Welcome to MilkyWay@home

Nice!


Advanced search

Message boards : Number crunching : Nice!
Message board moderation

To post messages, you must log in.

AuthorMessage
Stefan Ver3

Send message
Joined: 17 May 08
Posts: 16
Credit: 528,507
RAC: 0
500 thousand credit badge10 year member badge
Message 4349 - Posted: 22 Jul 2008, 17:54:47 UTC

I got a laugh out of this client.

Now thats a smoking cpu!#23471

Even managed to snip a w/u screen capture!


Must have rockets for rollerskates.


ID: 4349 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Odd-Rod

Send message
Joined: 7 Sep 07
Posts: 442
Credit: 1,397,909
RAC: 213
1 million credit badge10 year member badge
Message 4351 - Posted: 22 Jul 2008, 19:20:38 UTC

I've just downloaded and completed a WU in 10min 48sec. Seems there are a couple of short ones to be had.

It was amazing to watch the estimated remaining time drop so quickly. Unfortunately the next WU looks like a long one...
ID: 4351 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileThunder
Avatar

Send message
Joined: 9 Jul 08
Posts: 85
Credit: 44,842,651
RAC: 0
30 million credit badge10 year member badge
Message 4355 - Posted: 22 Jul 2008, 19:57:39 UTC

I think the issue is that you probably got 4 credits for that short one.

This client got 260 credits for what it reported as 17.5 minutes or so. With the reported processor, I don't see how it could have completed the task even if it started working on it the moment it was sent, but then the RAC is completely nuts for that processor as well so something is off.
ID: 4355 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John McLeod VII
Avatar

Send message
Joined: 27 Aug 07
Posts: 85
Credit: 405,705
RAC: 0
100 thousand credit badge10 year member badge
Message 4360 - Posted: 22 Jul 2008, 20:26:40 UTC - in response to Message 4355.  
Last modified: 22 Jul 2008, 20:29:50 UTC

I think the issue is that you probably got 4 credits for that short one.

This client got 260 credits for what it reported as 17.5 minutes or so. With the reported processor, I don't see how it could have completed the task even if it started working on it the moment it was sent, but then the RAC is completely nuts for that processor as well so something is off.

It is possible that someone is gaming the system since there is no [edit]apparent[/edit] validation done here.

If the user were to attach the header for the task to some stock data that was calculated for a previous task, there is no evident validation to stop them. S@H Classic had exactly the same problem, which is why S@H uses 2 tasks per WU and makes certain they match before granting credit.


BOINC WIKI
ID: 4360 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Stefan Ver3

Send message
Joined: 17 May 08
Posts: 16
Credit: 528,507
RAC: 0
500 thousand credit badge10 year member badge
Message 4364 - Posted: 22 Jul 2008, 20:38:20 UTC

Fish on! Fish on! Got me another one!

Nice Pentium4 2.8Ghz #23272

Smile!! You ought to be in pictures! :D



I'm on FIRE today!
ID: 4364 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileCrunch3r
Volunteer developer
Avatar

Send message
Joined: 17 Feb 08
Posts: 363
Credit: 258,227,990
RAC: 0
200 million credit badge10 year member badge
Message 4365 - Posted: 22 Jul 2008, 20:45:46 UTC - in response to Message 4364.  

Fish on! Fish on! Got me another one!

Nice Pentium4 2.8Ghz #23272

Smile!! You ought to be in pictures! :D



I'm on FIRE today!


And now what exactly makes you think that this is cheating ?



Join Support science! Joinc Team BOINC United now!
ID: 4365 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
John McLeod VII
Avatar

Send message
Joined: 27 Aug 07
Posts: 85
Credit: 405,705
RAC: 0
100 thousand credit badge10 year member badge
Message 4368 - Posted: 22 Jul 2008, 21:04:15 UTC

I did not say it was cheating, that has yet to be proved. I said it could be cheating, that has yet to be disproved. It could be other things as well (such as a win9X client that restarted causing the CPU time to go to 0). Until there is some sort of investigation, we won't know either way. I believe that it is probably only the project admins that can do the investigation, we can only bring the possibility of a problem to their attention.


