If Bits are Bits

[solderdude]

The file is riped from CD and read by MusicBee from a computer.
Either by USB or Bluetooth. The sound is the same.
The merchant https://portier-hifi.ch/ from whom I ordered the device also made the same observation as me but also on other products.

The Cary Audio DMC-600SE is a CDP that can also function as a DAC.
Have you tried the following:

CD directly played on Cary Audio DMC-600SE
same CD ripped to a file and then played back over the (same) Cary Audio DMC-600SE

It should not sound any different. When it does some processing is being done to it.
Faulty rips cannot sound different anyway, they will have ticks in it or errors but not a different sound. That's not how digital works.

When comparing 2 different DACs the reconstruction filters should be 'similar' and level matched within 1% (0.1dB) which is easy to verify using a PC/laptop with analog in.
You either want the 'mystery' to be solved or want it to perpetuate and validate the purchase of the Cary for yourself.
A small level difference can set you on the wrong foot. It needs to be verified by measuring the outputs (using a constant tone from say... a test CD)

When you want to convince ASR folks you need to come with technical proof or recordings, subjective opinions simply aren't accepted here. Even if others heard it to (that also had no proof other than their blue eyes.

Bits do not have a sound, they can not be timed correctly (jitter) which will not result in the observed sound difference.
When bits are flipped you will hear ticks, short caps or might duplicate the last sample.
Depending on which bit is flipped, because not all bits in a stream/rip are audio bits and it won't flip left and right bits in the same manner as channels are encoded in one stream (L and R samples after each other).

When the sound 'changes' character you can be sure that either the audio has been processed or there is a level difference or there are audible (and quite measurable) differences between the compared DACs. This could be completely different reconstruction filters or conversion or intended by the manufacturer in order to 'be different' as a sales argument or 'theory' they have.

 

[voodooless]

CD directly played on Cary Audio DMC-600SE
same CD ripped to a file and then played back over the (same) Cary Audio DMC-600SE

That is exactly what he did. At least as far as I can deduce.

 

[solderdude]

Seems like he did:

The DAC output of the DMC-600 is exactly the same regardless of the source.

If this is the case some audio-processing must be happening to either internal CDP or incoming digital file on purpose or it is just the usual sighted bias.

If @C3dr!c wants to get to the bottom of this he either must record the output (192/24 if possible) and or measure the output of a test CD with a constant tone (and rip that CD too) to find out if some processing of audio occurs (level difference or totally different reconstruction filter).

The observations point to some (deliberate ?) audio level difference being made or 'sighted bias' (knowing what is being played).

 

[voodooless]

What would be interesting in this case is to record the analog output (same song) at 192/24 and submit that to ASR for comparison.

We first have to check at the very least:
- he’s running the media player in exclusive mode for bit-perfect playback
- he doesn’t have replaygain or any other processing, including a volume change, enabled in the media player software
- the CD player is in bypass mode for the upsampling shenanigans, because we know it doesn’t work in USB mode
- the CD player has all the gains for the various inputs set to the exact same level

 

[solderdude]

- the CD player has all the gains for the various inputs set to the exact same level

Dunno.. that would only be logical when there would be conversion to DSD happening or different DAC chips are used for CDP and digital inputs that have a level difference or filtering.
I don't think possible inputs are all converted using own DACs and have level differences or some audio processing creating different levels depending on the used input.

But yes, if @C3dr!c would want to find out for real all aspects must be clear. Especially what is outputting the ripped CD to the digital input of the DAC.

 

[voodooless]

Dunno.. that would only be logical when there would be conversion to DSD happening.

The CD player has digital gain control for every input. Read the manual, page 17… why would DSD conversion be relevant here?

 

[goat76]

To be clear, we’re talking about only one device here, it can play CDs, but is also a USB DAC.

Thank you, I missed that information.

But it's still not clear to me what this player does with different types of source material, if the playback of the WAV file goes through the same circuitry as the CD, or if something may be affecting the levels differently.

It's too many uncertainties at play here.

 

[solderdude]

The CD player has digital gain control for every input. Read the manual, page 17… why would DSD conversion be relevant here?

Ah ... so some digital audio processing is present/optional (gain specifically)... did not know that and could certainly explain the observed phenomenon.
As the CDP would sound 'fuller' it would make sense the digital inputs would have some 'default' attenuation setting or the device playing has some digital attenuation applied.
This should be checked first (using a constant tone and measurements).

When converting PCM to DSD usually a different DAC chip is used and level differences are logical. Of course this is not the case here it seems.

 

[Koeitje]

Its so funny to me that people in 2024 still not understand how digital files work. It doesn't matter if its stored on a CD (which is worst in terms of errors during reading), HDD, SSD, Memory Card and from where it is played. Its still the same data, its just how that data is handled afterwards that can introduce changes.

 

[voodooless]

When converting PCM to DSD usually a different DAC chip is used and level differences are logical.

