Posts by Alan Ng

1) Message boards : Cafe SETI : Security camera records 'unexplainable' object (Message 592502) Posted 25 Jun 2007 by Alan Ng Post: The apparent "shadow" that follows the light as well as the "spinning motion" of the light itself look awfully like your typical, normal compression artifacts when you take any digital image and compress it highly to save space on disk, and as additionally happens with all Youtube videos regardless of the original video format.
2) Questions and Answers : Preferences : BOINC 5.10.7 not using General Preferences (Message 590702) Posted 22 Jun 2007 by Alan Ng Post: I found another thread reporting same problem: http://boinc.berkeley.edu/dev/forum_thread.php?id=1899
3) Questions and Answers : Preferences : BOINC 5.10.7 not using General Preferences (Message 590389) Posted 22 Jun 2007 by Alan Ng Post: In case it's useful, here's an example of the start-up messages I get when I restart BOINC 5.10.7: Fri Jun 22 09:57:52 2007\|\|Starting BOINC client version 5.10.7 for powerpc-apple-darwin Fri Jun 22 09:57:52 2007\|\|log flags: task, file_xfer, sched_ops Fri Jun 22 09:57:52 2007\|\|Libraries: libcurl/7.16.1 OpenSSL/0.9.7l zlib/1.2.3 Fri Jun 22 09:57:52 2007\|\|Data directory: /Library/Application Support/BOINC Data Fri Jun 22 09:57:52 2007\|SETI@home\|Found app_info.xml; using anonymous platform Fri Jun 22 09:57:52 2007\|\|Processor: 1 Power Macintosh Power Macintosh [Power Macintosh Model PowerMac10,1] [AltiVec] Fri Jun 22 09:57:52 2007\|\|Processor features: Fri Jun 22 09:57:52 2007\|\|Memory: 1.00 GB physical, 39.68 GB virtual Fri Jun 22 09:57:52 2007\|\|Disk: 74.41 GB total, 39.44 GB free Fri Jun 22 09:57:52 2007\|Einstein@Home\|URL: http://einstein.phys.uwm.edu/; Computer ID: 335019; location: home; project prefs: default Fri Jun 22 09:57:52 2007\|SETI@home\|URL: http://setiathome.berkeley.edu/; Computer ID: 1306287; location: home; project prefs: default Fri Jun 22 09:57:52 2007\|\|No general preferences found - using BOINC defaults Fri Jun 22 09:57:52 2007\|\|Preferences limit memory usage when active to 512.00MB Fri Jun 22 09:57:52 2007\|\|Preferences limit memory usage when idle to 921.60MB Fri Jun 22 09:57:52 2007\|\|Preferences limit disk usage to 9.31GB Fri Jun 22 09:57:53 2007\|Einstein@Home\|Restarting task h1_0488.35_S5R2__321_S5R2c_1 using einstein_S5R2 version 420
4) Questions and Answers : Preferences : BOINC 5.10.7 not using General Preferences (Message 590008) Posted 21 Jun 2007 by Alan Ng Post: On all of my machines (Win XP on P4, Intel Mac on Xeon, G4 Mac) the new BOINC 5.10.7 starts up and persists in saying "No general preferences found - using BOINC defaults" I double-checked that each machine is not using local preferences, including clicking the new "Clear" button in the new Boinc Manager - Preferences dialog. BOINC proceeds to operate without using the Web-based General Prefs I have always (and still) have set, even after using the Project Update feature to fetch prefs. Example: After installing the new BOINC, my Intel Mac started using 4 CPU cores instead of 3 as before and as dictated in my BOINC General Prefs in the Web interface of setiathome.berkeley.edu. Am I the only one having this problem?
5) Message boards : Technical News : Oopsie (Apr 11 2007) (Message 545080) Posted 12 Apr 2007 by Alan Ng Post: Looks like it's fixed here. Same here! I used Boinc Manager, Transfers tab, to "Retry" the stuck download, and now (unlike previously), the download succeeds, and shortly thereafter, this machine finally began requesting and getting new WUs. Thanks, Matt.
6) Message boards : Technical News : Oopsie (Apr 11 2007) (Message 545056) Posted 12 Apr 2007 by Alan Ng Post: Anecdotal evidence to support Arthur L. Smith's hypothesis: all of three my SETI-crunching machines, in geographically different locations and networks, with different processors, all have the same issues. In a few hours, all of my SETI machines will be doing no work for the SETI project because they're all unable to download workunits. Each machine is about to run out, or has already run out, of successfully-downloaded WUs. All machines have been reporting the following type of msg for some WUs since yesterday, and are not downloading new WU: Thu Apr 12 13:49:40 2007\|SETI@home\|[file_xfer] Temporarily failed download of 18se04aa.24984.27954.117328.3.8: file not found p.s. my BOINC manager version (all machines) is 5.8.17, all Mac OS 10.4.
