NOS DAC sound more natural

[Hiro]

I hope that everybody realizes that NOS PCM DACs produce lots of aliasing artifacts. That said, if anyone insists on listening to 44.1 pcm in NOS mode, no one is stopping them...

[toddrhodes]

Hiro, unfortunately what you said is a little out of my depth. I did 30 seconds of googling and found this (yea, it's wiki):

 

Temporal aliasing is a major concern in the sampling of video and audio signals. Music, for instance, may contain high-frequency components that are inaudible to humans. If a piece of music is sampled at 32000 samples per second (Hz), any frequency components above 16000 Hz (the Nyquist frequency) will cause aliasing when the music is reproduced by a digital to analog converter (DAC). To prevent this an anti-aliasing filter is used to remove components above the Nyquist frequency prior to sampling.

 

So if this aliasing happens above the Nyquist frequency of Redbook, wouldn't we not hear it?

[Miska]

So if this aliasing happens above the Nyquist frequency of Redbook, wouldn't we not hear it?

 

Intermodulation may make those frequencies audible...

 

Depending on analog reconstruction filter (or lack of) on a NOS DAC, those image frequencies can span far to MHz range.

 

Whole point of oversampling is to help those analog reconstruction filters remove more of the image frequencies.

[Jud]

Hiro, unfortunately what you said is a little out of my depth. I did 30 seconds of googling and found this (yea, it's wiki):

 

Temporal aliasing is a major concern in the sampling of video and audio signals. Music, for instance, may contain high-frequency components that are inaudible to humans. If a piece of music is sampled at 32000 samples per second (Hz), any frequency components above 16000 Hz (the Nyquist frequency) will cause aliasing when the music is reproduced by a digital to analog converter (DAC). To prevent this an anti-aliasing filter is used to remove components above the Nyquist frequency prior to sampling.

 

So if this aliasing happens above the Nyquist frequency of Redbook, wouldn't we not hear it?

 

As Miska has pointed out, there is also intermodulation; but back to aliasing for a moment:

 

You wouldn't hear aliasing (that is, the audible "mirror" frequencies of the ultrasonics) with a perfect filter. Of course there is no perfect filter, so the object is to drop the level of aliasing distortion below audibility, while at the same time not introducing audible time- or phase-based distortions. It is hard to design an effective, affordable filter operating as a "brick wall" at 22.05kHz to do this that will live inside a DAC. It is easier to design such a filter (1) if it can start operating at higher frequencies with a more gentle, non-"brick wall" slope; and (2) if it can live inside a computer with access to all the power of the CPU(s).

[Norton]

Hi Norton, I went with the standard dual Burr Brown option. I do like that I can change out boards for different sound if I am so inclined in the future though I have not inquired as to the cost of those boards...

 

Thanks. What's the detail retrieval like compared to other DACs you've owned? I listen 100% to classical and detail is really what I look for in a DAC.

 

I've tracked down a used La Voce 2 at a good price so very tempted....

[toddrhodes]

Thanks. What's the detail retrieval like compared to other DACs you've owned? I listen 100% to classical and detail is really what I look for in a DAC.

 

I've tracked down a used La Voce 2 at a good price so very tempted....

 

I was listening to Black Sabbath - War Pigs last night (I know, same as classical right? ) and for the first time I can remember, I could hear Ozzy's voice bouncing off walls on both sides of me. And I'm not talking about reflections, this was recorded information that very clearly defined the space in which he sang the song. It's subtle, but completely noticeable once you hear it.

 

I have to be honest I wasn't looking for detail last night as I was more trying to feel the music. With this DAC I find it much easier to unfocus my eyes (I find this to work better than closing them) and just feel my way through everything that's happening. It's a very engaging piece of equipment. To be fair, I'm also trying out a new amp so I'm sure that's contributing to my experience. It's not a super highend amp either, it's a Nuprime ST-10 ($1600), so this probably similar or even lower-fi than what a lot of folks here have. So anyway, I definitely caught more detail in songs I'm just having a hard time remembering specifics today. And last night was all about playing through my "old standards." Those songs I've heard a hundred times... in the last 6 months. I'm not exaggerating though, last night they all felt new again.

[groovybassist]

Can someone explain what an "aliasing artifact" actually sounds like during playback?

 

I've heard all the intellectual arguments on both sides, but have found I personally feel a deeper connection to the music and the artists' intent when listening to NOS DACs. I simply want to listen more and longer. I typically don't have that same response to DACs that perform various types of processing, up/oversampling or filtration.

 

This isn't a statement about every DAC, simply what I'm willing to put my hard earned money behind.

 

FWIW, my DAC uses the AD1865N-K chip.

