Welcome to MilkyWay@home

Computation errors


Advanced search

Message boards : Number crunching : Computation errors
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · Next

AuthorMessage
iancantwell

Send message
Joined: 28 Jul 13
Posts: 8
Credit: 2,398,088
RAC: 0
2 million credit badge6 year member badge
Message 59629 - Posted: 20 Aug 2013, 15:36:41 UTC - in response to Message 59539.  

I have just seen this message, I was wondering why they were still sitting on my system. I can let them finish the computation but one was due on 14th and other on 18th August, maybe they have been re-allocated since? No point doing them if they are being done elsewhere.
Please advise on this and if they have been re-allocated do I just abort them and hope I get new tasks or what should I do?
I assume the reason I am not getting new WU is because they are unfinished.
As a general note it would help to have reasonably accurate completion times from the beginning as this would avoid these kind of issues and misunderstandings arising
Regards
Ian Cantwell
ID: 59629 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profilemikey
Avatar

Send message
Joined: 8 May 09
Posts: 2273
Credit: 386,773,009
RAC: 349,409
300 million credit badge10 year member badgeextraordinary contributions badge
Message 59632 - Posted: 21 Aug 2013, 12:37:58 UTC - in response to Message 59629.  

I have just seen this message, I was wondering why they were still sitting on my system. I can let them finish the computation but one was due on 14th and other on 18th August, maybe they have been re-allocated since? No point doing them if they are being done elsewhere.
Please advise on this and if they have been re-allocated do I just abort them and hope I get new tasks or what should I do?
I assume the reason I am not getting new WU is because they are unfinished.
As a general note it would help to have reasonably accurate completion times from the beginning as this would avoid these kind of issues and misunderstandings arising
Regards
Ian Cantwell


If I take a look at one of your tasks I see this:
minimum quorum 2
initial replication 2
max # of error/total/success tasks 3, 9, 6

What that means is: minimum quorum 2 means each unit is sent to 2 different pc's initially. Max # of error/etc means each work unit can have 3 errors before the project tries to crunch it itself.

When a unit is sent to a user from most projects it is sent to more then one person to ensure each of our pc's is running accurately. Mine could be doing too many others things and throwing in too many errors to help with the math needed to crunch units effectively.

As for accurate completion times that is not possible with the way current units are being created. Each unit has multiple outcomes based on what is found as the crunching progresses thru. So the finish time is based on your past history of finishing units, so the more units you finish the more accurate the completion times will be. Of course since each unit could take longer, or shorter, then the one before it depending on how it is made, your completion times will vary a little bit.

As for units that are past there due date, just abort them, they will be resent to someone else to crunch. It is fairly normal for a project to have a bunch of these on a daily basis for any of a number of reasons. If it happens on a regular basis for you though you may wish to lower your cache a little bit though, just so YOU aren't a cause of alot of the resends. Don't get attached to an individual unit is what I am saying, in the big picture it means very little. I have been crunching for a long time and have had many pc's just crash, I used recycled hardware, and have many units resent to other pc's over the years.
ID: 59632 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Phil

Send message
Joined: 3 Dec 12
Posts: 6
Credit: 372,276
RAC: 0
100 thousand credit badge7 year member badge
Message 59636 - Posted: 21 Aug 2013, 23:31:22 UTC - in response to Message 59620.  

I have 3 computation errors on Milkyway@home N-body simulation 1.63 tasks.

I had 5 errors this morning when I turned on my computer. Can anyone tell me what's wrong and what I can do to fix it?

Two more computation errors today.
ID: 59636 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profilemikey
Avatar

Send message
Joined: 8 May 09
Posts: 2273
Credit: 386,773,009
RAC: 349,409
300 million credit badge10 year member badgeextraordinary contributions badge
Message 59637 - Posted: 22 Aug 2013, 10:43:25 UTC - in response to Message 59636.  

I have 3 computation errors on Milkyway@home N-body simulation 1.63 tasks.

I had 5 errors this morning when I turned on my computer. Can anyone tell me what's wrong and what I can do to fix it?

Two more computation errors today.


