Denon DCD-SA1 Review (CD/SACD Player)

[Mikig]

great job and great review of a really fascinating device. Always nice to see what's inside the box...

 

[krabapple]

Back in the day it was recommended by Scarlet Book DSD spec that players employ an output stage filter for DSD playback (i.e., from the analog outputs) at 50 kHz or 100 kHz to remove the hash above those frequencies.

Do you see evidence of that?

 

[NTTY]

Next time you want to measure DSD with perfect signals (in case its digital-in allows it) you can use Multitone Analyzer which has a full DSD generator with perfect signals at all rates.

Thanks for the review!

Sorry @Sokel, I forgot to reply.
Indeed, but I can’t send DSD signals from the generator since the player does not have a UBS input

 

[Scytales]

Back in the day it was recommended by Scarlet Book DSD spec that players employ an output stage filter for DSD playback (i.e., from the analog outputs) at 50 kHz or 100 kHz to remove the hash above those frequencies.

Just to be precise, the corner frequency of 100 kHz is not stated in the Super Audio CD System Description (commonly known as "Scarlet Book"), Part 2: Audio Specification, Version 2.0 of March 2004. This standard describes the characteristics of the low pass filter in Annex E, Audio Signal Recommendations, which is labelled Informative, contrary to Annex D Audio Signal Requirements, which is described as Normative.

The recommendations made in the Scarlet Book are :

"To protect analog amplifiers and loudspeakers, it is recommended that a Super Audio CD player
contain at its output an analog low pass filter with a cut-off frequency of maximum 50 kHz and a slope
of minimum 30 dB/Oct. For use with wide-band audio equipment, filters with a cut-off frequency of
over 50 kHz can be used." (Annex E2)

Do you see evidence of that?

From the frequency response plot of the DCD-SA1 in SA-CD mode published by the German magazine Stereoplay, the level is down 3 dB at about 50/55 kHz. So this Denon seems to follow the recommendation of the Scarlet book as far as the corner frequency of the output filter is concerned.

 

[NTTY]

Back in the day it was recommended by Scarlet Book DSD spec that players employ an output stage filter for DSD playback (i.e., from the analog outputs) at 50 kHz or 100 kHz to remove the hash above those frequencies.

Do you see evidence of that?

Have a look at the last measurement I put in the review. If there is analog filtering, it would be the reason to see the noise slowly going down at 70kHz. But it could be the noise shaping effect too, I can’t be sure. Since I have only pink noise form the Denon test SACD to play with, it’s not sufficient to analyze precisely.

 

[NTTY]

Just to be precise, the corner frequency of 100 kHz is not stated in the Super Audio CD System Description (commonly known as "Scarlet Book"), Part 2: Audio Specification, Version 2.0 of March 2004. This standard describes the characteristics of the low pass filter in Annex E, Audio Signal Recommendations, which is labelled Informative, contrary to Annex D Audio Signal Requirements, which is described as Normative.

The recommendations made in the Scarlet Book are :

"To protect analog amplifiers and loudspeakers, it is recommended that a Super Audio CD player
contain at its output an analog low pass filter with a cut-off frequency of maximum 50 kHz and a slope
of minimum 30 dB/Oct. For use with wide-band audio equipment, filters with a cut-off frequency of
over 50 kHz can be used." (Annex E2)


From the frequency response plot of the DCD-SA1 in SA-CD mode published by the German magazine Stereoplay, the level is down 3 dB at about 50/55 kHz. So this Denon seems to follow the recommendation of the Scarlet book as far as the corner frequency of the output filter is concerned.

30dB/octave from 50kHz could be what we see here from the pink noise wide band analysis.

 

[Henreid]

You can thank Philips for that. We know from an insider that Sony originally envisaged the SA-CD to be 2.0 stereo only. Just a kind of super CD, as the name implied. But at Philips (and also the electronic consumer sale department at Sony), they insisted that SA-CD get multichannel capacity.

Thanks for those insights! I had never researched the history of SACD's development, so I was unaware of these discussions concerning the advanced features the new format would provide. No doubt we can debate at length about which option was better, but I have always felt that well-implemented multi-channel musical recording is not a gimmick, and I tend to agree with those at Philips and Sony who thought that it would be tangible to more consumers. Unfortunately, due to many different factors, SACD has remained a niche product its designers had hoped to avoid.

