Modifying SqueezeBox hardware/jitter specs

[Michael Brouwer]

Has anyone out there looked at the jitter specs of the stock
SqueezeBox, or has anyone actually done any hardware mods to improve
jitter?

Since the SqueezeBox is a bit accurate "transport", assuming you ripped
your CDs without errors which is a whole separate issue, it could be
used to feed a high end system. Of course jitter is the factor besides
bit accuracy which needs to be taken into account.

Michael

[Mark Bennett]

Michael Brouwer wrote:

> Has anyone out there looked at the jitter specs of the stock SqueezeBox,

Yes, someone reported the jitter as being less than 20ppm on the sample
they tested - which is more accurate than many high-end audiophile CD
transports.

> or has anyone actually done any hardware mods to improve jitter?

There has been some suggestion of this, but it's not clear that anyone
has done it - certainly I haven't seen any reported results.

> Since the SqueezeBox is a bit accurate "transport", assuming you ripped
> your CDs without errors which is a whole separate issue, it could be
> used to feed a high end system. Of course jitter is the factor besides
> bit accuracy which needs to be taken into account.

There are a number of people on this forum doing this very thing. Most
who want high quality sound are just using a good quality external DAC
(as I am), which is pretty good. I've done comparisons between the SB
and an Audiolab CDM8000 transport, and using the Benchmark Media DAC
you can't tell which transport your listening to.

See a full summary at:
http://article.gmane.org/gmane.music.equipment.slimdevices.general/8640

Others have compared the SB to a high end Wadia transport and DAC and
have reported that they can tell the difference:
http://article.gmane.org/gmane.music.equipment.slimdevices.general/10026

However, when re-clocked by a dCS upsampler the difference was completely
eliminated:
http://article.gmane.org/gmane.music.equipment.slimdevices.general/9889

In summary, the Squeezebox is capabale of being the transport for
whatever system you want. Standalone for most people using the internal
DAC's, or using an external DAC for the audiophiles amongst us. Using
an external DAC the quality is going to depend on the external signal
path (Upsampler/DAC) more than the SB itself.

Having said that there are a couple of things to look out for:

Some versions of the server/firmware had some problems with channel
reversal - I think this is all fixed now (at least it is for me).

Some people are reporting occasional drop-outs when streaming uncompressed
over wireless links. This could be lurking bugs in server/firmware or it
could be straightforward interference on the wireless link. So, at least
for the time being a cabled connection is preferred (powerline seems to
work well from the reports). (Note - I'm using wireless and I've only
very rarely heard a dropout.)

[Mike Reeve]

Michael Brouwer <michael@...> writes:

>
> Has anyone out there looked at the jitter specs of the stock
> SqueezeBox, or has anyone actually done any hardware mods to improve
> jitter?

Michael

I don't have the test equipment to measure the SqueezeBox's jitter
- does anyone on the list have access to, for example, a Miller Analyser?

But I do address the jitter problem, if it is a problem with the SqueezeBox
(in the absence of measurements who knows ...),
using a Genesis Digital Lens
- between the SqueezeBox and a Musical Fidelity TriVista 21 DAC.

[For those unfamiliar with the Digital Lens,
(which is now, sadly, no longer in production)
it is a device that sits between the source (e.g. SqueezeBox, CD-transport, etc.)
and the (external) DAC.
It buffers the incoming stream from the source in a large RAM
and then sends it out to the DAC
using a high-precision (from the sample-rate perspective) and low jitter clock.
It thus corrects for sample-rate-clock errors and reduces jitter
- although clearly since it is external to the DAC
and not clock synced to the DAC
the link between the Digital Lens and the DAC will introduce its own jitter.]

The Digital Lens displays the sample-rate-clock error (but not the jitter)
and it indicates that my SqueezeBox has an error of 18ppm
(after the 20-second-from-power-up settling period 'discovered' by Michel).
Reports from other Digital Lens users indicate that
many 'audiophile' transports are up around 50ppm,
although I have heard that one of the current Mark Levinson transports
registers around 10ppm.]

One way to get the jitter measured would be to have Stereophile review
the SqueezeBox ... :-)

[As an aside: many of today's DACs have an asynchronous sample rate converter
on the input (typically the Crystal or Analog Devices chips)
and so, notionally, any jitter is removed by these devices
(the same chips/approach is used in the Monarchy, GW Labs, etc. add-on boxes).
There is of course a whole philosophical debate to be had on the use of
asynchronous sample rate converters for jitter reduction
but I don't want to go there ... :-O]

Regards

Mike