New faster application?
log in

Advanced search

Message boards : Number crunching : New faster application?

Previous · 1 . . . 8 · 9 · 10 · 11 · 12 · Next
Author Message
Profile GalaxyIce
Avatar
Send message
Joined: 6 Apr 08
Posts: 2018
Credit: 100,142,856
RAC: 0

Message 8646 - Posted: 19 Jan 2009, 7:52:14 UTC - in response to Message 8642.

200th post! Wow these new apps are faster!

The admins have set a credits limit of 108 credits per core/hour.

7 min a wu on my system. 60/7*12.5(aprox)=107. Ooooh. :D

8 min 34 secs per WU on my Core2.

But then its running 2 WUs at a time so I must on average be turning over a WU every 4 minutes and 17 seconds on my Core2. (Previously with the stock app it was approx. 35 minutes on average)


____________

Bill & Patsy
Avatar
Send message
Joined: 7 Jul 08
Posts: 47
Credit: 13,629,944
RAC: 0

Message 8650 - Posted: 19 Jan 2009, 10:20:16 UTC - in response to Message 8547.

Is anybody sharing this with Travis?

bump...
____________
--Bill

Sailor
Send message
Joined: 10 Feb 08
Posts: 32
Credit: 13,208,800
RAC: 0

Message 8657 - Posted: 19 Jan 2009, 12:51:03 UTC - in response to Message 8641.


I don't understand what you are saying. Should a faster processing (on GPU or CPU) get more credit or not?


Yes, my point was that a faster processor ofc should get more credits.


I'm not comparing the value of science either. How can I? My only point is that more work should give more credit, which is not the case in MW currently.


but more work IS getting more credit, the stock application doesnt get you 108 credits per hour.


About that stupid number, credit, I like it (that's me). I like looking at what I've done over the years processing for scientific projects. Plus, it's fun to play with this number and compare to countries, teams, people, computer technology over time, programming evolution over time etc. It's a good pastime. Gives me a better understanding of the world around me. But if it's capped or otherwise manipulated...


Well I do look at my stats aswell, but ill have to say I look at them from a different view. You say, the numbers tell the efford in science you did, but do they really? I mean I look at my numbers and I see MW now has caught QMC, but Ive spend much more processing time in QMC then in MW. So much for the comparison. I pretty much use credit to determine, in which project my CPU is efficient. Examples are : in Rosetta my CPU performs very poorly, the small L2 cache prevents better numbers, while Intels f.e. e8x00 series perform outstanding. So ive withdrawn from that project, not because I was looking for more credits, but because I think, my time can be spend better elsewhere - I think thats what credits are good for. In other projects you can tell from other things but credits, but Rosetta is one of those, where you can only tell from credits.

About the capped credits: I mean you are making 108 credits per hour and you talk about capped credits? And even pretty old CPUs getting these numbers. Show me any project that is even close to those numbers.

Profile Kevint
Avatar
Send message
Joined: 22 Nov 07
Posts: 285
Credit: 1,076,786,368
RAC: 0

Message 8689 - Posted: 19 Jan 2009, 23:25:19 UTC
Last modified: 19 Jan 2009, 23:26:16 UTC

No work?

I have been using the tested and approved opti app and saw there was no new work,
So I have removed the XML file and still get no new work..

Are we out of work ?

Profile Cori
Avatar
Send message
Joined: 27 Aug 07
Posts: 647
Credit: 27,592,547
RAC: 0

Message 8690 - Posted: 19 Jan 2009, 23:26:29 UTC - in response to Message 8689.

Currently server status says: "Results ready to send 0"... ;-(
____________
Lovely greetings, Cori

Misfit
Avatar
Send message
Joined: 27 Aug 07
Posts: 915
Credit: 1,503,115
RAC: 0

Message 8702 - Posted: 20 Jan 2009, 1:34:09 UTC

These optimized apps will have to be recoded.
____________
me@rescam.org

Profile Providence Christian School
Send message
Joined: 6 May 08
Posts: 7
Credit: 4,805,865
RAC: 2,688

Message 9355 - Posted: 29 Jan 2009, 15:03:15 UTC - in response to Message 8702.

So have apps which take advantage of the SSE variants been recompiled with the updated code changes? URL?

What would it take for the "stock" app to check for cpu capabilities and then download or use different code for the important parts? at least when it just comes to compiler switches. It seems silly to not use SSE(2,3,4) level instructions on machines which can take advantage of it and contribute a much greater benefit because of it.