 

[Sokel]

Sorry @Sokel, I forgot to reply.
Indeed, but I can’t send DSD signals from the generator since the player does not have a UBS input

You mean with Motu or in general?
Cause if DENON can receive DSD through SPDIF all you need is an interface to send it (capable of sending DSD through SPDIF,like XMOS XU216 for example) and then receive the analog results at the other end with MOTU (Multitone analyzer can use different ASIO devices as well,or ASIO and WASAPI,you can mix and play with it)

If with MOTU you're right of course,it can output up to 96kHz through SPDIF if I remember right.
And I'm not even sure if DENON can receive DSD through its SPDIF-in

 

[krabapple]

Just to be precise, the corner frequency of 100 kHz is not stated in the Super Audio CD System Description (commonly known as "Scarlet Book"), Part 2: Audio Specification, Version 2.0 of March 2004. This standard describes the characteristics of the low pass filter in Annex E, Audio Signal Recommendations, which is labelled Informative, contrary to Annex D Audio Signal Requirements, which is described as Normative.

The recommendations made in the Scarlet Book are :

"To protect analog amplifiers and loudspeakers, it is recommended that a Super Audio CD player
contain at its output an analog low pass filter with a cut-off frequency of maximum 50 kHz and a slope
of minimum 30 dB/Oct. For use with wide-band audio equipment, filters with a cut-off frequency of
over 50 kHz can be used." (Annex E2)

One might argue that the last sentence referring to higher cutoff as per Annex E, is part of a Recommendation, so....

But it's a semantic point. The main point was that lowpass filtering was advised at analog output.

 

[krabapple]

Thanks for those insights! I had never researched the history of SACD's development, so I was unaware of these discussions about the advanced features the new format would provide. No doubt we can debate at length about which option was better, but I have always felt that well-implemented multi-channel musical recording is not a gimmick, and I tend to agree with those at Philips and Sony who thought that it would be tangible to more consumers. Unfortunately, due to many different factors, SACD has remained a niche product its designers had hoped to avoid.

Going a little further back, DSD itself was not originally intended to be a consumer format. It was intended for archiving master tapes. The consumer product would be PCM downconverted from DSD (whose sample rate was a large integral multiple of CD rate)

But then someone with $$$ in their eyes who understood the gullibility of 'audiophiles' must've chimed in.....

 

[NTTY]

You mean with Motu or in general?
Cause if DENON can receive DSD through SPDIF all you need is an interface to send it (capable of sending DSD through SPDIF,like XMOS XU216 for example) and then receive the analog results at the other end with MOTU (Multitone analyzer can use different ASIO devices as well,or ASIO and WASAPI,you can mix and play with it)

If with MOTU you're right of course,it can output up to 96kHz through SPDIF if I remember right.
And I'm not even sure if DENON can receive DSD through its SPDIF-in

DSD can’t be sent via SPDIF, as far as I know.
The Denon processes only PCM via its inputs (coax and optical).

 

[Scytales]

Thanks for those insights! I had never researched the history of SACD's development, so I was unaware of these discussions about the advanced features the new format would provide. No doubt we can debate at length about which option was better, but I have always felt that well-implemented multi-channel musical recording is not a gimmick, and I tend to agree with those at Philips and Sony who thought that it would be tangible to more consumers. Unfortunately, due to many different factors, SACD has remained a niche product its designers had hoped to avoid.

Going a little further back, DSD itself was not originally intended to be a consumer format. It was intended for archiving master tapes. The consumer product would be PCM downconverted from DSD (whose sample rate was a large integral multiple of CD rate)

But then someone with $$$ in their eyes who understood the gullibility of 'audiophiles' must've chimed in.....

Yes, I have read in some pro-audio magazines of the nineties, such as Studio Sound, that story about DSD being originally intended for archiving master tapes .

But the gentleman who was kind enough to provide us with insight about the development of DSD has said this story was "propaganda fluff!", ie a story created for promotional purposes.

