S41.xx Observation Thread

ExtraTerrestrial Apes
ExtraTerrestria...
Joined: 10 Nov 04
Posts: 770
Credit: 362,469,641
RAC: 703,622

RE: Machine ist running

Message 30040 in response to message 30038

Quote:
Machine ist running 27/7

Damn, if only I could run my machine like this! ;)
And a bit more serious: it's better not to post mail adresses uncoded bc of spam.

MrS

Scanning for our furry friends since Jan 2002

Akos Fekete
Akos Fekete
Joined: 13 Nov 05
Posts: 561
Credit: 4,527,270
RAC: 0

RE: RE: Machine ist

Message 30041 in response to message 30040

Quote:
Quote:
Machine ist running 27/7
Damn, if only I could run my machine like this! ;)

Oh, yes. A time-extender would be good to me too...

comment: 27/7 means 27 hours each day on whole week... ;-)

Barrie
Barrie
Joined: 23 Mar 05
Posts: 219
Credit: 21,449,106
RAC: 0

RE: RE: RE: Machine ist

Message 30042 in response to message 30041

Quote:
Quote:
Quote:
Machine ist running 27/7
Damn, if only I could run my machine like this! ;)

Oh, yes. A time-extender would be good to me too...

comment: 27/7 means 27 hours each day on whole week... ;-)

It must be the Xeon with the Tardis core. The HT means it runs in HyperTime.

Dead men don't get the baby washed. HTH

Steve Cressman
Steve Cressman
Joined: 9 Feb 05
Posts: 104
Credit: 139,654
RAC: 0

My 2500+ is overclocked and

My 2500+ is overclocked and not getting errors with S41.06.
FSB is supposed to be at 166MHZ on these but am running at 200MHz.
I have noticed that this version does slightly increase temps over other versions, most likely because of the overlapping technique.

98SE XP2500+ @ 2.1 GHz Boinc v5.8.8

Mike Hewson
Mike Hewson
Moderator
Joined: 1 Dec 05
Posts: 6,233
Credit: 137,634,253
RAC: 17,519

RE: RE: RE: RE: Machi

Message 30044 in response to message 30042

Quote:
Quote:
Quote:
Quote:
Machine ist running 27/7
Damn, if only I could run my machine like this! ;)

Oh, yes. A time-extender would be good to me too...

comment: 27/7 means 27 hours each day on whole week... ;-)

It must be the Xeon with the Tardis core. The HT means it runs in HyperTime.


Quite right! But in the fine print of the license agreement ( which no-one ever seems to read - download it from here ) you'll note that it also delivers fewer weeks per year. However in the BIOS there is an option to restore that to the usual 52, but alas there is a shorter lifetime warranty on that..... :-(
Cheers, Mike.

I have made this letter longer than usual because I lack the time to make it shorter. Blaise Pascal

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5,513
Credit: 68,493,674,368
RAC: 58,876,438

RE: I'll report what

Message 30045 in response to message 30035

Quote:

I'll report what happens with the next one when it gets to run. There are two of these and they are separated so if the first fails again I'll have a further opportunity to go even lower with FSB.

Well the next result from the file z1_1269.0 has now been crunched and yet again it immediately errored out with the unhandled exception with the speed set @ 1800MHz (FSB = 112.5). I have one more of these results left and no further ones have been downloaded overnight.

For this final one which will start in less than two hours from now, I'll set the FSB @ 106.3 so the CPU will be @ 1700MHz - hardly an overclock at all when the machine processes all other results @ 1960MHz with no problems. There have been about 40 results from z1_1269.0 so I'm sorry for any disruptions caused to others also processing this same file who have had the normal credit granting cycle disrupted by these errors.

I'm happy to see the end of z1_1269.0 (hopefully) as I can now put the machine back to 1960MHz and let her fly through all the other results. The combination of S41.07 and approximately a 400MHz overclock means that these $USD70 Government castoffs can process the long results in just over 1.25hrs. With the stock app at stock speed (1600MHz) the time was close to 10 hours so the speedup factor is close to 8. Truly amazing - and all thanks to Akosf!!!

Cheers,

Cheers,
Gary.

Crun-chi
Crun-chi
Joined: 24 Feb 05
Posts: 7
Credit: 47,934
RAC: 0

Until few days, my Pentium IV

Until few days, my Pentium IV gives all valid results with U41.04, a it was fastest client I ever had.
When Boinc downloaded grid z1.118 I got pattern of valid and invalid results in order
----------------
1 wu valid
1 wu invalid
1 wu valid
1 wu invalid
1 wu ivvalid
1 wu valid
1 wu invalid
1 wu valid
1 wu valid
1 wu invalid
1 wu valid
1 wu invalid
1 wu invalid
----------------
repeat

I try s39L, also S41.07 and U41.04

Computer works fine in any other way. So I dont think that OC is causing a problem.
Invalid results are procceses in only one second and give error you all have here:
09/05/2006 04:21:23|Einstein@Home|Unrecoverable error for result z1_1118.0__80_S4R2a_0 ( - exit code -1073741819 (0xc0000005))

So it is posibble that z1_1118.0 is not valid?

Akos Fekete
Akos Fekete
Joined: 13 Nov 05
Posts: 561
Credit: 4,527,270
RAC: 0

RE: So it is posibble that

Message 30047 in response to message 30046

Quote:
So it is posibble that z1_1118.0 is not valid?

I don't think.

Here is one of your WUs: z1_1118.0__128_S4R2a

There are your invalid and three valid results.

Applications:
invalid: U41.04
valid: 4.37, S40.12, U41.04

So, U41.04 did good result on an other computer with the same WU.

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5,513
Credit: 68,493,674,368
RAC: 58,876,438

RE: For this final one

Message 30048 in response to message 30045

Quote:

For this final one which will start in less than two hours from now, I'll set the FSB @ 106.3 so the CPU will be @ 1700MHz

The final one from z1_1269.0 has now been processed with the FSB @ 106.3 (1700MHz). It too has immediately errored out with the unhandled exception error message. The machine is now up at 1960MHz and has already finished two results from r1_1221.5 successfully. These have been reported and have also validated so the machine has no apparent problems at 1960MHz so it seems. On searching back through the logs, I have seen the log entry where z1_1269.0 has been deleted so I wont be getting any more of those.

As I've said before, I have around 20 of this particular brand and configuration, running at much the same level of overclock, and this is the only example I've noticed of any problems. Of course, with this group alone producing around 350 results per day, there is no way I'm going to see even a small fraction of those results so it would take something drastic (like 40 rapidly errored out results) to even be noticed. It will be interesting to see if any more similar problems arise.

Cheers,

Cheers,
Gary.

Crun-chi
Crun-chi
Joined: 24 Feb 05
Posts: 7
Credit: 47,934
RAC: 0

Akosf if z1_1118.0 is valid ,

Akosf if z1_1118.0 is valid , and my computer works well, so where is problem. Why computer runs 24hours per day, at least 6 days non-stop with no errors. After z1_1118.0 is downloaded, problems start. Maybe combination od FSB, with your clinet and z1_1118.0 is very sensitive?

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.