RightMark Audio Analysis compares quality of all 3 SqueezeBox audio outputs

[homey]

Hello,
I have posted a comparison of all three of SqueezeBox 3 outputs [analog/SPDIF/Toslink] using the RightMark Audio Analyzer. The test measures several key audio parameters [such as frequency response, noise level, and THD] in a "real world" systems environment. It can be found at:
<url>
http://mysite.verizon.net/forumwebsp.../RightMark.htm
<url>

[PhilNYC]

Not to be critical here, but what exactly is a "real world" environment? I've visited a lot of people's homes and heard a lot of stereo systems, and there are plenty of different examples of "real world"...enough variety where you'd really have to describe the specifics for any test results to have any significance.

(for example, a "real world" environment in an NYC apartment is going to be vastly different from a "real world" environment in a similarly-priced huge home in Texas)...

[richidoo]

In the 2nd to last paragraph you mention more noise in right channel than left. That is due to interference from the built in plasma display. Turning this to OFF, at least during play is helpful.

[homey]

Hello Phil,
By "real world", I mean connected to a real entertainment system and not in a lab connected to nothing other than test equipment. One could overdesign or underdesign a particular audio component, but what is ultimately important is how it performs when connected to your system...

[seanadams]

Nice work there!

The analog measurements are in the same ballpark as I've measured, but I have to tell you this is not the right way to characterize the digital outputs. You're really just showing the RX-V2500's performance, so all we can conclude is that the digital outputs are working. It doesn't say anything about the s/pdif waveform except that it isn't completely out of whack.

[krochat]

Your comments in the report that the SB3 doesn't support more than
16-bit resolution not correct for all file formats. It will play
44/48kHz x 24 bit FLAC files just fine.

Regards,
Kim

[homey]

Hello Sean,
Thanks for your reply. With regard to the testing of digital outputs, doing the "systems approach" to the analysis is exactly my point. What is ultimately important is the quality of the sound reaching your ears. It will only be as good as the weakest link in audio rendering chain.

One approach to delivering the best quality is to invest in a high quality DAC/preamp/receiver [I'm not saying the RX-V2500 was the best choice, just available] and then use all digital connections to connect the various audio sources.

Can the latest clock recovery circuits [like the CS8416 part] then deliver audio performance similar to the analog connections even though there is a high degree of jitter on some of the inputs?

[gwak]

"The SqueezeBox should ALWAYS be set to its maximum volume level [40] for best signal quality"

thanks that tracks with what Sean had said in this thread
http://207.7.156.4/showthread.php?p=114967

needless to say i am purchasing a pre-amp or new integrated amplifier.

Anyone iterated in purchasing a Musical Fidelity A3CR in Europe?

-best
-gwak

[seanadams]

With regard to the testing of digital outputs, doing the "systems approach" to the analysis is exactly my point.

That's fine, but you've presented this as a comparison between SB3's three outputs, which it isn't at all. It is a comparison between SB3's DAC and the one in your particular receiver! The effects of jitter do not really come into play in these measurements.

Regarding the CS8416 - this is a standard s/pdif receiver, nothing special. The only approaches I've seen that measurably get rid of jitter are those which use an oscillator close to the DAC, i.e. word clock or asynchronoous resampling.

Although with lower jitter you will see a tiny improvement in THD+N figures, it is not a suitable test for this, because the distortion caused by jitter (unlike that of an amplifier) is not harmonic in nature. If you want to measure jitter sensitivity after the DAC, check out the tests that stereophile uses.