Sorry I've been off this for a few days, trying to work on other things.
Anyway it looks like we seemed to have settled down to a few people have
problems with 24/96 at higher compression when using wiFi to an external
server. In some cases wired fixes these. 

First off, does anybody have any problems with an form of 24/96 when
WIRED to an external server? 

>From what I'm reading it looks like its a wiFi issue. When using WiFi
wav seems to work and flac -0 seems to work. Higher compression wav
seems to work when transcoded to PCM. Does anybody have a situation
where flac -0 or wav does NOT work? 

If the above is true, then it looks like we just have to worry about
streaming over WiFi with flac greater than -0. Does this seem to be the
situation?

Given that wavs seem to work it does not sound like its a transmission
bandwidth issue. This seems to point to it being something inside the
touch, some sort of interaction between WiFi and higher level flac
decompression. 

IF the builtin server is not running the amount of free memory should
be plenty to handle the needs of this operation so I don't think that
is the first place to look. 

To me it seems like there are two reasons stuttering might be
happening, either data throughput is not fast enough for some reason,
or the code working with that data is not running (swapped out of
memory or process not running). Lets try and come up with some possible
scenarios and see if we can come up with some tests to try and pinpoint
whats happening. 

(the problem here is that I don't have the problem so I can't go off
and run these tests myself, its going to have to be those that DO have
the problem who run the tests)

Some of the previous tests have noticed that the CPU will max out
occasionaly, but only for short periods of time. One of the tests
seemed to show that low average with occasional max bursts works fine,
but much higher average without the max bursts does cause problems.
This sort of makes sense since there are fairly large buffers in the
player code, so short interruptions in the data flow should not be a
problem. One option to try here is increase the buffer size in
squeezeplay and see if that improves things. 

Another place I would like ot look at is the WiFi itself. There are
various levels of encription in use in WiFi, no encryption, WEP and
several levels of WPA. I would like to see if varying the encryption
type makes any difference. So could any of you that have the problem
try different encryptions levels on your WiFi and report back. 

Another thing to look at is process run statistics, see if either the
WiFi or jive process is not getting run for significant periods of
time. Another thing to check is if there is any significant swapping
going on. Given the fairly large amount of free memory there SHOULDN'T
be any swapping happening, but its a good idea to check. 

I'll leave it at that for now.

John S.


-- 
JohnSwenson
------------------------------------------------------------------------
JohnSwenson's Profile: http://forums.slimdevices.com/member.php?userid=5974
View this thread: http://forums.slimdevices.com/showthread.php?t=77080

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

Reply via email to