I think at some level, it's wireless for both ziggyb63 and nptohc.  Just
something that's irregular enough to not show up on a test reliably.

Okay, first ziggyb63:

I'm going to estimate mixed b/g real-world network speed as 9Mbps.  The
fact that your server is connected wirelessly is a very big deal--this
will increase the number of collisions and slow everything down more
than just another wireless client.  G-only operation is a function of
the access point.  You have to cut off your PSP's, or set up a separate
802.11b WLAN, to go G-only for your SB3.  9Mbps divided by 4-12 =
750kbps - 2.2Mbps per client, if everything shared the spectrum equally
and there were no collisions, which is not the way things really work. 
I'm not positive on SlimServer's settings, but FLAC's bitrate is in the
neighborhood of 800kbps per stream.  So even though the numbers are
inexact, I can tell you are working with very narrow margins on your
wireless network.  And we don't yet know if you have interference from
the neighbors.

Recommendations?  Step 1: connect your server to your home network with
a wired connection.  Step 2: consider a second wireless access point. 
This second access point can be specially designated for SB3s, operate
at G-only speeds, and work on a non-overlapping channel (1 or 12).  To
test this before laying down the cash, you should REALLY connect your
server via a wired connection anyway, then turn off every wireless
client in the house except the SB3's, then test sync.  I would also
recommend installing NetStumbler to find out what channels your
neighbors may be using too.  The only non-overlapping channels are 1, 6
and 12.  8 overlaps 6 and 12, etc.  6 is the default for most access
points, so 1 and 12 tend to be safer.  For the record, WEP and MAC
filtering is trivial security (but I understand you're limited by the
PSP's).  If you set up this second G-only network, you should use
WPA2/AES.  If spending more cash is absolutely out of the question, and
the server CAN'T be connected via a real ethernet cable for some reason,
the next best thing is to transcode to MP3, preferably at bitrates under
200kbps.

Now for nptohc:

Wireless access points on the same network can interfere with each
other just as badly as neighboring access points.  The fact that your
public network and your private network operate on non-overlapping
channels doesn't help when multiple access points on the same network
share the same channel and have overlapping ranges.  In particular, I'm
wondering if the two wireless routers on either side of the stone wall
are clobbering each other.  It's also worth checking if those cordless
phones operate in the 2.4GHz spectrum.

My testing recommendation here is: Unless you verify the phones are not
2.4 GHz, unplug them and remove the batteries.  Turn off every WAP on
channel 13 except one.  Move the squeezeboxes within range of this one
WAP, and test sync.  If the problem goes away, the solution is to
unclutter your wireless spectrum.  I'm not even sure transcoding to a
low bitrate would help here, depending on how badly your two access
points are clobbering each other.

Hope that helps.  I could obviously be wrong, but given the information
I have, I think it's a reasonable assessment.


-- 
CatBus
------------------------------------------------------------------------
CatBus's Profile: http://forums.slimdevices.com/member.php?userid=7461
View this thread: http://forums.slimdevices.com/showthread.php?t=35479

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

Reply via email to