[AnalogEuphoria]

I'm glad I ran across this thread as I was contemplating a Denafrips Pontus II based on the glowing reviews. Personally, I haven't heard a difference between a Lyr 3 going from the base DAC to the multibit, and not between that and the red Audioquest plug-in DAC. Then again, I can't say for certain that I've heard a sound improvement when I've rolled a couple of different tubes in the Lyr.

Unless you're doing blind testing, there's going to be some level of expectation and bias anyway. If you just spent $3000 on a new DAC, it darn well better improve the music, right?

Pretty much all modern DAC designs are very well engineered and unless broken I think most measured performance is inaudible, implementations im sure are making these subtle differences in sound not vanishingly low noise floor and perfect linearity etc. At end of the day its the headphone or speaker that has the hardest job to do and imo most of the money should be spent there.

Side note - I think its also very important to spend time looking for the best sounding masters of the recordings you like and not settling with whatever these streaming services hand you. Poor reissues and remasters can make a good system sound compressed and muddy or overly bright and harsh.

check dynamic range database and forums for information on which people find the best
https://dr.loudness-war.info

 

[iFi audio]

Couldn’t find much info on their topology other than definitely not using resistor ladders for conversion

It's all good, Nagra products are always a bit different. Great sound though!

 

[gregorio]

what I mean is that there are a lot of audiophiles who don't like having artificial sound effects on recordings that don't actually appear in the original musical performances, the recordings should be as close as possible to the original musical performances.
Audiophiles like these are the targets of DSD recordings.

Sorry but I don’t really understand that, for a number of reasons:

With a lot of recordings there isn’t an original musical performance, there’s just a bunch of takes/overdubs recorded at different times, then edited and processed together to sound like a musical performance.

Unless an audiophile can compare a processed recording with the same recording that hasn’t been processed, how can they possibly know which they prefer?

If engineers processed a recording and it sounded worse than without the processing why would anyone ever pay them to do that and how would they ever work again?

I can pretty much guarantee virtually no audiophiles could even tell what processing has been added or changed and if they could compare with an unprocessed version, they’d actually prefer the processed version.

And lastly, the majority of DSD recording have been processed and many of them heavily.

G

 

[PopZeus]

The value of DSD rests mostly on its native compatibility with delta-sigma DAC chips, based on my very limited experience. The benefit of R-2R is its more native compatibility with more common audio formats like mp3, aac, flac, etc, but there's more to it than just that. I do believe many R-2R DACs add 2nd and 3rd order harmonics to the analog signal, like tubes. With good headphones and amplification, as well as the right music tracks, the differences between DACs can be pretty striking.

 

[CiptoH]

And lastly, the majority of DSD recording have been processed and many of them heavily.

Of course the majority of DSD recordings will be processed through a digital editing and mixing stage (with an analog or native DSD mixer).
The important thing is do not add unnecessary sound effects which are not present in live acoustic performance where the recording was taken.
This is what I mean by "don't process", don't add unnecessary sound effect, sorry for the language barrier, I'm not a native English speaker.

 

[gregorio]

Of course the majority of DSD recordings will be processed through a digital editing and mixing stage (with an analog or native DSD mixer).

No, the majority of DSD recordings have been converted to PCM for mixing and editing. A few might use a native DSD mixer where appropriate and a few analogue mixing, although that’s a marketing gimmick which actually lowers resolution.

The important thing is do not add unnecessary sound effects which are not present in live acoustic performance where the recording was taken.

Are you really saying that we should stop making recordings of rock/pop and just about all genres other than some classical, jazz and ethnic music? Pretty much all popular genres of music rely on processing.

Even with classical/acoustic genres, I’m sorry but we’re not going to deliberately make recordings sound worse just because a few extremist audiophiles believe some false marketing about processing.

G

 

[CiptoH]

A few might use a native DSD mixer where appropriate and a few analogue mixing, although that’s a marketing gimmick which actually lowers resolution.

I don't think a native DSD mixer will lower the resolution.

Are you really saying that we should stop making recordings of rock/pop and just about all genres other than some classical, jazz and ethnic music? Pretty much all popular genres of music rely on processing.

Most audiophile recordings are of the jazz, classical, acoustic genre etc, and not electronic, metal, dance genre etc.
The target of DSD recordings are audiophiles and not everyone.

 

[gregorio]

I don't think a native DSD mixer will lower the resolution.

I was referring to analogue mixing reducing the resolution. Neither a native DSD mixer nor converting to PCM and conventional digital mixing will lower the resolution.

Most audiophile recordings are of the jazz, classical, acoustic genre etc, and not electronic, metal, dance genre etc.

About half of DSD recordings are classical, there’s a significant percentage of non-acoustic genres.

The target of DSD recordings are audiophiles and not everyone.

Historically, the target of DSD recordings were those who could afford the higher cost of SACD players and decent home sound systems. Today, DSD is an almost dead format and is just a tiny niche for those audiophiles who still believe the false marketing.

G

 

[CiptoH]

Today, DSD is an almost dead format and is just a tiny niche for those audiophiles who still believe the false marketing.

No false marketing, the DSD format has a higher dynamic range (approximately 120 dB vs 96 dB), higher resolution (approximately 24-bit vs 16-bit) and wider frequency response (up to 100 KHz vs 22.05 KHz) when compared to the majority of recordings (PCM 44.1 KHz 16-bit).