I enjoy running the various BOINC projects; I also enjoy squeezing every last bit out of my machines and taking advantage of all their capabilities. That is the premise of BOINC right? the primary reason my machines were purchased was not for BOINC, but I run BOINC to take advantage of all that unused capacity. The same optimization principles should apply to the BOINC apps, if the machine can do SSE3 then feed them an app which can use that unused capacity, rather than a lowest common denominator, at least for the major groups of machine classes. While discovering an optimized app for a project is fun, deploying it and maintaining it on a collection of machines is not efficient and if I can have a computer perform that repetitive task...

I always find it puzzling that, cpu manufacturers tout cool new cpu features, which take months or years to make it into general compilers and then take even longer to make into applications... So your cool new whiz bang i7 is mostly running code limited to a 486 instruction set.

Profile speedimic
Avatar
Send message
Joined: 22 Feb 08
Posts: 260
Credit: 57,387,048
RAC: 0

Message 9360 - Posted: 29 Jan 2009, 15:54:07 UTC - in response to Message 8702.

These optimized apps will have to be recoded.

you mean there, don't you...?
____________
mic.


jedirock
Avatar
Send message
Joined: 8 Nov 08
Posts: 178
Credit: 6,140,854
RAC: 0

Message 9361 - Posted: 29 Jan 2009, 16:27:55 UTC - in response to Message 9355.

What would it take for the "stock" app to check for cpu capabilities and then download or use different code for the important parts? at least when it just comes to compiler switches. It seems silly to not use SSE(2,3,4) level instructions on machines which can take advantage of it and contribute a much greater benefit because of it.

Einstein@Home does this, but from what I understand, you have to have different function names for each optimized function, compile them into object (.o) files with different compiler flags, and then link them with running code that can detect SSEx on the fly. Bernd over at E@H might have some insight or code samples to do that, but I have not done anything like that.
____________

Misfit
Avatar
Send message
Joined: 27 Aug 07
Posts: 915
Credit: 1,503,115
RAC: 0

Message 9369 - Posted: 29 Jan 2009, 22:34:59 UTC - in response to Message 9360.

These optimized apps will have to be recoded.

you mean there, don't you...?

Absolutely not.
____________
me@rescam.org

Profile The Gas Giant
Avatar
Send message
Joined: 24 Dec 07
Posts: 1947
Credit: 240,865,573
RAC: 0

Message 9567 - Posted: 2 Feb 2009, 20:23:48 UTC - in response to Message 8569.

zslip is updated.

Hi Ice.
Only a little hint:
You should mention on the site that these are only Windows apps.
Someone could mistaken them for Linux ones. ;-)
The most who come there will know it from here, but just to make it sure. ;-)

Good point DoctorNow. Thanks!

zslip is updated again.



WTF. A username and password are required!

Profile GalaxyIce
Avatar
Send message
Joined: 6 Apr 08
Posts: 2018
Credit: 100,142,856
RAC: 0

Message 9569 - Posted: 2 Feb 2009, 20:38:25 UTC - in response to Message 9567.
Last modified: 2 Feb 2009, 20:39:37 UTC

zslip is updated.

Hi Ice.
Only a little hint:
You should mention on the site that these are only Windows apps.
Someone could mistaken them for Linux ones. ;-)
The most who come there will know it from here, but just to make it sure. ;-)

Good point DoctorNow. Thanks!

zslip is updated again.



WTF. A username and password are required!

Yes clever, isn't it. But even if you had the password there is nothing there since nothing has been posted on these boards (that is current) to be downloaded. When something is posted on these boards (that is current), then there might be something to download on zslip. I will be happy to provide the additional download resource again.
____________

Profile Kevint
Avatar
Send message
Joined: 22 Nov 07
Posts: 285
Credit: 1,076,786,368
RAC: 0

Message 9570 - Posted: 2 Feb 2009, 20:42:42 UTC - in response to Message 9567.

zslip is updated.

Hi Ice.
Only a little hint:
You should mention on the site that these are only Windows apps.
Someone could mistaken them for Linux ones. ;-)
The most who come there will know it from here, but just to make it sure. ;-)

Good point DoctorNow. Thanks!

zslip is updated again.



WTF. A username and password are required!



I do think those are outdated apps ---

____________
.

Profile GalaxyIce
Avatar
Send message
Joined: 6 Apr 08
Posts: 2018
Credit: 100,142,856
RAC: 0

Message 9574 - Posted: 2 Feb 2009, 21:03:16 UTC - in response to Message 9570.

zslip is updated.

