Eric, News, Plans Please

Author	Message
Winterknight Volunteer tester Send message Joined: 15 Jun 05 Posts: 709 Credit: 5,834,108 RAC: 0	Message 23572 - Posted: 24 May 2007, 16:32:43 UTC Eric, Is there any chance you could give us an update on Multi-Beam and AP, and give us some idea of your plans for the near future. I assume from what has been said on the Main Site, that Multi-Beam will have to be introduced there soon. So can we expect a new pre-release version here? If so do how long do you plan to run it, assuming that there are no unforseen problems? I ask this as due to AP units, mine, and I'm sure everybody elses RDCF has sky rocketed. Mine were below 0.5 when doing Multi-beam only. Now both machines are about 5.0. Also because of the server problems, the AP units have had a clear run on the CPU and Beta now has very large short and long term debts. Plus the AP unit I still have left has been and is going to be in EDF, or whatever JM7 calls it these days, mode, further delaying any Beta downloads. So for my computers, at least, they are are not likely to download any Beta work for a week or two without intervention. Andy ID: 23572 ·

Winterknight Volunteer tester Send message Joined: 15 Jun 05 Posts: 709 Credit: 5,834,108 RAC: 0	Message 23831 - Posted: 28 May 2007, 10:52:44 UTC Further thoughts on MultiBeam, Have you any plans to increase sensitivity again, or any other methods of increasing the processing time? Other methods, larger units. The size of units, 350KByte, has been constant since beginning of Classic, when we were mainly on dial-up. As most of us, that can, are on broadband, is it possible to increase unit size and processing time? Reasons I ask, Seti-centric people with latest dual core 2 Xeons are already claiming that 4 cpu and 100 units/cpu are being reached. Personally I think this could only happen on the most extreme of days, but the next generation of cpu's, 3rd or 4th qtr 2007, will probably reach this limit. Andy ID: 23831 ·

zombie67 [MM] Volunteer tester Send message Joined: 18 May 06 Posts: 280 Credit: 26,477,429 RAC: 0	Message 23842 - Posted: 28 May 2007, 14:12:12 UTC - in response to Message 23831. Last modified: 28 May 2007, 14:12:46 UTC Reasons I ask, Seti-centric people with latest dual core 2 Xeons are already claiming that 4 cpu and 100 units/cpu are being reached. Personally I think this could only happen on the most extreme of days, but the next generation of cpu's, 3rd or 4th qtr 2007, will probably reach this limit. I have been hitting the limit frequently in recent days. But to be clear, my machine is a dual quad-core (8-way), not dual dual-core. If I understand correctly, the limit is 100 per core per day, with a total limit of 400. That works for machines with up to 4 cores. With an 8-way machine, that really means 50/core/day. And if you get a bunch of the little WUs (like we have been lately), it is very easy to hit the limit. Dublin, California Team: SETI.USA ID: 23842 ·

JKeck {pirate} Volunteer tester Send message Joined: 15 Jun 05 Posts: 69 Credit: 1,172,804 RAC: 0	Message 23880 - Posted: 29 May 2007, 5:42:54 UTC - in response to Message 23842. Reasons I ask, Seti-centric people with latest dual core 2 Xeons are already claiming that 4 cpu and 100 units/cpu are being reached. Personally I think this could only happen on the most extreme of days, but the next generation of cpu's, 3rd or 4th qtr 2007, will probably reach this limit. I have been hitting the limit frequently in recent days. But to be clear, my machine is a dual quad-core (8-way), not dual dual-core. If I understand correctly, the limit is 100 per core per day, with a total limit of 400. That works for machines with up to 4 cores. With an 8-way machine, that really means 50/core/day. And if you get a bunch of the little WUs (like we have been lately), it is very easy to hit the limit. A change was just checked in today (5/29/7) to make the number of cores limit configurable by the projects with the default at 8 cores. So the next time projects update their server code you should start seeing larger limits. BOINC WIKI BOINCing since 2002/12/8 ID: 23880 ·