The DSD format is not popular because the majority of consumers don't care about high-resolution recordings, this is also the reason why high-resolution PCM recordings (24-bit 96 KHz or higher) are also unpopular with the majority of consumers.

However, the DSD format will last as long as there are consumers who are willing to buy it. If the DSD market didn't exist, there certainly wouldn't be any software developers willing to make DSD editing and mixing (native DSD) software...
.....
"But the technology for mixing and editing in DSD without DXD is emerging rapidly"
https://positive-feedback.com/audio-discourse/dsd256/

 

[gregorio]

No false marketing, the DSD format has a higher dynamic range (approximately 120 dB vs 96 dB), higher resolution (approximately 24-bit vs 16-bit) and wider frequency response (up to 100 KHz vs 22.05 KHz) when compared to the majority of recordings (PCM 44.1 KHz 16-bit).

But that IS false marketing! The DSD format has the same dynamic range as CD (approx 120dB) and much less than 24bit. But that doesn’t make any difference because pretty much all recordings have a dynamic range roughly 1,000 less than that anyway. And 100kHz freq response is nonsense too, studio microphones can’t even record anything in that range and human hearing doesn’t extend beyond 20kHz anyway.

The DSD format is not popular because the majority of consumers don't care about high-resolution recordings, this is also the reason why high-resolution PCM recordings (24-bit 96 KHz or higher) are also unpopular with the majority of consumers.

Many consumers do care about high resolution recordings. The very highest resolution recordings can be contained in 16/44, the majority of consumers aren’t interested in resolution that is inaudible and beyond that of the recordings themselves anyway.

However, the DSD format will last as long as there are consumers who are willing to buy it.

Absolutely, as long as there are enough people who will pay for it, there will be small record labels catering to that demand.

If the DSD market didn't exist, there certainly wouldn't be any software developers willing to make DSD editing and mixing (native DSD) software...

The market does exist but it’s a tiny market, which is why software developers only make extremely simple, basic DSD editing and mixing software, which is little more than a toy compared to the numerous complex, comprehensive DAW software for PCM.

G

 

[DivineCurrent]

Only big difference I heard in a DAC was a DAT player (Yes, an old Digital Audio Tape player) Panasonic I believe. It sounded a lot warmer than every other DAC I was using at the time, both ESS and AKM Dacs. Not sure if it was a built in filter or something else, but there was no mistaking the difference. Currently, I really can't tell the difference among modern DACs. Even though I own an RME ADI-2 DAC, I bought it for it's features and not necessarily the high end DAC, even though it sounds excellent.

 

[CiptoH]

The DSD format has the same dynamic range as CD (approx 120dB) and much less than 24bit

The CD format only has 96 dB (16 bit) dynamic range !

"The compact disc’s dynamic range is about 90 decibels, compared with about 70 decibels on the best phonograph discs, thus accounting for the distinct, clear sound obtained from even the cheapest CD players"
https://www.britannica.com/technology/compact-disc/Analog-versus-digital-sound

and read this one...

Is DSD Hi-res ?

"So how high resolution is a DSD recording? There’s no exact way of striking equivalence between PCM and DSD but it’s around the same as 24-bit/88.2kHz PCM."

"In terms of dynamic range, DSD is claimed to be around 120dB in the audible frequency range. In comparison, CD measures around 96dB, and 24-bit/192kHz recordings have a theoretical maximum of around 144dB."
https://www.whathifi.com/advice/what-dsd-audio-how-it-works-where-to-download-files-and-more

 

[CiptoH]

And 100kHz freq response is nonsense too, studio microphones can’t even record anything in that range and human hearing doesn’t extend beyond 20kHz anyway.

The 100 KHz frequency response is not intended to be heard by humans, but so that the filter used can be of low-order (sloping) and not high-order (steep) filters like those used in CD players.
The low-order filter is much better (in terms of sound quality) and simpler than the high order filter.
The higher the filter order, the more phase shift of the audio signal at the output, and this is not good.

 

[gregorio]

The CD format only has 96 dB (16 bit) dynamic range !

16bit has a theoretical dynamic range of about 96dB but that’s excluding noise-shaped dither which increases it perceptually to about 120dB. With just 1bit, SACD/DSD has a theoretical dynamic range of only about 6dB but with very aggressive noise-shaped dither it increases perceptually also to around 120dB. 24bit has a theoretical dynamic range of about 144dB.

Again though, none of this makes any difference because the vast majority of recordings have a dynamic range of less than 50dB and the most dynamic recordings (symphony recordings) almost never have a dynamic range greater than about 60dB, which is about 1,000 times less than 16bit offers!

Human hearing doesn’t extend beyond 20kHz and if you’re an adult, probably not beyond 16kHz. So any resolution above that, is inaudible!

G

 

[gregorio]

The low-order filter is much better (in terms of sound quality) and simpler than the high order filter.

A slower roll-off filter is simpler but not better in terms of audible sound quality.

The higher the filter order, the more phase shift of the audio signal at the output, and this is not good.

There is no phase shift. For decades now linear phase filters have been used, which introduce zero phase shift, and this IS good!

G