Hi Ice.
Only a little hint:
You should mention on the site that these are only Windows apps.
Someone could mistaken them for Linux ones. ;-)
The most who come there will know it from here, but just to make it sure. ;-)

Good point DoctorNow. Thanks!

zslip is updated again.



WTF. A username and password are required!



I do think those are outdated apps ---

Yes, the ones I had which now don't work since they don't comply with the rules set down by Travis. I am not including anything on zslip that is not posted on these boards first, and is current, i.e., working. As I said, I will be happy to provide an additional download source if it's wanted.

____________

Profile banditwolf
Avatar
Send message
Joined: 12 Nov 07
Posts: 2425
Credit: 524,164
RAC: 0

Message 9575 - Posted: 2 Feb 2009, 21:07:46 UTC - in response to Message 9574.


Yes, the ones I had which now don't work since they don't comply with the rules set down by Travis. I am not including anything on zslip that is not posted on these boards first, and is current, i.e., working. As I said, I will be happy to provide an additional download source if it's wanted.


I do think it's great to supply another dl spot, also it's appreciated. Too bad there is nothing to dl yet.
____________
Doesn't expecting the unexpected make the unexpected the expected?
If it makes sense, DON'T do it.

Misfit
Avatar
Send message
Joined: 27 Aug 07
Posts: 915
Credit: 1,503,115
RAC: 0

Message 9685 - Posted: 4 Feb 2009, 22:58:26 UTC


Click the pic!
____________
me@rescam.org

Misfit
Avatar
Send message
Joined: 27 Aug 07
Posts: 915
Credit: 1,503,115
RAC: 0

Message 9705 - Posted: 5 Feb 2009, 3:46:31 UTC - in response to Message 9685.

I've updated my app_info.xml and replaced the versions 14 and 16 with 16 and 17. I don't know if this really makes a difference, but the app is running fine.
____________
me@rescam.org

zioriga
Send message
Joined: 30 Aug 07
Posts: 13
Credit: 15,694,454
RAC: 8,907

Message 9715 - Posted: 5 Feb 2009, 11:20:19 UTC

I tried the zslip optimized application in my AMD 5200 X2 - Windows XP 64b but all the Wu crashed.
I'm asking if those applications are accepted by the project administrator !!!
Now I discard those application and return to a normal application.

http://www.zslip.com/

Profile Gavin Shaw
Avatar
Send message
Joined: 16 Jan 08
Posts: 98
Credit: 1,371,299
RAC: 0

Message 9716 - Posted: 5 Feb 2009, 11:40:31 UTC - in response to Message 9715.
Last modified: 5 Feb 2009, 11:41:37 UTC

I tried the zslip optimized application in my AMD 5200 X2 - Windows XP 64b but all the Wu crashed.
I'm asking if those applications are accepted by the project administrator !!!
Now I discard those application and return to a normal application.

http://www.zslip.com/


Did you use the correct version? You need to use CPU-Z to find out what SSE version your processor supports and get that one. Then you have to stop Boinc and shut it down completely (so no services are running). Then install the app_info.xml and exe file in the milkyway projects folder and then start Boinc back up.

If you have any work units from the stock version and the stock version is higher than the app version then they will be trashed. If stock version is 0.17 and opt app version is 0.16 then Boinc will say that it can not process those units as the opt app is an 'older' version. They will be lost. Then try getting new work and everything should be good.

I have been using a opt app for a while and have had no trouble at all. If it is returning bad results then no one has told me.
____________
Never surrender and never give up. In the darkest hour there is always hope.

Profile caspr
Avatar
Send message
Joined: 22 Mar 08
Posts: 90
Credit: 501,728
RAC: 0

Message 9718 - Posted: 5 Feb 2009, 13:30:05 UTC - in response to Message 9715.

I tried the zslip optimized application in my AMD 5200 X2 - Windows XP 64b but all the Wu crashed.
I'm asking if those applications are accepted by the project administrator !!!
Now I discard those application and return to a normal application.

http://www.zslip.com/



I did the same with a 64bit '03 intel server w/3.0 xeon's and cant get any of these apps. to work so I guess I'll go back to linux and PG.
____________
A clear conscience is usually the sign of a bad memory



Previous · 1 . . . 8 · 9 · 10 · 11 · 12 · Next
Post to thread

Message boards : Number crunching : New faster application?


Main page · Your account · Message boards


Copyright © 2018 AstroInformatics Group