VALIDATION INCONCLUSIVE


log in

Advanced search

Message boards : Number crunching : VALIDATION INCONCLUSIVE

Previous · 1 · 2 · 3 · 4 · 5 · Next
Author Message
Profile Slavac
Volunteer tester
Avatar
Send message
Joined: 27 Apr 11
Posts: 1932
Credit: 17,952,639
RAC: 0
United States
Message 1297472 - Posted: 21 Oct 2012, 7:34:50 UTC - in response to Message 1297464.

That's a good point.

I'm curious to see how many project resources are being allocated to reprocessing invalid/error tasks.

I'll see if I can find anything out about this.

Next up on my list:

699798 (DG71)
8637650 (mamoru@igo-shogi)
9774123 (Nolan)
36267 (Eric Livingston)
285742 (3rik)
7888117 (HalbesHuhn)

My list for this month has now passed 100 (some have already named here) and if this month is like an other lately then I'll likely get another 15-25 (I must say though that I'm getting a better response to PM's now).

Cheers.


Should we set up a 'master list' thread?

I'd really like to help these folks sort out their machines. I would've abandoned SETI crunching all together if a few very smart gents wouldn't have helped me sort out my error/invalid issues.

The more people we can help get their machines up and running the more processing power SETI will have while at the same time reducing our total resources allocated to improperly configured machines.
____________


Executive Director GPU Users Group Inc. -
brad@gpuug.org

Profile Wiggo
Avatar
Send message
Joined: 24 Jan 00
Posts: 7130
Credit: 95,300,186
RAC: 73,973
Australia
Message 1297481 - Posted: 21 Oct 2012, 8:07:15 UTC - in response to Message 1297479.

At least now with the VALIDATION INCONCLUSIVE section people can easily see the source of the problem and hopefully more will PM those they come across.

It's certainly much better now than having to go through a hundred or so pages of pendings to find them.

Cheers.
____________

Terror Australis
Volunteer tester
Send message
Joined: 14 Feb 04
Posts: 1715
Credit: 205,151,357
RAC: 27,702
Australia
Message 1297485 - Posted: 21 Oct 2012, 8:23:44 UTC - in response to Message 1297479.

The only problem there, my friend, is a lot of folks were drawn into this project by it's initial promise.......
You download the software, you give it the go, and it 'runs in the background using resources only when you are not using them'. Period.
They don't have any regard or whim what happens after that. It was just like downloading a screensaver......which is old hat, for sure, but that is all many folks regard it as. They don't allocate resources to various GPUs, they trust the Seti Project will not download any software that would make their computers unusable in any way........

The promise was not to use any resources required by the user. You can't take that promise back now just because the apps have improved.
Unless it's implicit by user choice.

No-one is talking about taking anything from anybody. I'm sure that most of these people want to return valid results, there is no point in crunching otherwise.

A polite, general note to the effect that a software bug is effecting their video cards and pointing them to this forum for assistance. Maybe the V10 app should be upgraded to make it more 560 friendly and set to download automatically ? There are many benefits to the project to encourage this.

If nothing else, fixing this problem would save everybody a lot of bandwidth. 6,000 units (the worst case I've found so far) at 367KB each, equals 2.2GB of data. That's a lot bits to just pour down the sink.

T.A.

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5831
Credit: 59,460,284
RAC: 47,863
Australia
Message 1297489 - Posted: 21 Oct 2012, 8:38:51 UTC - in response to Message 1297487.
Last modified: 21 Oct 2012, 8:39:26 UTC

A polite, general note to the effect that a software bug is effecting their video cards and pointing them to this forum for assistance.

Is it a software bug, or is it the fact that their hardware isn't up to the task?


NO, NOT a freaking polite little note to say 'Geez, friend, you've joined our project but now we've screwed your computer a bit......could you fix it a bit???????'

Why not ask them to fix their faulty system?
Most of the problems i've seen addressed here aren't due to the the Seti software, but other issues with the users systems.
____________
Grant
Darwin NT.

Profile Wiggo
Avatar
Send message
Joined: 24 Jan 00
Posts: 7130
Credit: 95,300,186
RAC: 73,973
Australia
Message 1297494 - Posted: 21 Oct 2012, 8:49:07 UTC

I just have a standard letter that I use that I edit to suit the problem and add the host.

Hi there, Could you please check your computer, [url=http://setiathome.berkeley.edu/show_host_detail.php?hostid=####]ID number ####[/url], as it is producing a very high, unsatisfactory rate of errored and/or invalid results that in the end puts more strain on the servers. If you cannot solve the problem yourself then go to the [url=http://setiathome.berkeley.edu/forum_forum.php?id=10]Number Crunching Forum[/url] and start a new thread asking for help. Cheers, Wiggo.


The "####" is where I insert the host's ID number.

Cheers.
____________

Profile Slavac
Volunteer tester
Avatar
Send message
Joined: 27 Apr 11
Posts: 1932
Credit: 17,952,639
RAC: 0
United States
Message 1297499 - Posted: 21 Oct 2012, 8:51:31 UTC - in response to Message 1297489.

Mark you're not making sense here.

Just because someone has an issue with crunching on SETI we should ask them to completely abandon crunching for SETI?

It's impractical to suggest that due to the simple nature of the SETI project's origins we should expect everything to work flawlessly in 2012. Opt aps, hundreds of various GPU's, new forms of work, new drivers, ATI, Nvidia, AMD, Intel, so on and so forth all add to the complexity and thus potential issues.