Bob Guy Volunteer tester Send message Joined: 17 Jan 06 Posts: 114 Credit: 194,623 RAC: 0	Message 23886 - Posted: 29 May 2007, 7:53:14 UTC - in response to Message 23842. I have been hitting the limit frequently in recent days. Darn it, anyway! I would love to have your problem, you've got a great machine as well. I thought we were out of AP WUs (the loooooong ones). I was just finishing my last one, it's been running for 4..5 days, and so I thought I could then fix my DCF and start running some short 5.17 WUs because I need the credit! So anyway, the AP WU finished and what do I get? Darn it! Two more AP WUs, just my luck I guess. My current complaint (it's not a really big deal) is that the pending credit (especially for the AP WUs) is getting to be ridiculous, I've got 9000 pending credits on my one computer. It's taking a month or more for some of the AP WUs to get finished and returned by many users. I guess that's to be expected for Beta so don't go get all excited about my ranting. ID: 23886 ·

Demiurg Volunteer tester Send message Joined: 14 Apr 07 Posts: 506 Credit: 15,684 RAC: 0	Message 23895 - Posted: 29 May 2007, 10:28:03 UTC - in response to Message 23886. I have been hitting the limit frequently in recent days. Darn it, anyway! I would love to have your problem, you've got a great machine as well. I thought we were out of AP WUs (the loooooong ones). I was just finishing my last one, it's been running for 4..5 days, and so I thought I could then fix my DCF and start running some short 5.17 WUs because I need the credit! So anyway, the AP WU finished and what do I get? Darn it! Two more AP WUs, just my luck I guess. My current complaint (it's not a really big deal) is that the pending credit (especially for the AP WUs) is getting to be ridiculous, I've got 9000 pending credits on my one computer. It's taking a month or more for some of the AP WUs to get finished and returned by many users. I guess that's to be expected for Beta so don't go get all excited about my ranting. But just imagine what it will do for your RAC if all of your pending AP WUs punched in at the same day... Whoa! Carl ID: 23895 ·

Keith T. Volunteer tester Send message Joined: 9 Feb 07 Posts: 130 Credit: 27,850 RAC: 0	Message 23911 - Posted: 29 May 2007, 12:21:30 UTC - in response to Message 23886. I have been hitting the limit frequently in recent days. Darn it, anyway! I would love to have your problem, you've got a great machine as well. I thought we were out of AP WUs (the loooooong ones). I was just finishing my last one, it's been running for 4..5 days, and so I thought I could then fix my DCF and start running some short 5.17 WUs because I need the credit! So anyway, the AP WU finished and what do I get? Darn it! Two more AP WUs, just my luck I guess. My current complaint (it's not a really big deal) is that the pending credit (especially for the AP WUs) is getting to be ridiculous, I've got 9000 pending credits on my one computer. It's taking a month or more for some of the AP WUs to get finished and returned by many users. I guess that's to be expected for Beta so don't go get all excited about my ranting. I wish I could get at least 1 Astropulse WU, the server thinks I don't have enough memory: SETI@home Beta Test\|Message from server: Your computer has 255.48MB of memory, and a job requires 256.00MB I have done 1 previous AP task about a month ago with ver 4.12: http://setiweb.ssl.berkeley.edu/beta/result.php?resultid=1691738 http://setiweb.ssl.berkeley.edu/beta/workunit.php?wuid=437679 I thought I was still waiting for credit for this one, but I've just noticed on the result page it has been granted, but not on the WU page. I beleive the requirement is 144MB, would it be possible to configure the server so that computers with 256MB or 192MB could get AP tasks? ID: 23911 ·