You are running the 1.36 tasks not the 1.63 tasks and you might want to check out the News section as they talk about problems with those tasks in there. This means it may not be YOU but the UNITS instead, meaning the only choice is to crunch thru them or crunch elsewhere. Check the thread out first before you do anything though.
ID: 59637 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Guilherme Tomazini

Send message
Joined: 15 Jul 11
Posts: 9
Credit: 11,533,046
RAC: 0
10 million credit badge8 year member badge
Message 59642 - Posted: 23 Aug 2013, 2:59:12 UTC

I'm getting 100% computation error on Win 7 64, HD 6950 lastest drivers.
ID: 59642 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
iancantwell

Send message
Joined: 28 Jul 13
Posts: 8
Credit: 2,398,088
RAC: 0
2 million credit badge6 year member badge
Message 59643 - Posted: 23 Aug 2013, 7:06:25 UTC - in response to Message 59632.  

Many thanks for the reply, it's good to get an objective opinion as to what was happening. It was only those two units (out of six) that caused problems so it is not clear whether it as due to a failure on my side or in the original set up prior to allocation.
I followed your advice and aborted them and they were transmogrified to the cloud of dark matter that I'm sure is responsible for dead units. I have since received six more units with timings of 1/3 hours each with one exception that took one minute and four seconds to complete! So far no issues
I was not aware that completion times were based on previous work completion rates. My experience is that they are reasonably accurate give or take 10% over runs or under runs. I did however get one with a predicted 2,212 hour completion estimate and was given 10 days to do it in. As this was an obvious mistake on the part of the research team and their quality control I aborted that one too.
From trial and error I now focus on projects who have shorter units time-wise, interact with and respect their volunteers, and have reasonably trouble free projects which I choose from analyzing their message boards for complaints and responses
As a newbie to this I wish there was a "Computing for Dummies" hand book out there that answered obvious questions that everybody seems to take for granted. I also think that there should e some d=facility by where people with WU issues should be able to contact the project moderators directly and not have to rely on the vagaries of message boards
Again Thanks
Ian
ID: 59643 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profilemikey
Avatar

Send message
Joined: 8 May 09
Posts: 2273
Credit: 386,773,009
RAC: 349,409
300 million credit badge10 year member badgeextraordinary contributions badge
Message 59645 - Posted: 23 Aug 2013, 10:37:33 UTC - in response to Message 59643.  
Last modified: 23 Aug 2013, 10:45:47 UTC

Many thanks for the reply, it's good to get an objective opinion as to what was happening. It was only those two units (out of six) that caused problems so it is not clear whether it as due to a failure on my side or in the original set up prior to allocation.
I followed your advice and aborted them and they were transmogrified to the cloud of dark matter that I'm sure is responsible for dead units. I have since received six more units with timings of 1/3 hours each with one exception that took one minute and four seconds to complete! So far no issues
I was not aware that completion times were based on previous work completion rates. My experience is that they are reasonably accurate give or take 10% over runs or under runs. I did however get one with a predicted 2,212 hour completion estimate and was given 10 days to do it in. As this was an obvious mistake on the part of the research team and their quality control I aborted that one too.
From trial and error I now focus on projects who have shorter units time-wise, interact with and respect their volunteers, and have reasonably trouble free projects which I choose from analyzing their message boards for complaints and responses
As a newbie to this I wish there was a "Computing for Dummies" hand book out there that answered obvious questions that everybody seems to take for granted. I also think that there should e some d=facility by where people with WU issues should be able to contact the project moderators directly and not have to rely on the vagaries of message boards
Again Thanks
Ian


We have asked for this everywhere but most projects are run by Scientists not computer people. The Scientists contract, beg, borrow or steal the programmers to write the programs who then normally have very little to do with the running of the project. Not ALL are like that, but some are. Those that are couldn't help us if they wanted to, as they have no clue how to solve a pc problem. They are Scientists working on a problem in their little corner of the World and pc's just help others help them. They plug in the numbers and poof out come the results. They can tell you at what temperature water boils depending on how much pressure it is under to the nth degree, but pc's just isn't where there focus is.

That is why we users asked for an got these help groups and some of continue to drop in to help others. We ALL started where you are, but thru some trial and error, much help from other crunchers, and perseverance we are where we are today.

So ask away and we will try and help you as best we can.