7) Message boards : Number crunching : Computation errors on Seti Enhanced (Message 327652) Posted 5 Jun 2006 by Alan Ng Post: So far averaging 3 out 4 successful, but 1 out of 4 crash with exit code 250. This is happening on both of my machines (two different kinds of Macs, both running latest OS 10.4.6 and standard BOINC client 5.4.9). For details see: http://setiathome.berkeley.edu/results.php?userid=3337587
8) Questions and Answers : Macintosh : How do you install an older version of BOINC over a newer one? (Message 170131) Posted 20 Sep 2005 by Alan Ng Post: Hmm, I should clarify my question: How do I install the lower-numbered version without losing all my settings and history on this machine? Before I started this thread I had already tried deleting the main BOINCManager app file before running the 4.43 installation, but that didn't help - I still got "Nothing to install." The only other files I know about to delete would be those in the /Library/Application Support/BOINC Data/ folder. Deleting that would certainly erase all of the history for the statistics graphs. I realize that would be just a cosmetic loss, but surely there must be a way to install a lower-numbered version of the app file without having to delete all of those data files, too. Maybe there are just particular files in that data folder which one should delete, instead of "all files"?
9) Questions and Answers : Macintosh : How do you install an older version of BOINC over a newer one? (Message 170059) Posted 20 Sep 2005 by Alan Ng Post: I just tried out a beta version of BOINC (5.1.2) and now I would like to go back to the current Recommended (4.43). But when I attempt to install 4.43 the install process stops at the final step with a "Nothing to install" message. How does one install an older version of the BOINCManager GUI? I'm running OS 10.4.2.
10) Message boards : SETI@home Science : Are we doing science? Hasn't all this 2003/2004 data already been processed? (Message 167973) Posted 15 Sep 2005 by Alan Ng Post: (this is a remark from the original poster of this question) Thanks for the various responses. Some of them sound a bit defensive, which I assume comes not from my question, but from your past experiences with (aggressive) non-scientists who notice the odd chronology of the data. Let me admit that I posted my question partly because I had seen "in print" in (multiple) online sources over recent months rumors that the Seti@Home project was simply feeding "used" data tapes to the network without a good reason. I didn't mention these sources for the simple reason that I don't recall where that was (probably some BOINC-related online forum or other). I do recall running into comments about users preferring "fresh" data. In any case, upon searching this site for fifteen minutes or so, I found no explanation to the admittedly fair question about the backwards chronological order of the data, especially fair given that over the past (and my first) year of helping SETI the dates of the WUs had been slowly but consistently moving forwards, gradually catching up to the present. So: I suggest politely to S@H staff that rather than spend your time answering questions like this over and over, that some brief explanation of how you select the data to feed - particularly addressing the chronological-order question - be posted someplace more centrally and permanently than this message-board thread. Trust me: as a university historian and a Berkeley physics graduate myself, I have no misconceptions about the scientific value of "old" data! My question was whether "used" data was being run again, out of perhaps a lack of unprocessed data being available ... or for whatever fictitious reasons had/have been suggested elsewhere.
11) Message boards : SETI@home Science : Are we doing science? Hasn't all this 2003/2004 data already been processed? (Message 167227) Posted 13 Sep 2005 by Alan Ng Post: This question might sound aggressive, but its intent is friendly - I'm genuinely interested in doing SETI science. I'm curious why we have all been working on old SETI data tapes from 2003/2004 over the recent weeks/months. (I noticed SETI Classic has had at least some early 2005 data more recently). Are we all doing actual astronomy, or is all this communal, kilowatt-burning computing power just spinning its wheels in order to earn "credits" and in order to help SETI staff explore the - admittedly interesting and possibly useful - engineering problems of shuffling large amounts of data from left to right and back to left?
12) Questions and Answers : Windows : BOINC 4.32 fails to suspend a project when switching (Message 114660) Posted 25 May 2005 by Alan Ng Post: I'm having the same issue on a Win98SE machine using BOINC 4.43. It was not a problem on this machine in BOINC 4.25 and earlier. This machine is running S@H, Predictor, and E@H.