[Jud]

The aliasing due to leakage through a reasonably designed filter should be low enough in level that it's not obnoxious (except to some people over longer listening sessions, and a few over shorter sessions), and it will mostly be at quite high frequencies. People may experience this very high frequency energy as increased detail, air, warmth or energy. (Ringing may potentially increase the sense of air or space.) They may experience its absence as a lack of these things or as a quieter, "blacker" background with increased clarity.

[groovybassist]

This is interesting as my personal experience with NOS DACs is that they are quieter and have more natural detail than OS DACs, which sounds like the opposite of what you're describing. I find non-NOS (if that makes sense) as having forced detail, while sacrificing the flow of music.

[toddrhodes]

This is interesting as my personal experience with NOS DACs is that they are quieter and have more natural detail than OS DACs, which sounds like the opposite of what you're describing. I find non-NOS (if that makes sense) as having forced detail, while sacrificing the flow of music.

 

Groovy, what NOS DACs have you had experience with? I've only experienced one and I can't see myself going back. I'll take my Gungnir to work for my headphones and can only hope that NOS, Ladder/R2R DACs get better and come down in price. And, as with all things, it's probably the implementation that defines the experience, not the simple fact a DAC is NOS or isn't. I also avoid any unnecessary upsampling/DSP/etc...

[groovybassist]

I've owned three NOS DACs - the Red Wine Audio Bellina, Mojo Audio Mystique v1.0 and now the Mojo Audio Mystique v2.0, which is definitely the best DAC I've owned. I've had a few others in the past - a couple of ESS Sabre based and one Burr Brown.

 

I agree implementation is a very big determinant of sound quality, but for some reason, NOS DACs help me relax into the music - others, not so much. Wish I could try 'em all, but that's not so easy these days.

[toddrhodes]

Thanks. What's the detail retrieval like compared to other DACs you've owned? I listen 100% to classical and detail is really what I look for in a DAC.

 

I've tracked down a used La Voce 2 at a good price so very tempted....

 

So I just got through the first "track" on Rachmaninoff - Symphonic Dances (Dallas Symphony Orchestra) and I cannot more highly recommend something for classical music. It does a very good job, IMO of course, of placing instruments in space in all three dimensions. I really don't think you'll want for detail either. It's a special unit. But, full disclosure - my experience with such DACs is very limited, this La Voce being the best DAC I've owned. I've heard a much better one - dCS Vivaldi, but that's a silly comparison and I only heard the Vivaldi for about 5 minutes in an amazing room. Though it may be sacrilege to say, I daresay the musicality I get on the one track I've heard both on? I like the aqua more /runs and hides.

 

I thought of one more observation that might be relevant here - Jack Johnson - Middle Man. About 2 minutes into the track there are some bongos or something similar that show up in the right channel. With the Aqua, I can very distinctly place each individual bongo in its own space. This includes ones that are behind the others. It's a treat! I read about this track on a review, I think on DAR but it could have been on one of Chris's posts, and I had to check it out for myself. I didn't get that with my old DAC but now, it's plain as day.

[Norton]

Thanks again, definitely on my list now.

 

There's a used one for sale at a dealer about an hour's drive from where I 'm currently on holiday, but frustratingly it looks like they are shut for the next few weeks...

[Jud]

This is interesting as my personal experience with NOS DACs is that they are quieter and have more natural detail than OS DACs, which sounds like the opposite of what you're describing. I find non-NOS (if that makes sense) as having forced detail, while sacrificing the flow of music.

 

A couple of images from the "SRC [sample rate converter/conversion] Comparisons" web site (SRC Comparisons), just to give a visual "feel" of what signals without and with aliasing energy look like.

 

This is a "sweep" test. First a converter/filter that does relatively well with regard to aliasing:

 

 

Now one that does relatively poorly:

 

[groovybassist]

Intellectually interesting, but I'm not sure how the converters listed relate to the actual conversion happening in a given DAC, and still don't have a good feel for audibility or audible cues.

[toddrhodes]

Intellectually interesting, but I'm not sure how the converters listed relate to the actual conversion happening in a given DAC, and still don't have a good feel for audibility or audible cues.

 

Moral of story: Turn on music. Like? Continue listening. Else? Up your specs, dude.

 

Based on what I read in a previous post, it almost seems like aliasing is a psychoacoustic effect that may not really be audible (?) but can have a gradually straining effect on the listener. I don't get that out of my NOS DAC after 5+ hours of listening at various levels so I'm calling it good, at least for me. Regardless, more knowledge is NOT a bad thing.

[Jud]

Yep, not trying to tell anyone not to enjoy what they enjoy. Just providing a picture of what aliasing looks like with one filter/converter.

 

Clearly it's not audible (at least not subjectively as distortion) for those who enjoy NOS DACs.