As for the units you aborted they went back into the Projects Server's cache of available units to send back out again. With the total number of users it is VERY unlikely you personally will ever see it again, but someone else will. They might have the same problems, if it is a unit problem they WILL, but if it was a pc problem then they may crunch it just fine. It happens on a regular basis and some of the ones you have could be from others that aborted them. There IS a way to tell though....if you look at the actual name of the work unit you will see at the very end a dash and then a number. One unit I saw looks like this: Name de_separation_82_DR8_rev_3_2_1372784654_15477237_2

The dash 2 means it is the 3rd time the unit has been sent out for someone to crunch. Remember this is computer numbering and zero is the 1st number, then 1, then 2, etc. It doesn't mean anything to us crunchers though other then that. It is just a unit we are assigned and try to crunch and send back. There are often tens or even hundreds of thousands of units in a batch looking for the answer to the current problem.
ID: 59645 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Zydor
Avatar

Send message
Joined: 24 Feb 09
Posts: 620
Credit: 100,587,625
RAC: 0
100 million credit badge10 year member badgeextraordinary contributions badge
Message 59647 - Posted: 23 Aug 2013, 14:28:22 UTC
Last modified: 23 Aug 2013, 14:41:14 UTC

I did however get one with a predicted 2,212 hour completion estimate and was given 10 days to do it in. As this was an obvious mistake on the part of the research team and their quality control I aborted that one too.


Research Team would know nothing about it ...

This one is an Old BOINC chestnut as old as the hills.

That estimate is a BOINC calculation (not the Research Team - they know nothing about this) trying to estimate how long a WU will take. BOINC often has very little to go on ... ie what kind of CPU and/or GPU used, its speed setting, memory on board, cache used, what the program does, what the program is calculating, how long it actually takes to calculate the various parts of the program, how long it actually takes is completely unknown to BOINC - it has to try and guess the result from what it knows, which initially is very little to zero. Its impossible for BOINC to know how long a Project WU will take to run, right out the starting gate

What BOINC will do is basically take a wild stab at it, and over the subsequent WUs it "learns" from your previous results. Overtime the estimates get better and better, usually from one to a few days worth depending on the WU.

When you switch WU types, the whole process starts again, it learns from experience of WUs crunched. There is literally no other way to do it as the variables are far too many.

The best way to get round this is crunch the WU for around 30-50% of its length, then compute in your mind the time taken to get to that percentage - eg 9secs to get to 33% done means 18secs to go - whatever the "time remaining" counter says. As a few more are completed, estimates for time remaining get better through its experience of past WUs.

Until it learns about the WU on your individual PC, that's the only way its possible to do the calculation. You get round that by taking the time completed shown and the percentage completed shown, extrapolating that to 100%.

Usually after a few dozen / few hours worth of WUs from a Project are crunched, BOINC has learnt enough to give good estimates - until then do the calculation in your mind I outlined above.
ID: 59647 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profilemikey
Avatar

Send message
Joined: 8 May 09
Posts: 2273
Credit: 386,773,009
RAC: 349,409
300 million credit badge10 year member badgeextraordinary contributions badge
Message 59654 - Posted: 24 Aug 2013, 9:45:25 UTC

@iancantwell One other thing that can affect the time is takes is whether people crunch 24/7 or only some portion of that. Computers at work usually only crunch at night, some people work from home and that too can slow the crunching speed down as compared to when not using the pc. Some people game, some don't, lots of little things affect the crunching speed and as Zydor said Boinc will try to figure it out over time. Obviously the bigger the data sample the better the guesstimations.
ID: 59654 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
iancantwell

Send message
Joined: 28 Jul 13
Posts: 8
Credit: 2,398,088
RAC: 0
2 million credit badge6 year member badge
Message 59689 - Posted: 27 Aug 2013, 4:31:05 UTC - in response to Message 59645.  

Thanks for the updates. I obviously was making incorrect assumptions about how it is all organized. I can understand the issues better now. I had not known that it was BOINC (I thought they were only facilitators) who assessed how long WU are estimated to take and not the research project managers who I would have thought would have known.

I can see that there is trial and error here based on previous unit histories and that most must be computer generated. I find that the majority are reasonably accurate. Still I would have thought that it would be easy to sort out the contradiction between estimated time and completion deadline.

So I have learnt that I must be flexible, patient and not feel guilty if I have to be ruthless with a unit that is not working out on my system. That extra info about how to tell how often a unit has been attempted is a useful extra in assessing any issues

With regards to the extra long estimates I had tried it for half an hour and saw that the ratio was second to second but the percentage estimated a shorter time scale, the latter I had ignored since there is often differences between it and time remaining. Next time I will take the advice and run it for 30% for a more accurate overview.

Usage on my laptop is basic, nothing I do takes up much power. I have worked out how to crunch 24/7 which I do except when mobile. I only use two CPU, which appears to be the default setting so two units are crunched at any one time. I have all projects at 100 so they all get equal share - I did try to increase this but it did not seem to make any difference.

