Foldit.exe has stopped working. Windows is looking for a solution

Started by bob1928

spmm Lv 1

I am getting a lot of crashes 'program not responding' when I return a client from a minimised state when running a script.
I have learnt not to click anything until I have checked that the wheel is actually turning. Sometimes it comes good quickly, often after a minute wait, frequently never and I have to restart the client and redo the structure which is never saved. If a rank up or down message is appearing, possibly created by another client or team member, this seems to make it worse. I have very good system resources and no other programs running, but no graphics or sound cards. I quite like seeing the rank ups and downs :(
Running multiple clients minimised does not seem to make the restore failure worse or better.

http://msdn.microsoft.com/en-us/library/windows/desktop/dd744765(v=vs.85).aspx - this describes the effect.

Susume Lv 1

You might check your view options to make sure View: Show voids is turned off before minimizing the game. There is an old known bug where the game is unresponsive up to several minutes when restoring from minimized, if Show voids is turned on. Even if game sound is turned off, the sounds for all popping voids are saved up while minimized and insist on playing one after another when the window is restored, making the game completely unresponsive with a blank screen, sometimes for many minutes, after which it finishes drawing the screen and returns control to the player.

spmm Lv 1

ah thanks, I had heard of that but forgotten it :) usually show exposed and voids when I am running.
Possibly catching other players as well.

Bruno Kestemont Lv 1

I may be wrong, but I think it's sometimes linked to memory.
On a very big puzzle, I have this bug after an hour or two with an old computer on Windows xp with memory messages ("windows is running out of memory, trying to extend virtual memory" before crash).
see http://fold.it/portal/node/996512

On a new Mac, devprev, no problem.

But I see in the activity monitor, after 60 hours activity (long recipes):

For a 611 segment puzzle: Memory use = 1.67 Go (devprev)
For a 270 segments puzzle (hotspot) Memory use: 1.00 Go (New Chapter)
For a 101 residues puzzle (845) Memory use: 684 Mo (New Chapter)

So there is a correlation between memory use and length of the puzzle. May be something like, after sometimes, the system goes out of memory. A potential solution to save something on disk or server and refresh memory ?

I also have this problem like spmm when trying to share after a disconnection from server. If I'm patient, it may resolve after more or less time.

Another cause was when using similar recipes on same puzzle on different clients with same track.

Another cause was simply script errors (undefined variables or functions).

On a Mac, "non responding" occurs also, but quite rarely, in similar situations after a disconnection from server. (but disconnections occur once a day).

gitwut Lv 1

845 has crashed 3 times on me in the last 24-hours. At least twice I was running Acid Tweeker v2.4

I'm attaching a debug.txt file, since it appears to been updated around the time of the crash.

bob1928 Lv 1

Its back. (or still here)
Puzzle 897, crashed twice while quietly running long scripts.