Message boards :
Number crunching :
Much slower processing with SETI@home ver 4.5
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · 6 · Next
Author | Message |
---|---|
Divide Overflow Send message Joined: 3 Apr 99 Posts: 365 Credit: 131,684 RAC: 0 |
From what I've heard now, the developers are looking at the code differences between the SETI@Home Windows/x86 application v4.03 and v4.05. I’m confident that they will spot the issue that is causing the slowdown. The likely cause for the increased completion times is due to some debugging checks that were left on unintentionally. As soon as the source of the slowdown is found, we’ll probably have another update to the Windows/x86 application. Let’s hope that this happens sooner rather than later and we’ll be crunching WU’s away at faster speeds again! |
Neee-wom Send message Joined: 3 Apr 99 Posts: 23 Credit: 5,842 RAC: 0 |
Yep... same problems. With 4.03, WUs would take 3-4 hours to complete ... with 4.05 it's 5+ hours. When is beta testing due to finish? :( |
Paul Wais Send message Joined: 26 Oct 99 Posts: 2 Credit: 605,881 RAC: 0 |
To prove to myself, I timed off a portion of a workunit without touching my computer. Seti claimed 14 minutes had elapsed, when in reality 14 minutes and 23 seconds had elapsed.... which is strange because you'd think it would be the other way around? |
Paul D. Buck Send message Joined: 19 Jul 00 Posts: 3898 Credit: 1,158,042 RAC: 0 |
> To prove to myself, I timed off a portion of a workunit without touching my > computer. Seti claimed 14 minutes had elapsed, when in reality 14 minutes and > 23 seconds had elapsed.... which is strange because you'd think it would be > the other way around? And that is as it should be. BOINC should be tracking the actual computation time of the application working. SO, you had 23 seconds of non-essential programs running on your computer. So, you wasted 23 seconds of processing time! How could you do that! :) <p> For BOINC Documentation: Click Me! |
Walt Gribben Send message Joined: 16 May 99 Posts: 353 Credit: 304,016 RAC: 0 |
> To prove to myself, I timed off a portion of a workunit without touching my > computer. Seti claimed 14 minutes had elapsed, when in reality 14 minutes and > 23 seconds had elapsed.... which is strange because you'd think it would be > the other way around? > Where do you get the elapset time from Seti? The only column I see is for CPU time. Running for 14+ minutes and seeing only 23 seconds difference between CPU time and elapsed time is pretty good, thats less than three percent "wasted" time. That 23 seconds is the overhead from operating system stuff. |
Paul D. Buck Send message Joined: 19 Jul 00 Posts: 3898 Credit: 1,158,042 RAC: 0 |
> Where do you get the elapset time from Seti? The only column I see is for CPU > time. > > Running for 14+ minutes and seeing only 23 seconds difference between CPU time > and elapsed time is pretty good, thats less than three percent "wasted" time. > That 23 seconds is the overhead from operating system stuff. I am guessing that he used a brand new invention called a "watch" because when you are at work you "watch" it all the time to see if the meeting is over yet ... :) I could be wrong though ... he may have used the vacuum tube version called a clock ... :) <p> For BOINC Documentation: Click Me! |
Ulrich Metzner Send message Joined: 3 Jul 02 Posts: 1256 Credit: 13,565,513 RAC: 13 |
> I am guessing that he used a brand new invention called a "watch" because when > you are at work you "watch" it all the time to see if the meeting is over yet > ... > > :) > > I could be wrong though ... he may have used the vacuum tube version called a > clock ... > > :) > *ROTFLMAO* Sorry, couldn't resist ;) greetz, Uli |
Walt Gribben Send message Joined: 16 May 99 Posts: 353 Credit: 304,016 RAC: 0 |
> I am guessing that he used a brand new invention called a "watch" because when > you are at work you "watch" it all the time to see if the meeting is over yet > ... > > :) > > I could be wrong though ... he may have used the vacuum tube version called a > clock ... In his original post he said "Seti claimed 14 minutes had elapsed.....". Could it be the "watch" was eyeball to screen contact, tracking the progress of the CPU Time indicator. And just as "a watched pot never boils", maybe a watched SETI unit doesn't progress? Or at a slower rate? How does it know? :) |
Paul D. Buck Send message Joined: 19 Jul 00 Posts: 3898 Credit: 1,158,042 RAC: 0 |
|
huns0004 Send message Joined: 14 Jun 01 Posts: 46 Credit: 3,208,956 RAC: 1 |
> From what I've heard now, the developers are looking at the code differences > between the SETI@Home Windows/x86 application v4.03 and v4.05. I’m confident > that they will spot the issue that is causing the slowdown. The likely cause > for the increased completion times is due to some debugging checks that were > left on unintentionally. It would be nice if this could be confirmed by somebody in the know! A news item for the front page, perhaps? |
Misfit Send message Joined: 21 Jun 01 Posts: 21804 Credit: 2,815,091 RAC: 0 |
> In his original post he said "Seti claimed 14 minutes had elapsed.....". > Could it be the "watch" was eyeball to screen contact, tracking the progress > of the CPU Time indicator. > > And just as "a watched pot never boils", maybe a watched SETI unit doesn't > progress? Or at a slower rate? How does it know? Oh my. How will that effect the Einstein@Home project since gravity warps space/time? |
Petit Soleil Send message Joined: 17 Feb 03 Posts: 1497 Credit: 70,934 RAC: 0 |
> Oh my. How will that effect the Einstein@Home project since gravity warps > space/time? > Geodesic !!! And Don't wath too much the progress bar, to observe is to disturb. |
