and I think we had someone else here who complained about extreme runtimes on his Celeron. I checked and that box has 128 KB of cache...
I remember that, but the comment was made based on the forecast for the ETA, which can be off by miles. That Celeron was a Coppermine, we'll know only in a few days what the real runtime & credit was.
You are correct. That's my machine, and after 22 hours, the times are coming more in line with previous results. It was just very strange how the initial expected completion times were so different from the standard app. The standard app started at around 50 hrs while the unit actually took about 72. The beta started at 180hours, but current status shows 22hrs and 27% complete. That approximates 81 hrs to complete. We'll see what the final tally is, but this doesn't look like what Annika is seeing. :(
Looks like the Beta App isn't doing (much) worse than the 4.18 and it fixes the major source of the SIGABRT problem, which is the problem that wastes by far the most of computing time (and credit). I guess we'll make it official in the next hours (or latest tomorrow).
Annika, I don't have any clue yet what happens on your machine, but as I haven't seen a noticable slowdown on any other machine I got results from, I think that it's an issue limited to your machine. I'll start working on speeding up the App soon anyway.
When the 4.21 is official, I might post another version of the same code compiled with the compiler version I used previously - if I find the time to buid one before I have a code that is faster anyway.
Just started the first WU; completion time is around 46 hours... seems to have gotten a little faster towards the end, but still a bit slow... Completion time closely resembles my notebook, but that is a dual core and does two at a time, plus the Cores seem to suffer more from the current lack of optimization. Normally the notebook took around 70% longer, so that is about the difference between the two apps atm. The result validated okay, though.
Just started my second WU and it looks like it MIGHT be faster... as I said, I only just started, but it looks like it's going to take between 30 and 35 hours, which wouldn't be great but acceptable for this box (probably still better than Windows). Given it is another 500-credit-WU (which I can't check right now because the "my results" page won't load) it looks like it was some weird interaction between my system and this WU (stupid example but think of it like not being able to play a certain CD in a certain player although both are perfectly okay) which doesn't really have any significance. I'll watch the second result closely but I do hope it was a one-timer. Sorry about the trouble I gave everyone here.
[Edited for spelling]
[Edit: Just checked, my wingman hasn't finished the WU yet, either, so still no idea what it's worth- do hope it's a fat one cause that would hint at a performance increase on my box]
Important think first: Yes, it was a 500+ credit WU ! Wingman also has an AMD 64, but a faster one and on XP, finished it in less time.
So it seems that the "AMD kind of not so fast under Windows" problem is now spreading to Linux as well, but it's not as catastrophic as initially thought.
@Bernd: if a new version is officially released and auto-downloaded to clients, will this cause the WUs currently in progress under the beta apps to crash (as was the case after installing the beta app), or will they run to completion?
if a new version is officially released and auto-downloaded to clients, will this cause the WUs currently in progress under the beta apps to crash (as was the case after installing the beta app), or will they run to completion?
As long as the app_info.xml is present, your Client won't download a new App automatically, it will continue to use the 4.21.
To get back to the automatic update path you need to stop the client, remove the app_info.xml file and start the client again. If you do this while the official App has the same version number as the one you were using, the client should find the App already present on your system and don't download it, and the App will just continue from the checkpoint.
There are, however, some BOINC clients that have problems with this transition, because actually the platform changes (from anonymous to linux-x86). The safest way would be to set the project to "no new work", let the client finish all tasks it got, then stop the client, remove the app_info.xml, start the client again and set the project to "allow more work".
Almost 8% in 2.5 hours, meaning it is still on track... maybe I'll really be able to finish this WU in a reasonable amount of time. I'll be at uni now and leave the WU alone, check on it again some time in the evening...
Yep, 44% in about 13 hours, which still hints at a completion time little over 30 hours. I think we can definitely file my previous problems under "strange glitches that occur once and never again" or if you prefer it "shit happens" ;-) Really stupid it had to happen in a beta test, of all possible unsuitable moments...
Yep, 44% in about 13 hours, which still hints at a completion time little over 30 hours. I think we can definitely file my previous problems under "strange glitches that occur once and never again" or if you prefer it "shit happens" ;-) Really stupid it had to happen in a beta test, of all possible unsuitable moments...
Oh oh... sh*t just happened again....
My Palomino Athlon XP 1800+ is now only halfway thru a workunit which seems to be a tiny 170 credits one. But it has already spent more than 60.000 CPU seconds on it, which used to be more than enought to complete a unit that size !! Either the wingman was "under-claiming" or the "Annika-effect" just hit my AMD box.