Judging by the disc space stats I probably could do 3/4 CPU but have not yet worked out how I can do this simply, I don't want to get into manipulating configuration files which I have no experience with and I would be worried that I would make some simple mistake and everything would go wrong. Still I would like more options so that I could run longer units - my focus at the moment is anything under six hours as they are the easiest to manage. Any advice here welcome.

One final question. Is there any particular project you would care to recommend? I currently run Pogs, Malaria control, Poem, World Community Grid, Rosetta, Milky Way, boincsimap, correlizer & mind modelling. and have suspended Lattice Project, RNA World and Cosmology since I was having problems with them.

Also any that have cool screen saver animations. Rosetta is excellent, World Community and RNA average, and mind modelling, with its simulation of a brain in a microwave, awful.
Thanks again to community members who have taken the time to advise
Regards
Ian
ID: 59689 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
iancantwell

Send message
Joined: 28 Jul 13
Posts: 8
Credit: 2,398,088
RAC: 0
2 million credit badge6 year member badge
Message 59690 - Posted: 27 Aug 2013, 4:32:22 UTC - in response to Message 59647.  

Thanks for the reply, I have made one answer above
ID: 59690 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
iancantwell

Send message
Joined: 28 Jul 13
Posts: 8
Credit: 2,398,088
RAC: 0
2 million credit badge6 year member badge
Message 59692 - Posted: 27 Aug 2013, 8:04:26 UTC - in response to Message 59690.  

There will be no shortage of crunching to do when this comes on stream http://www.bbc.co.uk/news/science-environment-23779294
ID: 59692 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profilemikey
Avatar

Send message
Joined: 8 May 09
Posts: 2273
Credit: 386,773,009
RAC: 349,409
300 million credit badge10 year member badgeextraordinary contributions badge
Message 59696 - Posted: 27 Aug 2013, 11:13:18 UTC - in response to Message 59689.  

Thanks for the updates. I obviously was making incorrect assumptions about how it is all organized. I can understand the issues better now. I had not known that it was BOINC (I thought they were only facilitators) who assessed how long WU are estimated to take and not the research project managers who I would have thought would have known.


Yes kinda, the project builds its own workunits that we then crunch, and in that process they guesstimate how long a unit will take. But since they don't have the pc's to run the units on, they may have a couple but together we users have hundreds of thousands, each slightly different from the other, the Boinc software manages the process of using the algorithms to make it all work. For instance you laptop is a dual core running at 2.2ghz with 4gb of ram in it. My laptop is an I7 with 4 cores but thinks it has 8 cores thru hyper-threading, it has 8gb of ram and is running at 2.10ghz. Since ram is a large part of how fast a unit runs we should expect my pc to run the units about the same speed as yours as I am only using 4 of the cores, giving me an average of 2gb of ram per core.

I can see that there is trial and error here based on previous unit histories and that most must be computer generated. I find that the majority are reasonably accurate. Still I would have thought that it would be easy to sort out the contradiction between estimated time and completion deadline.

So I have learnt that I must be flexible, patient and not feel guilty if I have to be ruthless with a unit that is not working out on my system. That extra info about how to tell how often a unit has been attempted is a useful extra in assessing any issues

With regards to the extra long estimates I had tried it for half an hour and saw that the ratio was second to second but the percentage estimated a shorter time scale, the latter I had ignored since there is often differences between it and time remaining. Next time I will take the advice and run it for 30% for a more accurate overview.

Usage on my laptop is basic, nothing I do takes up much power. I have worked out how to crunch 24/7 which I do except when mobile. I only use two CPU, which appears to be the default setting so two units are crunched at any one time. I have all projects at 100 so they all get equal share - I did try to increase this but it did not seem to make any difference.


Once they are all the same it makes no difference if they are are 10, 20, 50, 100, or 1000. The same is the same. If you want one project to do less then another project just lower one projects numbers lower then the other, ie leave one project at 100 and change another to 50 and the first will crunch twice as much work as the second.

Judging by the disc space stats I probably could do 3/4 CPU but have not yet worked out how I can do this simply, I don't want to get into manipulating configuration files which I have no experience with and I would be worried that I would make some simple mistake and everything would go wrong. Still I would like more options so that I could run longer units - my focus at the moment is anything under six hours as they are the easiest to manage. Any advice here welcome.