Siran d'Vel'nahr Send message Joined: 23 May 99 Posts: 7379 Credit: 44,181,323 RAC: 238 |
> I don't know exactly if the processing time has doubled, but I noticed that > the new WU, I downloaded was 5.09 Mb big instead of usually about 600 kb. And > it runs more equally (it this the right word??). Instead of finishing the > first about 65 % in 3 minutes, it has finished about 30 % in about 30 minutes. > > All my WUs are roughly 354KB in size, and my processing time has almost doubled. The old WUs were just under the above size, if I remember right. L8R.... --- Rick A. - BOINCing right along now.... It can only get better! "There is no fate except that which we create for ourselves." Live Long and Prosper.... |
Siran d'Vel'nahr Send message Joined: 23 May 99 Posts: 7379 Credit: 44,181,323 RAC: 238 |
> > > > > > > I LIKE your avatar!!!! > > > > L8R.... > > Me too ;) > It's from a moderator in a german firefox forum. I hope he won't sue me for > that ... LOL > > BTW: Tried several versions of Opera and Netscape but Firefox is the first > alternative browser i instantly fell in love with ;) > > > > greetz, Uli > > I started out with Netscape when it was v3.xx and quit using it at v4.72. I only used IE when I absolutely had to such as going to a Microscuz site. After v4.72 Netscape I got Mozilla and haven't looked back since. I don't even have to use IE on Microscuz sites any more. I like open source programming. My office suite is open source from OpenOffice.org as is Mozilla and I think, FireFox. Opera is free (ad supported) or shareware ($$). I will use shareware but not ad supported software. L8R.... --- Rick A. - BOINCing right along now.... It can only get better! "There is no fate except that which we create for ourselves." Live Long and Prosper.... |
Misfit Send message Joined: 21 Jun 01 Posts: 21804 Credit: 2,815,091 RAC: 0 |
Well I finally ran outta 4.03 WU's and forced to upgrade. :( With 4.03 a WU would take 3Hrs 3Min average. Now with 4.05 (4.09) it takes 4hrs 15min average. Some even continue to run for awhile after showing 100%. Why Oh Why did Boinc have to fix something that wasnt broken? |
Siran d'Vel'nahr Send message Joined: 23 May 99 Posts: 7379 Credit: 44,181,323 RAC: 238 |
> Well I finally ran outta 4.03 WU's and forced to upgrade. :( > With 4.03 a WU would take 3Hrs 3Min average. > Now with 4.05 (4.09) it takes 4hrs 15min average. > Some even continue to run for awhile after showing 100%. > Why Oh Why did Boinc have to fix something that wasnt broken? > > I ranted about that very thought in another thread last night. I was crunching like mad with v4.05 core client and v4.03 SETI client. Now, it's as you stated above. My WUs have almost doubled in processing time and run after the 100% completion. I've even noticed BOINC doing some other crazy things. For one thing I can now run 3 WUs at the same time. Now, that I really don't mind at all. But, BOINC has locked up, well, not locked up, just quit responding. I ended up re-installing v4.09 CC which fixed it, but it has happened once since then. I really don't think that v4.09 CC is all that stable as v4.05 was. L8R.... --- Rick A. - BOINCing right along now.... It can only get better! "There is no fate except that which we create for ourselves." Live Long and Prosper.... |
Walt Gribben Send message Joined: 16 May 99 Posts: 353 Credit: 304,016 RAC: 0 |
> Well I finally ran outta 4.03 WU's and forced to upgrade. :( > With 4.03 a WU would take 3Hrs 3Min average. > Now with 4.05 (4.09) it takes 4hrs 15min average. > Some even continue to run for awhile after showing 100%. > Why Oh Why did Boinc have to fix something that wasnt broken? Don't worry about how long it takes to process, the question is, will the server accept the results when uploaded? Other than that, I agree with you, BOINC 4.05 (4.06 on Win98) with Seti 4.03 worked just fine. WU's on my PIII machine went from under 12 hours to over 16. |
CyberGoyle Send message Joined: 2 Jun 99 Posts: 160 Credit: 3,622,756 RAC: 26 |
Has anyone considered that the 4.05 client is performing more calculation that the 4.03 client, thereby increasing WU processing time? Berkeley has done this in the past, specifically from the 2.x Classic client to the 3.x Classic client. <a> |
Bill & Patsy Send message Joined: 6 Apr 01 Posts: 141 Credit: 508,875 RAC: 0 |
> Has anyone considered that the 4.05 client is performing more calculation that > the 4.03 client, thereby increasing WU processing time? Berkeley has done this > in the past, specifically from the 2.x Classic client to the 3.x Classic > client. > > > > > <a> > How could we possibly consider this very sensible possibility when Berkeley is once again stone-walling us!!?? This thread started over a week ago. Yet in all this time they have given no explanation, nor done anything visible to cure the problem (if it is a problem). It is certainly a problem of some sort when so many users are this upset about it. How about it, Berkeley?????????? --Bill Z. |
Siran d'Vel'nahr Send message Joined: 23 May 99 Posts: 7379 Credit: 44,181,323 RAC: 238 |
> Has anyone considered that the 4.05 client is performing more calculation that > the 4.03 client, thereby increasing WU processing time? Berkeley has done this > in the past, specifically from the 2.x Classic client to the 3.x Classic > client. > > I can't disagree with you there. My WUs for "Classic" more than doubled. I also believe that the upgrade from v2.x to v3.x was also to eliminate cheating, wasn't it? L8R.... --- Rick A. - BOINCing right along now.... It can only get better! "There is no fate except that which we create for ourselves." Live Long and Prosper.... |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.