BOINC WIKI
ID: 4368 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileCrunch3r
Volunteer developer
Avatar

Send message
Joined: 17 Feb 08
Posts: 363
Credit: 258,227,990
RAC: 0
200 million credit badge10 year member badge
Message 4369 - Posted: 22 Jul 2008, 21:06:32 UTC - in response to Message 4368.  

I did not say it was cheating, that has yet to be proved. I said it could be cheating, that has yet to be disproved. It could be other things as well (such as a win9X client that restarted causing the CPU time to go to 0). Until there is some sort of investigation, we won't know either way. I believe that it is probably only the project admins that can do the investigation, we can only bring the possibility of a problem to their attention.


There is no problem with this nor is it cheating. It's just ... doing the math a little bit faster that's all... and of course its valid.




Join Support science! Joinc Team BOINC United now!
ID: 4369 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Stefan Ver3

Send message
Joined: 17 May 08
Posts: 16
Credit: 528,507
RAC: 0
500 thousand credit badge10 year member badge
Message 4371 - Posted: 22 Jul 2008, 21:15:34 UTC - in response to Message 4369.  

I did not say it was cheating, that has yet to be proved. I said it could be cheating, that has yet to be disproved. It could be other things as well (such as a win9X client that restarted causing the CPU time to go to 0). Until there is some sort of investigation, we won't know either way. I believe that it is probably only the project admins that can do the investigation, we can only bring the possibility of a problem to their attention.


There is no problem with this nor is it cheating. It's just ... doing the math a little bit faster that's all... and of course its valid.




Don't get me wrong now, I never said there was cheating. If I suspected there was cheating, I would have flat out said it was.
I need me a farm of those Pentium 4's! LOL
ID: 4371 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileCrunch3r
Volunteer developer
Avatar

Send message
Joined: 17 Feb 08
Posts: 363
Credit: 258,227,990
RAC: 0
200 million credit badge10 year member badge
Message 4372 - Posted: 22 Jul 2008, 21:16:54 UTC - in response to Message 4371.  


Don't get me wrong now, I never said there was cheating. If I suspected there was cheating, I would have flat out said it was.
I need me a farm of those Pentium 4's! LOL


the farm of P4s won't help... you need the application to do this ;)



Join Support science! Joinc Team BOINC United now!
ID: 4372 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Stefan Ver3

Send message
Joined: 17 May 08
Posts: 16
Credit: 528,507
RAC: 0
500 thousand credit badge10 year member badge
Message 4374 - Posted: 22 Jul 2008, 21:23:40 UTC - in response to Message 4372.  


Don't get me wrong now, I never said there was cheating. If I suspected there was cheating, I would have flat out said it was.
I need me a farm of those Pentium 4's! LOL


the farm of P4s won't help... you need the application to do this ;)



MINT! Crunch3r :D :D
ID: 4374 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileThunder
Avatar

Send message
Joined: 9 Jul 08
Posts: 85
Credit: 44,842,651
RAC: 0
30 million credit badge10 year member badge
Message 4375 - Posted: 22 Jul 2008, 21:46:01 UTC - in response to Message 4369.  

There is no problem with this nor is it cheating. It's just ... doing the math a little bit faster that's all... and of course its valid.


Glad to know you've been continuing the optimization Crunch3r!

I must admit that I suspected there were some serious opportunities for it when I saw that a 6 year old Celeron 2Ghz was only taking about 40% longer to do a task than a Xeon 5120 on the stock app. ;)

So are you hopefully going to share with the whole class and provide these to the developers? :D
ID: 4375 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileCrunch3r
Volunteer developer
Avatar

Send message
Joined: 17 Feb 08
Posts: 363
Credit: 258,227,990
RAC: 0
200 million credit badge10 year member badge
Message 4377 - Posted: 22 Jul 2008, 22:04:42 UTC - in response to Message 4375.  


