Analyze this: data on the latest version of the Airport Express DAC

[MondoAudio]

The DAC in the original version of the Apple Airport Express wasn't very good. But Ken Rockwell measured and reviewed the newer 2014 version of the Airport Express and came to this conclusion: "At any price, the Apple AirPort Express is a great DAC. It sounds great, with a completely silent background, full 2 V output, smooth detailed sound and full infrasonic bass response." Here's his data and full review.

Would you come to the same conclusion based on his data? (While I understand the basics of interpreting DAC measurements, I don't have experience to come to any firm conclusions based on the data myself. So my question here is a sincere one.)

EDIT: Ken Rockwell's assessment was done with a 2012 version. It is the last version of the Airport Express. I mistakenly refer to it as the 2014 version.

 

[restorer-john]

Definitely. His tests are comprehensive and the test gear used is excellent- just as good or better than AP's gear.

He can be a polarizing figure to some, and he makes some mistakes, but if you read through the words, there is decent content in there.

 

[amirm]

Would you come to the same conclusion based on his data?

He is mixing subjective and objective in that remark so you can't just go by his data.

Using his data, he is getting 90 dB SINAD (assuming his THD numbers include noise). That would put it in fourth tier of performance in this table of mine:

So way off from top class performance. The minimum performance I like to see is high 90s.

 

[restorer-john]

Using his data, he is getting 90 dB SINAD (assuming his THD numbers include noise). That would put it in fourth tier of performance in this table of mine:

He's using the original CBS-1 test disc as a source. It's a Redbook CD and is 16/44 material.

You routinely use 24/44 from the AP. Not a fair comparison.

 

[MondoAudio]

Interesting point, @restorer-john. So would your follow-up conclusion be that devices tested with 24/44 material are 1) being held to a more rigorous standard in terms of their ability to turn bits into sine waves or 2) being given better input material and therefore giving them better opportunity to create better sine waves?

@amirm, thanks for putting his data into a comparative chart. Very helpful.

 

[Darwin]

I’ve seen this article before.
The thing is he says there is no jitter but the AP is known for high jitter. I experienced this myself when using it with airplay attached to powered speakers. I had to replace it with a 3rd gen Apple TV which had an optical connection and better jitter rates. The AP jitter was audible.
A google search shows a lot of discussion of this.

 

[MondoAudio]

@Darwin, did you experience the audible jitter using the old (pre-2012) or new (2012) Airport Express?

 

[amirm]

You routinely use 24/44 from the AP. Not a fair comparison.

If the source signal is the limiting factor, we should see a SINAD of 96 dB. To the extent the DAC only goes up to 90 dB, most likely its own performance is limiting as it is not able to have better dynamic range than 16 bits.

So while using 24 bit signals may improve things a bit (the source noise is additive to DAC's), it won't change the picture much.

Why is he testing with 16 bit signals anyway?

 

[svart-hvitt]

If the source signal is the limiting factor, we should see a SINAD of 96 dB. To the extent the DAC only goes up to 90 dB, most likely its own performance is limiting as it is not able to have better dynamic range than 16 bits.

So while using 24 bit signals may improve things a bit (the source noise is additive to DAC's), it won't change the picture much.

Why is he testing with 16 bit signals anyway?

Apple AP X truncates?

 

[amirm]

Apple AP X truncates?

His 90 dB I thought was for non-bluetooth performance.

 

[svart-hvitt]

His 90 dB I thought was for non-bluetooth performance.

You mean AirPlay?

 

[amirm]

You mean AirPlay?

I am assuming that is what was used:

I guess he would be limited to 16 bits regardless of test tone resolution if it does use airplay.

 

[restorer-john]

Why is he testing with 16 bit signals anyway?

Probably because 16bit 44.1 is representative of what the majority of digital source material is in. As far as I know, there is no commercial distribution of 24bit 44.1 lossless content.

Also, because the CBS-1 test disc is a reference Redbook source, like the Sony YEDs and the Nippon Columbia (Denon) and he can zip from test track to test track very quickly. Perhaps it's not a piece of cake to set up the R&S to output multiple test signals quickly through the SPDIF output on it?

Although, putting the disc on a network and streaming the files across wi-fi is not exactly a 'reference source' either, IMO.

 

[MondoAudio]

Is there a relationship between the bit rate and SINAD values? If you were to compare a higher-tier DAC and an Airport Express fed the same 16/44 signal, would the relative SINAD values change much? (I ask because most of my sources are 16/44 and I wonder if a change in DAC would make a noticeable difference.)

 

[amirm]

Is there a relationship between the bit rate and SINAD values? If you were to compare a higher-tier DAC and an Airport Express fed the same 16/44 signal, would the relative SINAD values change much? (I ask because most of my sources are 16/44 and I wonder if a change in DAC would make a noticeable difference.)

As I explained earlier, higher performance DACs reach a SINAD of 96 dB when fed 16 bit data.

 

[MondoAudio]

Ah, I understand what you meant now. Thanks for clarifying. Key takeaway: if I want to improve my audio chain noticeably, I should first improve my source material.