[PD-dev] [ pure-data-Bugs-1845770 ] PiDiP ascii renderer eats mac hard disks

[SourceForge.net]

Bugs item #1845770, was opened at 2007-12-06 15:15
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1845770&group_id=55736

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: stereopticon (stereopticon)
Assigned to: Nobody/Anonymous (nobody)
Summary: PiDiP ascii renderer eats mac hard disks

Initial Comment:
...iknow, it sounds like a tabloid headline.

I'm running OSX 10.4.11 on a Macbook Pro  2.16 GHz Intel Core 2 Duo.  I upgraded to pd  0.39.3-extended-rc5 recently and began messing around with the PiDiP ascii renderer. Early in a day of testing I loaded an 11-second, 640x480 quicktime clip compressed with the H.264 codec into  pdp_ascii-help.pd and got…nothing. The patch became completely unresponsive to mouse clicks and I ended up having to quit and relaunch pd and X11.  I eventually discovered that pdp_ascii works with the Sorenson video 3 codec and spent a happy hour turning bits of things into ascii art.

Then I noticed something strange. The hard drive was nearly full.  As I watched in shock, it filled up completely.  A couple of minutes of searching found the culprit: my console log file for the session was 36 gigs.  Turns out that all that space was taken up by a single message, “skipping corrupted frame”, spit out over and over again at the ridiculous rate of 4 megs/sec.  Trying to load the H.264 quicktime is what initiated the loop. I’ve been able to replicate it and came up with the following info:  the bug runs independent of pd and X11, hogs 80% of the CPU, and since for some reason the process is tied to the OSX loginwindow process I was only able to kill it by logging out.

Cheers,

Michael


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=478070&aid=1845770&group_id=55736