Okay, you were right, the difference is linux versus Windows. Although I
thought these files were not working in Windows, I just wasn't giving it
enough time. Normally (as you know), if a file is analyzed locally, the
mipcore process starts up, CPU usage spikes, and then after a long time
of this, the file is analyzed. This is still what happens with files
that work both on linux and Windows.

For files that don't work, however, genpuid starts, then there is a
very long pause with no CPU activity. On linux, that pause never ends,
and when it went on for a while on Windows, I assumed it was the same
thing and killed genpuid. The difference is that on Windows, after a
while, mipcore does start and can analyze the file. For each of these
files, you get the "Error communicating with server" message in the log
file. 

So it seems to me that whatever is causing the "Error communicating
with server" message is not arbitrary but is linked with specific
characteristics of the song that is being analyzed. In other words,
something about a specific file triggers this message.  When this
happens on linux, the program gets stuck (I never see the message in
the log on my linux system); when it happens on Windows, it somehow can
continue.

I checked this by taking a song that can be analyzed on linux and when
analyzing it on windows, I don't get the "error communicating with
server" message. It just starts mipcore and analyzes the data.

So it's still very confusing, but I can work around this by using
Windows just to run genpuid.

One other quick question, how do you look at the tags that genpuid
creates? I can't find the added tags in picard, puddletag, or easytag;
the only thing I can see is a MusicIP Fingerprint in picard. But I
can't see the data tag in any of those programs.


-- 
slamhound
------------------------------------------------------------------------
slamhound's Profile: http://forums.slimdevices.com/member.php?userid=19473
View this thread: http://forums.slimdevices.com/showthread.php?t=91499

_______________________________________________
ripping mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/ripping

Reply via email to