When I look at your pc details I see this:
Number of processors 2
That means that NO you cannot run more then two units at a time using the cpu. If you had more cpu cores then yes you could.

One final question. Is there any particular project you would care to recommend? I currently run Pogs, Malaria control, Poem, World Community Grid, Rosetta, Milky Way, boincsimap, correlizer & mind modelling. and have suspended Lattice Project, RNA World and Cosmology since I was having problems with them.


This is pretty normal, most of us have tried out projects that we left later on, there is no stigma in trying and not liking something as much as something else. As for which project to run we each must make that decision based on our own needs and wants. Some people only run socially helpful projects, some people run anything and everything. Here is a list of the Distributed Computing active projects, but I don't think it has been updated in a while. The Boinc ones are noted:
http://www.distributedcomputing.info/projects.html

Also any that have cool screen saver animations. Rosetta is excellent, World Community and RNA average, and mind modelling, with its simulation of a brain in a microwave, awful.
Thanks again to community members who have taken the time to advise
Regards
Ian


I personally turn all screen savers off anyway as if the pc is drawing fancy pictures it is not crunching, and I like crunching better. If you like looking for aliens Seti has a cool one, the MalariaControl one to me is funky as it is from the perspective of the mosquito flying around though. I have not tried any other so sorry can't help there.
ID: 59696 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Phil

Send message
Joined: 3 Dec 12
Posts: 6
Credit: 372,276
RAC: 0
100 thousand credit badge7 year member badge
Message 59724 - Posted: 28 Aug 2013, 21:46:54 UTC - in response to Message 59637.  

I have 3 computation errors on Milkyway@home N-body simulation 1.63 tasks.

I had 5 errors this morning when I turned on my computer. Can anyone tell me what's wrong and what I can do to fix it?

Two more computation errors today.


You are running the 1.36 tasks not the 1.63 tasks and you might want to check out the News section as they talk about problems with those tasks in there. This means it may not be YOU but the UNITS instead, meaning the only choice is to crunch thru them or crunch elsewhere. Check the thread out first before you do anything though.


Pardon my typo It's 1.36. I looked in the news thread and all they talk about is validation. Nothing there about computation errors.
I will just abort those until someone tells me otherwise.
ID: 59724 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profilemikey
Avatar

Send message
Joined: 8 May 09
Posts: 2273
Credit: 386,773,009
RAC: 349,409
300 million credit badge10 year member badgeextraordinary contributions badge
Message 59731 - Posted: 29 Aug 2013, 11:19:09 UTC - in response to Message 59724.  

I have 3 computation errors on Milkyway@home N-body simulation 1.63 tasks.

I had 5 errors this morning when I turned on my computer. Can anyone tell me what's wrong and what I can do to fix it?

Two more computation errors today.


You are running the 1.36 tasks not the 1.63 tasks and you might want to check out the News section as they talk about problems with those tasks in there. This means it may not be YOU but the UNITS instead, meaning the only choice is to crunch thru them or crunch elsewhere. Check the thread out first before you do anything though.


Pardon my typo It's 1.36. I looked in the news thread and all they talk about is validation. Nothing there about computation errors.
I will just abort those until someone tells me otherwise.


Assuming they are unit problems someone will either have to dump them all or people will have to crunch thru them finding the bad ones before they make another batch. If some units are good it may be helpful to keep going, depending on your error rate. 2 units out of 30 per day is not too bad, 2 units out of 3 would make me move on and come back later.
ID: 59731 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Zydor
Avatar

Send message
Joined: 24 Feb 09
Posts: 620
Credit: 100,587,625
RAC: 0
100 million credit badge10 year member badgeextraordinary contributions badge
Message 59763 - Posted: 30 Aug 2013, 17:01:09 UTC

.... run it for 30% for a more accurate overview.


The second "trap" is to get sucked into too strong a competitive edge. The latter has two consequences

- Hammering hardware far too strongly, laptops are particularly vulnerable to that. Burning out a Laptop itsnt worth it .... the world will move on and you get left with the Bill :)

- Take part in your Team Events and contribute, BOINC was not founded on speed records, it was Founded on many PCs contributing a little. Its the sum of the whole that matters.

As you move forward, make sure you use software tools to monitor the Laptop performance, its critical with basic hardware else you can strain it far too much. Your Prime indicator for that is Heat. Do NOT take any chances with heat. Keep it well ventilated, learn how to use the fans correctly. Monitor it carefully. Once you have settled in you'll be fine, its no drama, but it can be if you ignore the basics. Burning a laptop is not worth it.