13) Message boards : Number crunching : When is Seti classic going away (Message 87638) Posted 18 Mar 2005 by Alan Ng Post: Regarding the larger question: why do people want to run BOINC on ancient machines? I, for example, have an old extra laptop (Pentium 120 MHz, 32 MB) which I leave in the basement, running constantly, with a wireless network card, so I have easy Web access for various trivial household reasons. Doesn't it make sense to have that computer doing something useful when not being used? Currently it's running SETI Classic quite happily. I suggest that this project send out advice to SETI Classic users who have archaic machines, suggesting projects (hopefully SETI Boinc) where these machines can be put to use. Similarly, I suggest reducing or removing specific minimum hardware requirements for SETI/Boinc as far as possible (eventually). As mentioned in other posts in this thread, 64 MB seems arbitrary and unnecessarily high. Just a thought.
14) Questions and Answers : Windows : BOINC 4.19 not asking for more work when it has no work. (Message 76819) Posted 5 Feb 2005 by Alan Ng Post: UPDATE - problem fixed. I went into General Preferences for this machine on the SETI server and noticed that "Use % of virtual memory" was blank. I believe this is a new preference that must have been added recently, thus no value would have been possible to inherit from my older preference. Once I filled in a value (my first try of 50% was sufficient) BOINC finally realized it had insufficient work. Perhaps this is a useful bug report insofar as the "% of virtual memory" preference should have a default value filled in by BOINC instead of a blank. Here's a full, complete transcript after a clean reboot of the machine and before I fixed things as per above report. ProteinPredictorAtHome - 2005-02-05 00:36:36 - Project prefs: no separate prefs for home; using your defaults SETI@home - 2005-02-05 00:36:36 - Host ID is 13949 ProteinPredictorAtHome - 2005-02-05 00:36:36 - Host ID is 31527 --- - 2005-02-05 00:36:36 - General prefs: from unknown project http://climateprediction.net/ (last modified 2005-02-03 15:49:42) --- - 2005-02-05 00:36:36 - General prefs: using separate prefs for home --- - 2005-02-05 01:04:01 - Suspending computation and network activity - user request --- - 2005-02-05 01:04:05 - Resuming computation and network activity --- - 2005-02-05 01:04:26 - Running CPU benchmarks --- - 2005-02-05 01:04:26 - Suspending computation and network activity - running CPU benchmarks --- - 2005-02-05 01:05:27 - Benchmark results: --- - 2005-02-05 01:05:27 - Number of CPUs: 1 --- - 2005-02-05 01:05:27 - 418 double precision MIPS (Whetstone) per CPU --- - 2005-02-05 01:05:27 - 944 integer MIPS (Dhrystone) per CPU --- - 2005-02-05 01:05:27 - Finished CPU benchmarks --- - 2005-02-05 01:05:28 - Resuming computation and network activity ProteinPredictorAtHome - 2005-02-05 01:07:00 - Sending request to scheduler: http://predictor1.scripps.edu/predictor_cgi/cgi ProteinPredictorAtHome - 2005-02-05 01:07:03 - Scheduler RPC to http://predictor1.scripps.edu/predictor_cgi/cgi succeeded ProteinPredictorAtHome - 2005-02-05 01:07:03 - Project prefs: no separate prefs for home; using your defaults SETI@home - 2005-02-05 10:34:29 - Sending request to scheduler: http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi SETI@home - 2005-02-05 10:34:33 - Scheduler RPC to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi succeeded SETI@home - 2005-02-05 10:34:33 - Project prefs: using separate prefs for home
15) Questions and Answers : Windows : BOINC 4.19 not asking for more work when it has no work. (Message 76740) Posted 5 Feb 2005 by Alan Ng Post: I'm participating in both SETI and Predictor. My BOINC client (4.19), running on Win98SE, 128MB RAM, >1GB free disk space, finished all its work units over a day ago, and did not ask for any more. I rebooted a few times. Still didn't ask for any work. I used "update" and it talked to both the SETI and Predictor schedulers just fine, but didn't request any work. Then I "reset" both projects. No change. I ran benchmarks - no change. 4.19 was running fine on both projects for a few weeks up to now (or whenever it was that 4.19 was posted on Predictor). Any advice?
16) Questions and Answers : Wish list : Mac OS X Jaguar CLI (Message 19387) Posted 29 Aug 2004 by Alan Ng Post: Same wish all over again, but now for Boinc 4 on Jaguar. My Powerbook has lots of hours to donate ... please let it do so.
17) Questions and Answers : Macintosh : How do I run it? (Message 780) Posted 23 Jun 2004 by Alan Ng Post: Ok, StuffIt automatically gunzipped it, giving a file which I moved to Applications and renamed "boincSETI." I then did the following: - run terminal - go to Applications folder (using cd command) - type in "chmod +x boincSETI" which seems to work fine - type "boincSETI" I get the response "boincSETI: command not found." Uh, what am I missing. A step-by-step writeup for MacOSX newbies would be excellent. I only know the Windows world.

©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.