Results 1 to 17 of 17

Thread: Client test version #2 available

  1. #1

    Client test version #2 available

    A second test release of the client is now available from the ftp site: ftp.mshri.on.ca/pub/distribfold/download. The versions available are just for Windows and Linux, respectively named foldtrajlite.zip and foldtrajlite.gz

    Get ftp://ftp.mshri.on.ca/pub/distribfol...ldtrajlite.zip Windows version or ftp://ftp.mshri.on.ca/pub/distribfol...oldtrajlite.gz Linux version (RedHat 8.0 and up).

    Simply replace the executable in the existing installation of the client or test version (make sure to keep a backup version), and please delete filelist.txt to force it to start over at generation zero.

    Changes from the last test version, mostly arising from discussions between FEEDB0B0 and I, are again to the energy function - it now focuses on getting the structure compact first, and then worries about pseudo-energy values. We shall see if this gets us any closer to the folded state. Again the main purpose of pre-releasing this is that we would appreciate detailed feedback about any bugs. Thanks to all those who volunteer their time to test! If all goes well, we will be releasing essentially the same program to everyone as an official update, this coming Thursday.
    Howard Feldman

  2. #2
    Member
    Join Date
    Apr 2003
    Location
    Germany
    Posts
    59
    I'm just downloading it

  3. #3
    Target Butt IronBits's Avatar
    Join Date
    Dec 2001
    Location
    Morrisville, NC
    Posts
    8,619
    9 hrs 45 minutes, 7.6 million page faults, still running.

  4. #4
    OCworkbench Stats Ho
    Join Date
    Jan 2003
    Posts
    519
    All looking good so far, no hangs, no errors, no crashes. Very stable on my Dual 2400 MP Win2k.
    I am not a Stats Ho, it is just more satisfying to see that my numbers are better than yours.

  5. #5
    I installed the new test client #2 on 3 machines. Deleted thr filelist.txt, and started at zero. After 4 hours one machine is up to 28 generations, while another can't finish the "10K run". It gets to 9546/454 or close to those numbers, and then reverts back to 8541/1459. I watched it do this countless times. The error log has not been updated since I started the client and shows no problems...

    Within an hour the 3ed machine is running fine...
    Too many computers, too little time......

  6. #6
    let it run on my home cpu all last night, completely stable.
    im going to put it on my work computers and see what happens, today.
    Driving home the sky accelerates and the clouds all form a geometric shape.

  7. #7
    Electric fence operator
    Join Date
    Dec 2003
    Location
    Indianapolis, IN
    Posts
    379
    Are we back to 50 structures per generation (and if so, is it permanent, and 100 was just temporary), or did I make a mistake in replacing the foldtrajlite file? This is possible, as I was only partially awake when doing so.
    Beyond that, I've not encountered any bugs with whatever version I'm running, and I made it through gen. 0 (and now 1) without problems.

  8. #8
    so is the 2nd test client any faster compare to the 1st ?

  9. #9
    Electric fence operator
    Join Date
    Dec 2003
    Location
    Indianapolis, IN
    Posts
    379
    Sorry, double post

  10. #10
    I just got this error:

    ========================[ Jan 9, 2000 10:16 PM ]========================
    Starting foldtrajlite built 2003.12.12
    Sun Jan 9 22:16:49 2000 FATAL ERROR: [065.017] {asnio.c, line 1101} XXXXXXXX_protein_147.trjInput
    TGraph.<datafpt>
    Premature end of file


    It ran fine for 147 structures with -rt and -it options on mandrake 9.1. I unpluged the network cable to upload from my laptop. It buffered 2 generations before the network cable was pluged in. I restarted the client to upload and got this error.

  11. #11
    Boinc'ing away
    Join Date
    Aug 2002
    Location
    London, UK
    Posts
    982
    Got this error:

    Wed Dec 17 06:52:46 2003 ERROR: [001.001] {bbox.c, line 445} Tried to BDRemove non-existent (7.65458,24.01711,2.24812)
    Client died without removing .lock or progress.txt...I can upload a copy of the directory if required

    Client was running on Windows 2000, extra RAM, as a service.

  12. #12
    Senior Member
    Join Date
    Jul 2002
    Location
    Kodiak, Alaska
    Posts
    432
    Anteraan: If you take a quick look at your error.log file, it'll mention the date the foldtrajlite.exe file was created. The latest beta should say, "Starting foldtrajlite built Dec 12 2003" It seems to be making 100 structure generations, just like the first beta.

    I let dfGUI run last night for awhile - (from gen 61 to 128) and noticed the client was now hitting a minimum energy reading at 29.620. The value showed up on gen 80 and then from gen 92 through gen 104. I mention this because I assumed that the minimum energy reading was going to be lowered from the 1st beta.
    www.thegenomecollective.com
    Borging.. it's not just an addiction. It's...

  13. #13
    Ol' retired IT geezer
    Join Date
    Feb 2003
    Location
    Scarborough
    Posts
    92

    2nd Test Still Does Plateaus



    I'm not sure if the energy level gets lower in general with this release, but the plateau effect still is in force... As the above graph from dfGUI shows...

    Ned

  14. #14
    Originally posted by dano
    I just got this error:

    ========================[ Jan 9, 2000 10:16 PM ]========================
    Starting foldtrajlite built 2003.12.12
    Sun Jan 9 22:16:49 2000 FATAL ERROR: [065.017] {asnio.c, line 1101} XXXXXXXX_protein_147.trjInput
    TGraph.<datafpt>
    Premature end of file


    It ran fine for 147 structures with -rt and -it options on mandrake 9.1. I unpluged the network cable to upload from my laptop. It buffered 2 generations before the network cable was pluged in. I restarted the client to upload and got this error.
    Take a look at the size of the file mentioned - is it zero, or significantly smaller than other .trj files in the directory? Also make sure you have sufficient disk space and write permissions.
    It also seems, as a side note, that your clock settings are almost 4 years behind or so (look at the date stamp of the error).
    Elena Garderman

  15. #15
    Originally posted by pfb
    Got this error:



    Client died without removing .lock or progress.txt...I can upload a copy of the directory if required

    Client was running on Windows 2000, extra RAM, as a service.
    We are looking into this one, the directory is not needed in this case, but thank you for keeping it handy as backup - always a good idea!
    Elena Garderman

  16. #16
    OCworkbench Stats Ho
    Join Date
    Jan 2003
    Posts
    519
    We are still seeing filelst.txt completely vanishing from the hard drive without a trace Usually after a blackout shuts the PC down. Also, the test Client seems more vulnerable to filelst corruption after a blackout or net loss during upload
    I am not a Stats Ho, it is just more satisfying to see that my numbers are better than yours.

  17. #17
    Originally posted by Stardragon
    Take a look at the size of the file mentioned - is it zero, or significantly smaller than other .trj files in the directory? Also make sure you have sufficient disk space and write permissions.
    It also seems, as a side note, that your clock settings are almost 4 years behind or so (look at the date stamp of the error).
    Sorry please ignore my last post. It looks like the hard drive is dying.
    I can`t eaven backup the DF directory without getting I/O errors.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •