Yamaha receiver DSD question

[scott9873]

Hello all - this is my first post on this message board

I recently figured out that my Yamaha Aventage RX-A880 could play DSD files or a DSD stream over HDMI, even multichannel! So I got excited and bought a multichannel SACD - my Blu-ray player happens to also play SACDs and output a DSD stream over HDMI.
I got everything working perfectly, and the SACD sounds brilliant - but is there a chance that my receiver is internally converting the DSD stream to PCM? I think this might be the case because it has a digital volume control, and usually you cannot modify a DSD audio stream/file.
Thoughts?

 

[Audiofire]

If it is not in the manual, you probably have to contact the manufacturer. DSD over PCM (DoP) could be what you have.

 

[scott9873]

The manual doesn't state anything about DSD over PCM, although that could be a possibility? Another thing to note is when I'm playing a SACD and I go to the info page on the receiver, it shows it is receiving a 2.8mhz multichannel DSD stream. Nothing there mentioned PCM. I guess the receiver must be converting it to PCM when it is not in 'pure direct' mode as it has to do processing such as user specified EQ settings and its room correction settings..

 

[Audiofire]

There is no EQ with DoP, but DoP does not convert DSD to PCM. The extra bit depth in PCM is used for volume control in DoP.

 

[Scytales]

Hello all - this is my first post on this message board

I recently figured out that my Yamaha Aventage RX-A880 could play DSD files or a DSD stream over HDMI, even multichannel! So I got excited and bought a multichannel SACD - my Blu-ray player happens to also play SACDs and output a DSD stream over HDMI.
I got everything working perfectly, and the SACD sounds brilliant - but is there a chance that my receiver is internally converting the DSD stream to PCM? I think this might be the case because it has a digital volume control, and usually you cannot modify a DSD audio stream/file.
Thoughts?

Honestly, you shouldn't be concerned by this topic of "converting DSD to PCM" (in quotation marks, because without any other details, it is an ambiguous statement from my point of view). This entire topic is clogged by dogmatism and faith in the audiophile market, to the point that almost anything that can be read on the Internet is far remote from any objective technical analysis.

As far as the Yamaha RX-A880 is concerned, to my mind, the volume control is done in the analogue domain in an integrated circuit: a Rohm BD34703 that I have highlighted in a red rectangle on the block diagram below (out of the relevant service manual). So your concern is moot, anyway.

 

[scott9873]

Wow, thank you for the very in-depth response! I would’ve never expected an IC in such a modern receiver to do the analog attenuation for multiple channels. I really agree with you about the state of DSD and the “audiophiles”.. I’m just having fun with the limits and diminishing returns of high end digital audio, I really don’t think anything can improve the reliability and fidelity of PCM audio.

 

[antcollinet]

As stated by @Scytales it really doesn't matter. PCM is perfectly capeble of retaining all of the audible music signal present in the DSD. Converted or not, there will be no audible difference. Not for you, nor for any other human being on the planet - assuming the conversion is half way competent.

 

[A Surfer]

Why people think DSD offers any actual audible advantage is mystifying.

 

[SoundsGood64]

Two (or three) additional aspects to consider:

1. If you are playing a 5 or 6 channel SACD, then you might also want to use the room correction feature of your receiver. Probably the receiver must convert from DSD to PCM before room correction processing happens so the event order is something like this: DSD -> DSD to PCM -> Room correction with PCM -> Digitial to Analog Conversion.
2. On the block diagram that Scytales posted above, the DAC chip for Main Zone is PCM5102A. I believe this particular chip cannot receive DSD, unlike some other DAC chips that are used in receivers.
(3. Enjoy that multichannel sound.)
Corrections to this post are welcome.

 

[Scytales]

2. On the block diagram that Scytales posted above, the DAC chip for Main Zone is PCM5102A. I believe this particular chip cannot receive DSD, unlike some other DAC chips that are used in receivers.

Yes, you're right.

I failed to check that point.

 

[Tre2023]

Some receiver manufacturers will state the conversion to PCM of DSD signals while others do not do so.

I’ve set my player at times to output everything in PCM which resulted in the multichannel DSD signal of SACDs being recognized as multichannel PCM 176Hz by the receiver. There was no audible difference in the presentation compared to inputting a DSD signal.

 

[Audiofire]

PCM 176Hz

176 kHz

 

[Scytales]

More exactly 176.4 kHz (4x44.1 kHz or, in case of decimating DSD, 2.8224 MHz/16).

 

[Descartes]

Does anyone has any day comparing DSD vs PCM? I always thought that DSD was superior to PCM! Please explain why you don’t think that is the case!

 

[DVDdoug]

Does anyone has any day comparing DSD vs PCM? I always thought that DSD was superior to PCM! Please explain why you don’t think that is the case!

I'm not aware of specific listening tests but "CD quality" (16-bit/44.1kHz) has been demonstrated to be generally better than human hearing in blind ABX tests where a high-resolution original is compared to a copy down-sampled to 16/44.1. So, 24/96 PCM is certainly better than human hearing. Under "artificial conditions" you MAY hear a difference with CD quality but not with 24/96 which is also better than any hardware.

With the right setup, I believe foobar2000 can play DSD, and there is an ABX plug-in for foobar2000 so if you have a way to convert DSD to PCM you might be able to run your own ABX test. (Just make sure to loudness-match the files it's important to do the conversion yourself so you know is the exact same recording/mix/master.)

What is a blind ABX test?.
Controlled Audio Blind Listening Tests (video)


...You might be surprised how difficult it is to hear the difference between a good-quality (high bitrate) MP3 and high-resolution (or CD quality) original in a proper blind ABX test. You usually have to listen very carefully, if you can reliably hear a difference at all.

 

[Descartes]

I'm not aware of specific listening tests but "CD quality" (16-bit/44.1kHz) has been demonstrated to be generally better than human hearing in blind ABX tests where a high-resolution original is compared to a copy down-sampled to 16/44.1. So, 24/96 PCM is certainly better than human hearing. Under "artificial conditions" you MAY hear a difference with CD quality but not with 24/96 which is also better than any hardware.

With the right setup, I believe foobar2000 can play DSD, and there is an ABX plug-in for foobar2000 so if you have a way to convert DSD to PCM you might be able to run your own ABX test. (Just make sure to loudness-match the files it's important to do the conversion yourself so you know is the exact same recording/mix/master.)

What is a blind ABX test?.
Controlled Audio Blind Listening Tests (video)


...You might be surprised how difficult it is to hear the difference between a good-quality (high bitrate) MP3 and high-resolution (or CD quality) original in a proper blind ABX test. You usually have to listen very carefully, if you can reliably hear a difference at all.

Thank you for your explanation and suggestion. I am able to hear a difference between mp3: and regular CD recordings as long as it is properly mastered.

 

[antcollinet]

Thank you for your explanation and suggestion. I am able to hear a difference between Mo: and regular CD recordings as long as it is properly mastered.

Not sure what you mean by Mo: assuming this is some source of DSD:

Are you able to hear that difference double blind and accurately level matched in an ABX test scoring at least 8 out of 10 correct? And when it is confirmed that exactly the same master is used (generally the only way to do this is to carefully down convert the high resolution file to the redbook file using a method that doesn't introduce errors)

It seems unlikely.

 

[Descartes]

Not sure what you mean by Mo: assuming this is some source of DSD:

Are you able to hear that difference double blind and accurately level matched in an ABX test scoring at least 8 out of 10 correct? And when it is confirmed that exactly the same master is used (generally the only way to do this is to carefully down convert the high resolution file to the redbook file using a method that doesn't introduce errors)

It seems unlikely.

I meant mp3 and cd!

 

[antcollinet]

I meant mp3 and cd!

OK - then that is possible, especially with lower bit rate MP3. You might find it interesting though to try one of the online tests at a higher bitrate, and quality encoder, to see if you can then.

Here is an example.

ABX Lossless vs Lame MP3 test list

A set of browser-based ABX tests to compare Lame MP3 encoded and lossless music samples

abx.digitalfeed.net

Personally I can't detect a difference with 256kb/s or higher when listening via speakers. I haven't tried with headphones.

 

[peng]

The manual doesn't state anything about DSD over PCM, although that could be a possibility? Another thing to note is when I'm playing a SACD and I go to the info page on the receiver, it shows it is receiving a 2.8mhz multichannel DSD stream. Nothing there mentioned PCM. I guess the receiver must be converting it to PCM when it is not in 'pure direct' mode as it has to do processing such as user specified EQ settings and its room correction settings..

Unless you have a way to connect to a Yamaha customer support rep that is at or above (preferred) level 2, such as someone in their engineering/design team, you will not likely get an accurate answer.

You mentioned using HDMI input right? In that case, block diagrams, even schematics will show that the signal get routed to the DSP so it is quite possible that the signal gets converted to PCM, but it is also possible that in direct/pure direct mode, 2 channel DSD may remain DSD. For multichannel signal, especially if at above 2.8 MHz, I think it will most likely be converted to PCM.