Glad to know you've been continuing the optimization Crunch3r!

I must admit that I suspected there were some serious opportunities for it when I saw that a 6 year old Celeron 2Ghz was only taking about 40% longer to do a task than a Xeon 5120 on the stock app. ;)

So are you hopefully going to share with the whole class and provide these to the developers? :D


just take a look at this thread... i think the answer to your question becomes quite clear...

besides that the "developers" do know what can be done to improve the apps performance cuz i told them a few months ago...


Join Support science! Joinc Team BOINC United now!
ID: 4377 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileJayargh
Avatar

Send message
Joined: 8 Oct 07
Posts: 289
Credit: 3,690,838
RAC: 0
3 million credit badge10 year member badge
Message 4379 - Posted: 22 Jul 2008, 22:17:07 UTC - in response to Message 4377.  


Glad to know you've been continuing the optimization Crunch3r!

I must admit that I suspected there were some serious opportunities for it when I saw that a 6 year old Celeron 2Ghz was only taking about 40% longer to do a task than a Xeon 5120 on the stock app. ;)

So are you hopefully going to share with the whole class and provide these to the developers? :D


just take a look at this thread... i think the answer to your question becomes quite clear...

besides that the "developers" do know what can be done to improve the apps performance cuz i told them a few months ago...


Yes give the 'enemy' more ammunition while at war! :D

But seriously last time it was done the project adjusted credit accordingly ;)
ID: 4379 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileThunder
Avatar

Send message
Joined: 9 Jul 08
Posts: 85
Credit: 44,842,651
RAC: 0
30 million credit badge10 year member badge
Message 4393 - Posted: 22 Jul 2008, 23:31:17 UTC - in response to Message 4377.  

just take a look at this thread... i think the answer to your question becomes quite clear...

besides that the "developers" do know what can be done to improve the apps performance cuz i told them a few months ago...


Uh... I'm lost. What does yet another "Wahhh... credit is too <low/high/sideways/unfair/shiny>" thread have to do with improving the task execution times?

That's at least a 20-fold increase in speed! They could increase the precision to the level of the 373 tasks and still be completing them in only slightly longer than the time of the original tasks! Damn! :D

What the heck are they waiting on? Are you positive these are producing identical output on identical tasks? (or at least close enough for the science value?)
ID: 4393 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileCrunch3r
Volunteer developer
Avatar

Send message
Joined: 17 Feb 08
Posts: 363
Credit: 258,227,990
RAC: 0
200 million credit badge10 year member badge
Message 4394 - Posted: 22 Jul 2008, 23:53:36 UTC - in response to Message 4393.  
Last modified: 22 Jul 2008, 23:54:06 UTC


Uh... I'm lost. What does yet another "Wahhh... credit is too <low/high/sideways/unfair/shiny>" thread have to do with improving the task execution times?


there are a couple of other things involved in that decision. seeing the whiners here is just the tip of the iceberg...


That's at least a 20-fold increase in speed! They could increase the precision to the level of the 373 tasks and still be completing them in only slightly longer than the time of the original tasks! Damn! :D


might be even faster... i could get a rac of 450K/day out of a single V8 xeon if it would let it crunch 24h... but i dont.


What the heck are they waiting on?


i guess money/time plays a part..


Are you positive these are producing identical output on identical tasks? (or at least close enough for the science value?)


results were identical in my off-line tests.

Join Support science! Joinc Team BOINC United now!
ID: 4394 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
ProfileThunder
Avatar

Send message
Joined: 9 Jul 08
Posts: 85
Credit: 44,842,651
RAC: 0
30 million credit badge10 year member badge
Message 4401 - Posted: 23 Jul 2008, 1:45:10 UTC - in response to Message 4394.  

Well, I have at least a few older machines that I could theoretically do testing on if you're needing to do any validation on less common processors.

Shoot me a message if I can ever be of assistance. Until then, I'll just hope the project folks can make use of your improvements.

Good luck! :)
ID: 4401 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : Number crunching : Nice!

©2019 Astroinformatics Group