:46:48] - Autosend completed
[16:52:08] Completed 20000 out of 250000 steps  (8%)

I see no point in keeping this sucking up juice when it won't complete in the time alloted

[codysluder]

Once a WU has been completed, there's no point it having a Mod report it as bad. It has already removed itself from future distribution just like if it's reported. The next gen has been produced and FAH has passed that point. There's no way that can remove a WU that has already been assigned to you.

[mdk777]

Yeah? Then why have I been assigned it twice?

ETA on a 6 core is 5 days. 23 hours

Clearly defective.

Code: Select all

Note: Please read the license agreement (Folding@home-Win32-x86.exe -license). F
urther
use of this software requires that you have read and accepted this agreement.

6 cores detected
If you see this twice, MPI is working
If you see this twice, MPI is working
[12:30:51] MPIexec present


--- Opening Log file [June 28 12:30:51 UTC]


# Windows SMP Console Edition #################################################
###############################################################################

                       Folding@Home Client Version 6.30

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01
Executable: C:\Program Files (x86)\Folding@Home Windows SMP Client V1.01\Folding
@home-Win32-x86.exe
Arguments: -smp

[12:30:51] - Ask before connecting: No
[12:30:51] - User name: mdk777 (Team 36837)
[12:30:51] - User ID: 12378AC4F1DE79F
[12:30:51] - Machine ID: 1
[12:30:51]
[12:30:52] Loaded queue successfully.
[12:30:52]
[12:30:52] + Processing work unit
[12:30:52] Core required: FahCore_a3.exe
[12:30:52] Core found.
[12:30:52] Working on queue slot 04 [June 28 12:30:52 UTC]
[12:30:52] + Working ...
[12:30:52]
[12:30:52] *------------------------------*
[12:30:52] Folding@Home Gromacs SMP Core
[12:30:52] Version 2.22 (Mar 12, 2010)
[12:30:52]
[12:30:52] Preparing to commence simulation
[12:30:52] - Ensuring status. Please wait.
[12:31:01] - Looking at optimizations...
[12:31:01] - Working with standard loops on this execution.
[12:31:01] - Previous termination of core was improper.
[12:31:01] - Files status OK
[12:31:02] - Expanded 977970 -> 10427873 (decompressed 1066.2 percent)
[12:31:02] Called DecompressByteArray: compressed_data_size=977970 data_size=104
27873, decompressed_data_size=10427873 diff=0
[12:31:02] - Digital signature verified
[12:31:02]
[12:31:02] Project: 6013 (Run 0, Clone 50, Gen 140)
[12:31:02]
[12:31:02] Entering M.D.
[12:31:29] Completed 0 out of 250000 steps  (0%)

[AtwaterFS]

Sorry for the tardy reply but nope I didnt edit Fahlog - str8 c&p, I guess that's just how messed up Fah was...

I miss the days of SMP2, back when it "just worked"... I know some hard work went into SMP3, but it really seems like its got a ways to go in terms of stability.

[bruce]

This WU has now been completed by two people plus one person got 0 points for EUE report.

Project 6013, Run 0, Clone 50, Gen 140
Days taken to complete WU: 8.24
Days taken to complete WU: 0.03 (zero points)
Days taken to complete WU: 6.04

I don't know what hardware those people have.

There's no doubt that this project takes longer that other A3 projects, but the deadlines are longer, too. Both the points and deadlines are based on the time taken on an i5 (or a comparable i7 with HT disabled) and they should be consistent with other projects.

[mdk777]

Bruce,

I really don't understand your post. I really am not trying to pick a fight, I just really can't follow your logic.

6 days is what I reported the ETA was. So your results are consistent with my ETA.
However, The deadline is 3 days.(or was it 4?)

I just downloaded 6012 (r2,c189,g140)

On the same hardware, ETA is 6hours and 15 minutes.deadline 3 days.

So the WU runs 24X slower and you are indicating that it should be consistent with other projects?

Your own results show that it takes anywhere from 6 to 8 days. This is not a difference of 25-30%, this is a difference of 24 TIMES longer.

"Should be" and are have obviously diverged.

[jima13]

The deadline on these is 3-6 days, but this is moot since 6013 is no longer listed on the project page.

[muziqaz]

The deadline on these is 3-6 days, but this is moot since 6013 is no longer listed on the project page.

since when? because mdk777s log clearly shows, that he got 6013 WU yesterday (June 28th)

[mdk777]

locked this thread with no answer:

viewtopic.php?f=58&t=15076&start=0

Perhaps they have just now been pulled it and aren't going to admit it?

He was running a

REAL 8-core Xeon (5430@3100MHz)

so it is not a compiled for Intel rather than AMD issue.

Just a horrible WU and they have concluded that they would rather stonewall rather than give "complainers the satisfaction" ?

I can't fathom the motivation.

[bruce]

locked this thread with no answer:

viewtopic.php?f=58&t=15076&start=0

Perhaps they have just now been pulled it and aren't going to admit it?

He was running a

REAL 8-core Xeon (5430@3100MHz)

so it is not a compiled for Intel rather than AMD issue.

Just a horrible WU and they have concluded that they would rather stonewall rather than give "complainers the satisfaction" ?

I can't fathom the motivation.

Well, if you are talking about Project: 6013 (Run 0, Clone 160, Gen 98) it's an entirely different with different characteristics. Your link points to a discussion of that different WU. Yes, there was a locked topic but I don't know how it got that way but it has now been merged with another topic regarding the same WU. Yes, it was bad and I just reported it.

This topic is talking about Project: 6013 (Run 0, Clone 50, Gen 140). There's no need to report this because it has been successfully complete by two people (one would be enough) and isn't bad. When a WU is completed, it "pulls" itself out of distribution and the next Gen goes into distribution. (Gen 141 is now being worked on.)

There's no conspiracy to pull WUs without admitting it or stonewalling going on, no matter how much you'd like to believe there is. Maybe it's your motivation that's unfathomable.

(Oh, I'm being too direct. I must be getting tired. I've been at this keyboard for about 16 hours. I apologize. Please don't flame me -- but Mods/Admins have feelings, too.)

[mdk777]

And yet you post the results of the discussion here:

viewtopic.php?f=19&t=14957#p149778

No need to thank the dedicated folders who worked tirelessly to try to bring this problem to PG attention.

PS, I'm not trying to take any credit. I am talking about the people like Wrish, Mr.Nosmo and others who have written detailed posts.
And of course these people:

viewtopic.php?f=19&t=14997&start=0

Why hasn't someone from the PG responded to the above thread and thanked these people?

My attitude may be overly negative, but few people have the time or energy to research threads like this.
They get a work unit that runs 24x slower and they report it as defective.
Being told that it is not defective goes against all reason; have spent the time to try and help the project, and they are told they just don't understand.
Finally, after exhausting all logic, they loose their minds like CSM.

Call me a negative complainer all you want, but this system of telling people that "everything is benched correctly, learn to live with it", is not working.

PSS. While it may certainly come across as a personal attack, I mean it as an exposure of the system. (not you Bruce)
As I mentioned in previous posts, I see you as being put in the position of trying to defend the indefensible.

The attitude of the PG on this issue needs to change, and not that of the donors, mods/admins.