[groovybassist]

I'm very happy with my NOS DAC and feel no need to up my specs. Just trying to understand audibility of aliasing and why folks that are negative on NOS DACs are always trotting aliasing out like it's some obvious distortion we unwitting NOS supporters aren't hearing.

 

My interest in listening to music and the length of my listening are both greater when listening NOS - guess that says it all.

[Jud]

folks that are negative on NOS DACs are always trotting aliasing out like it's some obvious distortion

 

Well, if you're looking at measurements or scope traces it can literally be as obvious as the comparison between the two pictures I posted above. That's not to say it must necessarily be either audible at all to everyone, or even for those to whom it may be audible, necessarily bad sounding.

 

Some NOS DACs are bad sounding to some people - the criticism of early CD players was essentially a criticism of the sound of the NOS DACs of the day built into CD players, and was the reason why the industry soon standardized on 8x oversampling DAC chips. But lots of folks these days love the way their NOS DACs play music, many of them people whose taste I greatly respect, and how can I (and why would I) tell them they're wrong? A whole lot of this hobby comes down to matters of taste.

[accwai]

[...] Some NOS DACs are bad sounding to some people - the criticism of early CD players was essentially a criticism of the sound of the NOS DACs of the day built into CD players, [...]

 

What are the current NOS DACs that you find are showing audible distortion as a result of the aliasing artifact?

[Jud]

What are the current NOS DACs that you find are showing audible distortion as a result of the aliasing artifact?

 

I have no idea - the only one I've heard is the Phasure NOS1a, and that was exclusively with the manufacturer's XXHighEnd software providing oversampling/filtering in the PC prior to sending the signal to the DAC, as he recommends.

 

This is yet another very good reason why I cannot tell anyone that aliasing must be either (1) audible to them, or (2) if audible, bad sounding.

 

That said, I have done a lot of experimentation with oversampling filters in Audirvana Plus, and have set up some filters that *did* allow aliasing. These didn't sound bad to me. When I looked at "pictures" of what my filters were doing and altered them to better eliminate aliasing, I experienced the changes I noted in post #60 above.

[groovybassist]

Perception is a funny thing. I can point to a number of OS DACs I think sound worse than NOS. Of course, OS DAC proponents will say the current measurements support their position. The fact we have people on both sides of this argument indicates we don't yet fully understand the signal/ear/brain interface well enough.

[Jud]

The fact we have people on both sides of this argument indicates we don't yet fully understand the signal/ear/brain interface well enough.

 

One thing that's evident is that there are as many signal/ear/brain interfaces as there are people, and they vary, possibly quite a lot. So do the systems in which our DACs operate. Thus we can have a large group of folks in the 80s and very early 90s saying the NOS DACs in CD players of the era sounded "digital," and a large group of folks here in the 'teens saying NOS DACs are a relief from the "digital" sound of delta-sigma oversampling DACs.

[toddrhodes]

One thing that's evident is that there are as many signal/ear/brain interfaces as there are people, and they vary, possibly quite a lot. So do the systems in which our DACs operate. Thus we can have a large group of folks in the 80s and very early 90s saying the NOS DACs in CD players of the era sounded "digital," and a large group of folks here in the 'teens saying NOS DACs are a relief from the "digital" sound of delta-sigma oversampling DACs.

 

It could just be a matter of product evolution too. Perhaps the output stages on these NOS DACs (and IMO, NOS seems to start showing up more once you crest the $2k threshold, so IMO starting to get into the lower tier of "high end" components), along with power supplies and just chip implementation themselves are better than what were in those original NOS CD players.

 

I'm curious, is there any evidence to support that NOS is a better fit for 24/96 and up "high res" files? That maybe there is so much information in those files, the need for oversampling just doesn't exist? I'm talking out of my depth here, admittedly. And regardless of the answer, I do prefer redbook through this new DAC than through my old D-S so my question is merely a talking point, not really seeking a point to prove.

[Miska]

I'm curious, is there any evidence to support that NOS is a better fit for 24/96 and up "high res" files? That maybe there is so much information in those files, the need for oversampling just doesn't exist? I'm talking out of my depth here, admittedly. And regardless of the answer, I do prefer redbook through this new DAC than through my old D-S so my question is merely a talking point, not really seeking a point to prove.

 

If you download 176.4/192k hires files, they are effectively 4x "oversampled" already. If you download 352.8k DXD files, they are 8x "oversampled" which is equivalent of the oversampling ratio in most oversampling DACs. Difference is mostly that there is actually also content in the higher frequencies too, compared to something that has been converted from lower sampling rates.

 

So more hires you go with, less you need oversampling and more you sort of close gap to the oversampling DACs.