Is it? I don't think we see many DACs with different chips for PCM and DSD? However, even with the same DAC chip, DSD levels are generally different, depending on the chip used.

Anyway, this CD player actually uses two sets of DAC chips, AK4495 and AK4490. The manual is very vague about why and how. But it's safe to say that neither is SOTA, as one would expect from a device in such a price bracket.

 

[antcollinet]

Looking at the manual, you have your answer
View attachment 395766

https://www.caryaudio.com/wp-content/uploads/2016/12/DMC-600-and-DMC-600SE_manual.pdf

Note the rest of the ASR folk. Don’t be so hard on people who say they hear differences between two things that should not be different.

Upsampling, if done correctly, and it usually is, will be audibly transparent.

 

[JiiPee]

Upsampling, if done correctly, and it usually is, will be audibly transparent.

Come on... When You upsample a digital stream that is already transparent, the result will not be just transparent, but super-transparent!

 

[AnalogSteph]

Looking at the manual, you have your answer
View attachment 395766

https://www.caryaudio.com/wp-content/uploads/2016/12/DMC-600-and-DMC-600SE_manual.pdf

Uh-oh. The thing has an ASRC, but only in CD playback? Then they basically had two options:
1. introduce 1-3 dB of digital attenuation to ward off intersample-overs - which would make CD playback a bit softer than external sources.
2. play CD material as-is and live with the 0 dBFS hard-limiting typical for ASRCs, which could make peak loudness war era material sound even more crappy.

Since this is a relatively recent device, I'd be putting my money on the first option. If so, a bit of digital attenuation on the computer side to match levels should sort the differences out.

 

[somebodyelse]

Upsampling, if done correctly, and it usually is, will be audibly transparent.

Downsamplers measured - with some performing much worse than you might expect. Have a look at Ableton Live 7 for example. This underscores the 'correctly' part, and most of the underperformers have improved in later versions.
https://src.infinitewave.ca/

 

[Blumlein 88]

Uh-oh. The thing has an ASRC, but only in CD playback? Then they basically had two options:
1. introduce 1-3 dB of digital attenuation to ward off intersample-overs - which would make CD playback a bit softer than external sources.
2. play CD material as-is and live with the 0 dBFS hard-limiting typical for ASRCs, which could make peak loudness war era material sound even more crappy.

Since this is a relatively recent device, I'd be putting my money on the first option. If so, a bit of digital attenuation on the computer side to match levels should sort the differences out.

So once again we see that level matching is job #1 for listening comparisons. Plus level matching is super easy to do. Checking frequency response is job #2, but slightly more difficult for most people to check. Job #1 and job #2 catch about 99% of reasons something actually sounds different.

 

[voodooless]

So once again we see that level matching is job #1 for listening comparisons

Not quite: the #1 job is to make sure that the whole setup is properly configured first, like: https://www.audiosciencereview.com/forum/index.php?threads/if-bits-are-bits.39484/post-2097949 .

Once you've done that it's time to grab some test files and a DMM.

 

[FrantzM]

bits are bits.. End of the story.
Applications that are order of magnitudes more sensitive than mere audio applications, function without any issues going through hundreds of interfaces, yes, hundreds, without any errors.

It is quite possible for a device to have different signal paths for different inputs.. and that could result in different sound from those interface, possible.
Else.. Same paths though same processing.. bit will be bits.
The END.


Peace.

 

[GXAlan]

Sort of reposting because everyone is hammering the OP with the same thing without reading context.

Instruction manual says that even if you turn on the upsampling, it does not apply it to the USB only physical media playback. This confirms that there are different signal paths, that the OP’s subjective impressions may reflect differences in signal processing.

Going beyond this, we have to measure and analyze with DeltaWave

 

[HammersRocco]

bits are bits.. End of the story.
Application that are rder of magnitudes more sensitive than mere audio applications , function without any issues going through hundreds of interfaces, yes, hundreds, without any errors.

It is quite possible for a device to have different signal paths for different inputs.. and that could result in different sound from those interface, possible.
Else.. Same paths though same processing.. bit will be bits.
The END.


Peace.

Sort of reposting because everyone is hammering the OP with the same thing without reading context.

Instruction manual says that even if you turn on the upsampling, it does not apply it to the USB only physical media playback. This confirms that there are different signal paths, that the OP’s subjective impressions may reflect differences in signal processing.

Going beyond this, we have to measure and analyze with DeltaWave

OP question was phrased in the typical manner of some posts - kind of dropping the mic on ASR - I heard a difference... WHAT DO YOU THINK OF THAT??!!! [mic drops]

Plus the subject tends to make ASRs skip reading!


Because even with the clarification there is a high likelihood of any differences being in the posters mind, given what we know. All things being equal of course. The streamer might be malfunctioning!

 

[mhardy6647]

I dunno.
Some of us are actually drawn to "I can't measure it*, but I know what I heard" threads -- if only to watch the sequelae.

____________
* and neither can you