Some people will have an issue here or there, so let's fix those issues for the betterment of everyone rather than raising the white flag while standing atop a roadblock in the middle of the road to progress.
____________


Executive Director GPU Users Group Inc. -
brad@gpuug.org

Profile Wiggo
Avatar
Send message
Joined: 24 Jan 00
Posts: 7130
Credit: 95,300,186
RAC: 73,973
Australia
Message 1297505 - Posted: 21 Oct 2012, 9:10:58 UTC - in response to Message 1297496.

I just have a standard letter that I use that I edit to suit the problem and add the host.

Hi there, Could you please check your computer, [url=http://setiathome.berkeley.edu/show_host_detail.php?hostid=####]ID number ####[/url], as it is producing a very high, unsatisfactory rate of errored and/or invalid results that in the end puts more strain on the servers. If you cannot solve the problem yourself then go to the [url=http://setiathome.berkeley.edu/forum_forum.php?id=10]Number Crunching Forum[/url] and start a new thread asking for help. Cheers, Wiggo.


The "####" is where I insert the host's ID number.

Cheers.

And if I ever got such a message, I'd tell you what to insert and where.


Well I'm no literary expert but I'm open to any suggestions that will make it more polite and friendly.

Cheers.
____________

Profile Slavac
Volunteer tester
Avatar
Send message
Joined: 27 Apr 11
Posts: 1932
Credit: 17,952,639
RAC: 0
United States
Message 1297506 - Posted: 21 Oct 2012, 9:17:11 UTC - in response to Message 1297504.

Mark you're not making sense here.

Just because someone has an issue with crunching on SETI we should ask them to completely abandon crunching for SETI?

It's impractical to suggest that due to the simple nature of the SETI project's origins we should expect everything to work flawlessly in 2012. Opt aps, hundreds of various GPU's, new forms of work, new drivers, ATI, Nvidia, AMD, Intel, so on and so forth all add to the complexity and thus potential issues.

Some people will have an issue here or there, so let's fix those issues for the betterment of everyone rather than raising the white flag while standing atop a roadblock in the middle of the road to progress.

My point was, Brad........
I do not dismiss the problems involved.
I only invoked the initial promise of the project.
The holy grail, so to speak.

Am I wrong in that?
If thine eye offends thee, then pluck it out.


I see where you're coming from but that's just not practical. It's similar to a web hosting service advertising guaranteed 100% uptime. Almost every one of these companies will experience some downtime at some point. Sometimes the best intentions simply aren't practical.

What we can do here is help these users who are having an issue for whatever reason then move on to more important tasks like enjoying a nice frothy beverage while our machines hum quietly in the background, error free.
____________


Executive Director GPU Users Group Inc. -
brad@gpuug.org

Terror Australis
Volunteer tester
Send message
Joined: 14 Feb 04
Posts: 1715
Credit: 205,151,357
RAC: 27,702
Australia
Message 1297509 - Posted: 21 Oct 2012, 9:23:24 UTC - in response to Message 1297489.
Last modified: 21 Oct 2012, 9:28:01 UTC

Is it a software bug, or is it the fact that their hardware isn't up to the task?

In the case of 560 Ti's I think you could call it a bug. The stock V10 app does not play well with them at all. From what I've found so far, the top 5 invalid producers are all 560 Ti GPU's


NO, NOT a freaking polite little note to say 'Geez, friend, you've joined our project but now we've screwed your computer a bit......could you fix it a bit???????'

Why not ask them to fix their faulty system?
Most of the problems i've seen addressed here aren't due to the the Seti software, but other issues with the users systems.

There is no point crunching if your not producing valid results. It's just a waste of time, electricity, bandwidth and enthusiasm otherwise.

EDIT: This is why a revised stock app that is more 5xx/6xx friendly is probably the best solution. It could be sent out as a download with a batch of work units and nobody would even know.

T.A.

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5831
Credit: 59,460,284
RAC: 47,863
Australia
Message 1297520 - Posted: 21 Oct 2012, 10:03:23 UTC - in response to Message 1297493.

You are not getting my point, friend.

Nope, i'm not following at all.


The entire premise of the Seti project was that it would run in the background on any and all computers when they were not being otherwise utilized for their usual tasks.

Which it does, pretty much.
Like all software, there are some issues that relate to certain configurations. It would be nice if Seti could produce perfect software, but as most other companies with much larger budgets aren't able to, i don't see why you should expect it of Seti.


What part of not being used for other tasks do you not understand?

What part of not being used for other tasks has anything to do with the problems being discussed here?


The user notes I am sure did not include 'uhhh.....some adjustments to improve your user experience after loading the Seti program might be involved...'

Which would make it the only piece of software on the planet where that isn't the case.
____________
Grant
Darwin NT.

fscheel
Send message
Joined: 13 Apr 12
Posts: 73
Credit: 11,135,641
RAC: 0
United States
Message 1297524 - Posted: 21 Oct 2012, 10:44:52 UTC

In progress (1787) · Validation pending (2850) · Validation inconclusive (188) · Valid (1060) · Invalid (0) · Error (41)

Since most of the 41 errors are left over from the download issues, I'll just keep on keeping on. While I would certainly like to see the validation pending number be a lot less, it certainly is not a deal breaker. I just hope my pc's don't start contributing to the Invalid column.


:)
Frank

Previous · 1 · 2 · 3 · 4 · 5 · Next

Message boards : Number crunching : VALIDATION INCONCLUSIVE

Copyright © 2014 University of California