Demiurg Volunteer tester Send message Joined: 14 Apr 07 Posts: 506 Credit: 15,684 RAC: 0	Message 23914 - Posted: 29 May 2007, 12:44:20 UTC - in response to Message 23911. I have been hitting the limit frequently in recent days. Darn it, anyway! I would love to have your problem, you've got a great machine as well. I thought we were out of AP WUs (the loooooong ones). I was just finishing my last one, it's been running for 4..5 days, and so I thought I could then fix my DCF and start running some short 5.17 WUs because I need the credit! So anyway, the AP WU finished and what do I get? Darn it! Two more AP WUs, just my luck I guess. My current complaint (it's not a really big deal) is that the pending credit (especially for the AP WUs) is getting to be ridiculous, I've got 9000 pending credits on my one computer. It's taking a month or more for some of the AP WUs to get finished and returned by many users. I guess that's to be expected for Beta so don't go get all excited about my ranting. I wish I could get at least 1 Astropulse WU, the server thinks I don't have enough memory: SETI@home Beta Test\|Message from server: Your computer has 255.48MB of memory, and a job requires 256.00MB I have done 1 previous AP task about a month ago with ver 4.12: http://setiweb.ssl.berkeley.edu/beta/result.php?resultid=1691738 http://setiweb.ssl.berkeley.edu/beta/workunit.php?wuid=437679 I thought I was still waiting for credit for this one, but I've just noticed on the result page it has been granted, but not on the WU page. I beleive the requirement is 144MB, would it be possible to configure the server so that computers with 256MB or 192MB could get AP tasks? Isn't it actually easier if you looked for a used memory-board instead? a pair of 256MBs are possible to get for 20-30USD nowadays. ID: 23914 ·

Keith T. Volunteer tester Send message Joined: 9 Feb 07 Posts: 130 Credit: 27,850 RAC: 0	Message 23919 - Posted: 29 May 2007, 13:38:00 UTC - in response to Message 23914. I have been hitting the limit frequently in recent days. Darn it, anyway! I would love to have your problem, you've got a great machine as well. I thought we were out of AP WUs (the loooooong ones). I was just finishing my last one, it's been running for 4..5 days, and so I thought I could then fix my DCF and start running some short 5.17 WUs because I need the credit! So anyway, the AP WU finished and what do I get? Darn it! Two more AP WUs, just my luck I guess. My current complaint (it's not a really big deal) is that the pending credit (especially for the AP WUs) is getting to be ridiculous, I've got 9000 pending credits on my one computer. It's taking a month or more for some of the AP WUs to get finished and returned by many users. I guess that's to be expected for Beta so don't go get all excited about my ranting. I wish I could get at least 1 Astropulse WU, the server thinks I don't have enough memory: SETI@home Beta Test\|Message from server: Your computer has 255.48MB of memory, and a job requires 256.00MB I have done 1 previous AP task about a month ago with ver 4.12: http://setiweb.ssl.berkeley.edu/beta/result.php?resultid=1691738 http://setiweb.ssl.berkeley.edu/beta/workunit.php?wuid=437679 I thought I was still waiting for credit for this one, but I've just noticed on the result page it has been granted, but not on the WU page. I beleive the requirement is 144MB, would it be possible to configure the server so that computers with 256MB or 192MB could get AP tasks? Isn't it actually easier if you looked for a used memory-board instead? a pair of 256MBs are possible to get for 20-30USD nowadays. I do have a spare 256MB, but it is SDRAM and the 256MB which is fitted is DDR. My MoBo is an ASRock K7VT2 which has 2 DDR slots and 2 SDRAM slots, but the manual says you can only fit 1 type or the other, not both. Any opinions on which type is better to get? ID: 23919 ·