Some more fascinating history:

• First multichannel recording session with (bulky!) Philips prototype DSD recorders ;
• Early SA-CD test pressings and format history ;

 

[Sokel]

DSD can’t be sent via SPDIF, as far as I know.

It can now,here's an example (there's lots,all new DACs,bridges,etc more or less) :




(link)

(that's DoP,didn't search about native,maybe there's not)

 

[krabapple]

Yes, I have read in some pro-audio magazines of the nineties, such as Studio Sound, that story about DSD being originally intended for archiving master tapes .

But the gentleman who was kind enough to provide us with insight about the development of DSD has said this story was "propaganda fluff!", ie a story created for promotional purposes.

Ah, you mean Black Elk over on Hoffman's board...

where they all seem to accept as a given matter of faith that DSD itself, rather than the care taken in sourcing and mastering audio, is what makes (some) SACDs sound better than their CD counterparts.

And I fail to see how selling a false story that DSD was originally meant for archiving, is very 'promotional' to SACD. Though it's admittedy more compelling than 'hey, we needed a new revenue stream and were getting killed by CD ripping, so we decided to properly remaster some albums and sell them on a pricey and higly copy protected new delivery format thet we'll claim makes the sound better'.

 

[Scytales]

Yes, I mean Black Elk, because the man behind the pseudonym was deeply involved at the heart of the creation and deployment of DSD and SA-CD, ie inside Philips then Sony, and is therefore an invaluable witness to the events of the time and the key people who set them in motion (or standstill!).

 

[Scytales]

And I fail to see how selling a false story that DSD was originally meant for archiving, is very 'promotional' to SACD.

I imagine it was in order to get traction in the music industry. At the time, Sony was a big player in town, both in software and hardware, in the pro audio as well as in the consumer market. To let anyone believe that Sony was so confident in his new system that it was ready to transfer all its vault to DSD was probably a trick to get the other players in town to think about doing the same or at least let appear that the deployment of DSD was already in an advanced stage or was promised a bright future (after all, Sony had been so successful with the CD Audio). But as Black Elk let us known, the suggestion that Sony Music systematically makes a DSD transfer of any rework or new work was actually never acted upon. So much for DSD as Sony's archiving format.

 

[NTTY]

It can now,here's an example (there's lots,all new DACs,bridges,etc more or less) :


View attachment 391955

(link)

(that's DoP,didn't search about native,maybe there's not)

Interesting, thanks. It is DSD over PCM indeed, but that means it’s possible. I didn’t know.

 

[Andreas007]

Thanks NTTY! I love those vintage reviews.
I'm tempted to offer my Kenwood DP-7090 for measurement...

 

[Scytales]

Interesting, thanks. It is DSD over PCM indeed, but that means it’s possible. I didn’t know.

But of course, DoP cannot be send to the Denon, which don't have the receiving circuitry that is able to identify and handle it properly, because the Denon was designed and put on the market several years before the "invention" of DoP . If DoP were to be sent to the Denon, all that would come out of the analogue or digital outputs would be noise.

 

[Haskil]

I imagine it was in order to get traction in the music industry. At the time, Sony was a big player in town, both in software and hardware, in the pro audio as well as in the consumer market. To let anyone believe that Sony was so confident in his new system that it was ready to transfer all its vault to DSD was probably a trick to get the other players in town to think about doing the same or at least let appear that the deployment of DSD was already in an advanced stage or was promised a bright future (after all, Sony had been so successful with the CD Audio). But as Black Elk let us know, the suggestion that Sony Music systematically makes a DSD transfer of any rework or new work was actually never acted upon. So much for DSD as Sony's archiving format.

And to my knowledge, Philips, which was also a major record publisher - owner at that time of Decca, Polydor, DGG, Philips Classics, etc. -, did not publish SACD in Europe... Only in Japan.
At the same time, the first Sony SACDs were single layer and were mostly reissues of analog recordings. They were single layer because the representatives of musicians in the United States would have demanded to receive rights for each layer: CD and DSD...
The launch of the SA-CD was a fiasco even worse than that of the two quadraphonic systems in the 1970s: SQ and CD4...