There are a number of heat monitoring tools around, each with their own "thing", you'll come across them over time, together with the other useful tools. For now take safety first, and a basic heat monitoring tool - having safeguarded the laptop that way, you can move forward and investigate what BOINC has to offer .... and there's a lot.

There are many "Tools" - for now safety first, load GPU Temp and Core Temp, that'll watch your back until you get chance to settle in more:

http://www.gputemp.com/

http://www.techpowerup.com/realtemp/

If you don't know the safe temperatures, Post for help, you'll soon get an answer from someone - and doubtless more and different types of tools that do various things. For now, stay safe and use GPU Temp & RealTemp (or another if you found one) .... but DO use one for CPU and GPU.

Last for now - make sure you have security tools loaded - there are nasty folk around. The basic tools from Windows are free - not perfect to say the least, but they do an effective job for the threat level you face. Use them ....
ID: 59763 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
OldArmyGuy

Send message
Joined: 18 Sep 12
Posts: 1
Credit: 3,068,740
RAC: 0
3 million credit badge7 year member badge
Message 59773 - Posted: 31 Aug 2013, 21:29:58 UTC

I'm getting all sorts of computation errors on "de_modfit_18_3s_126testwrap_2_13727846..." Is there a "fix" for this or do I just ignore and keep on crunching?

There is no problem with de_separation_**_DR_******.

System:
Dell E521
AMD Athlon 64 X2 Dual Core 5600+
nVidia Geforce GT 640 1 Gig GDDR 5
4 Gigs RAM
Windows XP Pro (SP3)

Thank You
ID: 59773 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profilemikey
Avatar

Send message
Joined: 8 May 09
Posts: 2273
Credit: 386,773,009
RAC: 349,409
300 million credit badge10 year member badgeextraordinary contributions badge
Message 59786 - Posted: 2 Sep 2013, 11:13:36 UTC - in response to Message 59773.  

I'm getting all sorts of computation errors on "de_modfit_18_3s_126testwrap_2_13727846..." Is there a "fix" for this or do I just ignore and keep on crunching?

There is no problem with de_separation_**_DR_******.

System:
Dell E521
AMD Athlon 64 X2 Dual Core 5600+
nVidia Geforce GT 640 1 Gig GDDR 5
4 Gigs RAM
Windows XP Pro (SP3)

Thank You


Are you overclocking or anything like that? You could also try a machine restart but first do a full stop and let it stay off for 2 to 3 minutes, this will clear any onboard caches that could be causing any problems. The other thing is you could go into your preferences and turn off the modified fit units and just crunch the others.
ID: 59786 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
KIDH

Send message
Joined: 3 Aug 08
Posts: 19
Credit: 10,327,719
RAC: 181
10 million credit badge10 year member badge
Message 59794 - Posted: 3 Sep 2013, 16:03:24 UTC - in response to Message 59786.  

I'm getting all sorts of computation errors on "de_modfit_18_3s_126testwrap_2_13727846..." Is there a "fix" for this or do I just ignore and keep on crunching?

There is no problem with de_separation_**_DR_******.

System:
Dell E521
AMD Athlon 64 X2 Dual Core 5600+
nVidia Geforce GT 640 1 Gig GDDR 5
4 Gigs RAM
Windows XP Pro (SP3)

Thank You


Are you overclocking or anything like that? You could also try a machine restart but first do a full stop and let it stay off for 2 to 3 minutes, this will clear any onboard caches that could be causing any problems. The other thing is you could go into your preferences and turn off the modified fit units and just crunch the others.


Just take a look at this thread: http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=3347

Everything should be fine with his computer, it's a problem with the app or the WUs, mostly seen on WinXP.

regards
KIDH
ID: 59794 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileConan
Avatar

Send message
Joined: 2 Jan 08
Posts: 105
Credit: 65,396,973
RAC: 0
50 million credit badge10 year member badge
Message 59832 - Posted: 6 Sep 2013, 23:05:45 UTC

Modfit errors for me as the system keeps sending me the Modfit Separation work units to my 32 Bit Windows computer.
I believe that they are 64 bit for Windows, hence why they are failing.

Have de-selected them now.

Conan
ID: 59832 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 · 2 · 3 · 4 · Next

Message boards : Number crunching : Computation errors

©2019 Astroinformatics Group