Demiurg Volunteer tester Send message Joined: 14 Apr 07 Posts: 506 Credit: 15,684 RAC: 0	Message 23969 - Posted: 30 May 2007, 9:44:31 UTC - in response to Message 23919. I have been hitting the limit frequently in recent days. Darn it, anyway! I would love to have your problem, you've got a great machine as well. I thought we were out of AP WUs (the loooooong ones). I was just finishing my last one, it's been running for 4..5 days, and so I thought I could then fix my DCF and start running some short 5.17 WUs because I need the credit! So anyway, the AP WU finished and what do I get? Darn it! Two more AP WUs, just my luck I guess. My current complaint (it's not a really big deal) is that the pending credit (especially for the AP WUs) is getting to be ridiculous, I've got 9000 pending credits on my one computer. It's taking a month or more for some of the AP WUs to get finished and returned by many users. I guess that's to be expected for Beta so don't go get all excited about my ranting. I wish I could get at least 1 Astropulse WU, the server thinks I don't have enough memory: SETI@home Beta Test\|Message from server: Your computer has 255.48MB of memory, and a job requires 256.00MB I have done 1 previous AP task about a month ago with ver 4.12: http://setiweb.ssl.berkeley.edu/beta/result.php?resultid=1691738 http://setiweb.ssl.berkeley.edu/beta/workunit.php?wuid=437679 I thought I was still waiting for credit for this one, but I've just noticed on the result page it has been granted, but not on the WU page. I beleive the requirement is 144MB, would it be possible to configure the server so that computers with 256MB or 192MB could get AP tasks? Isn't it actually easier if you looked for a used memory-board instead? a pair of 256MBs are possible to get for 20-30USD nowadays. I do have a spare 256MB, but it is SDRAM and the 256MB which is fitted is DDR. My MoBo is an ASRock K7VT2 which has 2 DDR slots and 2 SDRAM slots, but the manual says you can only fit 1 type or the other, not both. Any opinions on which type is better to get? Evil idea, sometimes manuals say things like that since they during development have discovered that doing something causes mystifying problem X when doing Y, but otherwise it works. Testing something like using both types of memory very rarely breaks anything permanently. So I am 99.9 percent shure that the worst thing that will happen if you try is that the computer wont start up with both, but will do so when you remove one. But there is a fairly large chance you will see your comp counting up 512MB... See it as a beta test:-) Other idea is to see if you can swap with someone. There is probably someone around with the exact opposite problem needing an SDRAM and has a DDR. Carl ID: 23969 ·

Pappa Volunteer tester Send message Joined: 13 Nov 05 Posts: 1724 Credit: 3,121,901 RAC: 0	Message 24029 - Posted: 30 May 2007, 18:58:02 UTC I recieved a reply from Eric... Yes I know he has been busy... One thing that you should notice shortly is that Your Pending Credit has dropped and you should have credit for the bulk of the AP Results... Eric managed fit into "all those things happening" a working AP Validator... Josh is looking at the AP Results return thus far, scratching his head and trying to make some sense out of what he is looking at... Then what adjustments can be made and what has to be done... Eric does have the proposed changes from Joe towards 5.18... Getting them incorporated is high the list of things to get done... A couple of other things to be noted are if you are having problem with contacting the Beta Server, You might need to shutdown BOINC and Restart or Restart the Computer... It would appear that BOINC Collected some trash it refuses to get rid of even though the Servers are Up and Running... A reminder about the broken Server Status Page... There are enough things going on that it will probably be a while before it is fixed. Pappa Thanks to Paul and Friends Please consider a Donation to the Seti Project ID: 24029 ·

Winterknight Volunteer tester Send message Joined: 15 Jun 05 Posts: 709 Credit: 5,834,108 RAC: 0	Message 24035 - Posted: 30 May 2007, 19:21:50 UTC - in response to Message 24029. I recieved a reply from Eric... Yes I know he has been busy... One thing that you should notice shortly is that Your Pending Credit has dropped and you should have credit for the bulk of the AP Results... Eric managed fit into "all those things happening" a working AP Validator... Josh is looking at the AP Results return thus far, scratching his head and trying to make some sense out of what he is looking at... Then what adjustments can be made and what has to be done... Eric does have the proposed changes from Joe towards 5.18... Getting them incorporated is high the list of things to get done... A couple of other things to be noted are if you are having problem with contacting the Beta Server, You might need to shutdown BOINC and Restart or Restart the Computer... It would appear that BOINC Collected some trash it refuses to get rid of even though the Servers are Up and Running... A reminder about the broken Server Status Page... There are enough things going on that it will probably be a while before it is fixed. Pappa Pappa, Thanks for passing on the news. It is appreciated. One other small thing that needs looking at is the bad exit on some windows computers after finding 30 pulses. This may not be an AP only error as it is seen on Multi-Beam and in the standard app Enhanced app (I have not seen it in optimised app). This AP unit 461257 is an example. Three hosts exited gracefully and Validated, 4 host exitted with - exit code -1073741819 (0xc0000005). All 7 hosts found 30 pulses and all took over 120k sec. Andy ID: 24035 ·