RE: RE: and I think we
)
You are correct. That's my machine, and after 22 hours, the times are coming more in line with previous results. It was just very strange how the initial expected completion times were so different from the standard app. The standard app started at around 50 hrs while the unit actually took about 72. The beta started at 180hours, but current status shows 22hrs and 27% complete. That approximates 81 hrs to complete. We'll see what the final tally is, but this doesn't look like what Annika is seeing. :(
Second WU on beta app
)
Second WU on beta app validated w/o problem and w/ good performance :-).
I've two older 32bit Athlon XPs working w/ the beta app now.
CU
BRM
Looks like the Beta App isn't
)
Looks like the Beta App isn't doing (much) worse than the 4.18 and it fixes the major source of the SIGABRT problem, which is the problem that wastes by far the most of computing time (and credit). I guess we'll make it official in the next hours (or latest tomorrow).
Annika, I don't have any clue yet what happens on your machine, but as I haven't seen a noticable slowdown on any other machine I got results from, I think that it's an issue limited to your machine. I'll start working on speeding up the App soon anyway.
When the 4.21 is official, I might post another version of the same code compiled with the compiler version I used previously - if I find the time to buid one before I have a code that is faster anyway.
BM
BM
Just started the first WU;
)
Just started the first WU; completion time is around 46 hours... seems to have gotten a little faster towards the end, but still a bit slow... Completion time closely resembles my notebook, but that is a dual core and does two at a time, plus the Cores seem to suffer more from the current lack of optimization. Normally the notebook took around 70% longer, so that is about the difference between the two apps atm. The result validated okay, though.
Just started my second WU and it looks like it MIGHT be faster... as I said, I only just started, but it looks like it's going to take between 30 and 35 hours, which wouldn't be great but acceptable for this box (probably still better than Windows). Given it is another 500-credit-WU (which I can't check right now because the "my results" page won't load) it looks like it was some weird interaction between my system and this WU (stupid example but think of it like not being able to play a certain CD in a certain player although both are perfectly okay) which doesn't really have any significance. I'll watch the second result closely but I do hope it was a one-timer. Sorry about the trouble I gave everyone here.
[Edited for spelling]
[Edit: Just checked, my wingman hasn't finished the WU yet, either, so still no idea what it's worth- do hope it's a fat one cause that would hint at a performance increase on my box]
Wingman's result is in
)
Wingman's result is in now!
Important think first: Yes, it was a 500+ credit WU ! Wingman also has an AMD 64, but a faster one and on XP, finished it in less time.
So it seems that the "AMD kind of not so fast under Windows" problem is now spreading to Linux as well, but it's not as catastrophic as initially thought.
@Bernd: if a new version is officially released and auto-downloaded to clients, will this cause the WUs currently in progress under the beta apps to crash (as was the case after installing the beta app), or will they run to completion?
CU
BRM
RE: if a new version is
)
As long as the app_info.xml is present, your Client won't download a new App automatically, it will continue to use the 4.21.
To get back to the automatic update path you need to stop the client, remove the app_info.xml file and start the client again. If you do this while the official App has the same version number as the one you were using, the client should find the App already present on your system and don't download it, and the App will just continue from the checkpoint.
There are, however, some BOINC clients that have problems with this transition, because actually the platform changes (from anonymous to linux-x86). The safest way would be to set the project to "no new work", let the client finish all tasks it got, then stop the client, remove the app_info.xml, start the client again and set the project to "allow more work".
BM
BM
Almost 8% in 2.5 hours,
)
Almost 8% in 2.5 hours, meaning it is still on track... maybe I'll really be able to finish this WU in a reasonable amount of time. I'll be at uni now and leave the WU alone, check on it again some time in the evening...
Yep, 44% in about 13 hours,
)
Yep, 44% in about 13 hours, which still hints at a completion time little over 30 hours. I think we can definitely file my previous problems under "strange glitches that occur once and never again" or if you prefer it "shit happens" ;-) Really stupid it had to happen in a beta test, of all possible unsuitable moments...
RE: Yep, 44% in about 13
)
Oh oh... sh*t just happened again....
My Palomino Athlon XP 1800+ is now only halfway thru a workunit which seems to be a tiny 170 credits one. But it has already spent more than 60.000 CPU seconds on it, which used to be more than enought to complete a unit that size !! Either the wingman was "under-claiming" or the "Annika-effect" just hit my AMD box.
Go figure!
CU
BRM
That doesn't sound too got.
)
That doesn't sound too got. I'll keep my fingers crossed that we don't have a "global" problem here...