Pappa Volunteer tester Send message Joined: 13 Nov 05 Posts: 1724 Credit: 3,121,901 RAC: 0	Message 24045 - Posted: 30 May 2007, 22:52:25 UTC - in response to Message 24035. Andy, et al If we can get them catalogged I will run them past Eric Pappa, Thanks for passing on the news. It is appreciated. One other small thing that needs looking at is the bad exit on some windows computers after finding 30 pulses. This may not be an AP only error as it is seen on Multi-Beam and in the standard app Enhanced app (I have not seen it in optimised app). This AP unit 461257 is an example. Three hosts exited gracefully and Validated, 4 host exitted with - exit code -1073741819 (0xc0000005). All 7 hosts found 30 pulses and all took over 120k sec. Andy Thanks to Paul and Friends Please consider a Donation to the Seti Project ID: 24045 ·

Misfit Volunteer tester Send message Joined: 15 Jun 05 Posts: 3400 Credit: 1,026,406 RAC: 0	Message 24048 - Posted: 31 May 2007, 3:02:17 UTC - in response to Message 24029. You missed one me@rescam.org ID: 24048 ·

keyboards Volunteer tester Send message Joined: 11 Feb 06 Posts: 309 Credit: 4,022,545 RAC: 0	Message 24087 - Posted: 31 May 2007, 13:39:25 UTC Last modified: 31 May 2007, 13:44:11 UTC Another one missed (host 7966 on this one). A different issue on This one . No credit granted / check skipped (host 4872) even though it exited gracefully with 30 pulses. *!!Stupidity should be PAINFUL!!* ID: 24087 ·

Winterknight Volunteer tester Send message Joined: 15 Jun 05 Posts: 709 Credit: 5,834,108 RAC: 0	Message 24098 - Posted: 31 May 2007, 21:03:01 UTC - in response to Message 24087. Another one missed (host 7966 on this one). This one errored out with the - exit code -1073741819 (0xc0000005), which I talked about earlier. A different issue on This one . No credit granted / check skipped (host 4872) even though it exited gracefully with 30 pulses. This ones different though, it probably didn't validate because of 'too many total results'. Which will need manual intervention to grant credits, which I believe has not been done yet. Andy ID: 24098 ·

Misfit Volunteer tester Send message Joined: 15 Jun 05 Posts: 3400 Credit: 1,026,406 RAC: 0	Message 24105 - Posted: 1 Jun 2007, 0:31:38 UTC And another Of course this was uploaded and reported today - instantly awarded zero. me@rescam.org ID: 24105 ·

Pappa Volunteer tester Send message Joined: 13 Nov 05 Posts: 1724 Credit: 3,121,901 RAC: 0	Message 24110 - Posted: 1 Jun 2007, 0:48:01 UTC Sorry Folks I should have been looking, "we" have very many reports that are Ap validate issues... Eric got this working inbetween various outage things and something he is doing in the background. I created a thread in Astropulse for the links to these issues. Validate Errors with AP Please repost there so I can point Eric at them... Thanks to Paul and Friends Please consider a Donation to the Seti Project ID: 24110 ·

Wander Saito Volunteer tester Send message Joined: 6 Feb 06 Posts: 105 Credit: 188,885 RAC: 0	Message 24116 - Posted: 1 Jun 2007, 2:01:33 UTC - in response to Message 24029. A reminder about the broken Server Status Page... There are enough things going on that it will probably be a while before it is fixed. Pappa So I guess this nearly-all-green status page is an "optical illusion"? :) Too bad, I don't remember when was the last time I saw it like that. Regards, Wander ID: 24116 ·

Fuzzy Hollynoodles Volunteer tester Send message Joined: 17 Jun 05 Posts: 2087 Credit: 257,732 RAC: 0	Message 24239 - Posted: 2 Jun 2007, 18:18:47 UTC - in response to Message 24105. And another Of course this was uploaded and reported today - instantly awarded zero. Mine as well. :-/ [b]"I'm trying to maintain a shred of dignity in this world." - Me[/b] ID: 24239 ·

©2025 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.