Why do so few DACs support 88.2 and/or 176.4 kHz?

[K3cT]

I used to have a Lavry DA11 for several weeks and I can say that I'm not sure how I feel about it. Sure it's transparent and detailed but I don't know, music has a sharpness/edginess to it like you're forced to listen to the transition of notes. It was very apparent when I A/B-ed to AMB γ2 which seemed to have a more free-flowing sound which although perhaps not as technically capable, it's definitely more enjoyable to listen to.

I had a similar impression with a friend's Sabre-based Buffalo DAC and he has sold this DAC away ever since for similar reasons.

[nnotis]

I'm too lazy to dig up a link, but Dan Lavry has written a lot about high sample rate digital audio. He's even done so on Head-Fi. In short, he thinks that going above 96 Khz actually degrades performance. I’m pretty sure he stated his ideal sample rate as something around 60 Khz. That makes 88.2 the best pick from what’s available.

I do use a DA-10, which is connected to my Mac via an RME card that outputs either SPDF or Toslink/ADAT. When listening to anything in the standard 44.1 Khz sample rate, I double it into the Lavry to prevent any high frequency roll off. And I’m pretty sure that it does sound slightly better. I’ve never been diligent enough to do a blind test though.

If Dan Lavry is right, and I’ve not seen anyone challenge him, than all digital audio products with uber high sample rates are just gimmicks. Perhaps it’s a bit like saying your IEM has more drivers than the others do. It must reproduce sound more accurately then, right?

[grawk]

Thanks for your insight.

[Eric5676]

If Dan Lavry is right, and I’ve not seen anyone challenge him,

Yes and no. If you want your head to spin look up on his own forums and Gearslutz some of the old and back and forths from the past. Got pretty soap opera-ish.

Regardless of how someone may feel on the issue, they do owe it to themselves to look over some his musings on the subject.

...than all digital audio products with uber high sample rates are just gimmicks. Perhaps it’s a bit like saying your IEM has more drivers than the others do. It must reproduce sound more accurately then, right?

Good questions. I'm not sure where I fully stand on it myself. All I know is my 384kHz Cambridge 840c sounds good.

[nnotis]

I have previously spent some time reading his posts on gearslutz. But now I’ll have to go back and find dissenting opinions. I’d really be interested in hearing Cambridge explain the benefit of sampling at 384kHz. I’d like to give them the benefit of the doubt. But a justification seems warranted. Have you played around with the different rates? Do you notice any differences?

[Eric5676]

I have previously spent some time reading his posts on gearslutz. But now I’ll have to go back and find dissenting opinions. I’d really be interested in hearing Cambridge explain the benefit of sampling at 384kHz. I’d like to give them the benefit of the doubt. But a justification seems warranted. Have you played around with the different rates? Do you notice any differences?

Honestly? Nothing terribly earth shattering to my ears. Now that's hardly a scientific treatise and I wouldn't take that all the way to the bank sight unseen. Others may swear by it and it may be the case in their setups with their ears. I admit that I have to watch out myself with "Oh look! Higher numbers! Must mean it's better!" when that may not necessarily be the case.

As it stands, I did sell my 840c this morning because this buyer really made me an offer that I couldn't refuse, especially given that it was two years old and outside of warranty. I really can't believe it.

Whatever the 840c does overall is pretty darned good. It could be a little dry and sterile, especially towards the higher end. It might be a situation where some people would describe the sound as very "digital" instead of "analog-like."

That being said: I remember my Apogee Mini-DAC that preceded it that was "only" 24/192 instead of 384 and I'm going to be honest with you: It may have been just a smidge better if not at least equal to my ears. And that thing was $800 and change and the 840c is $1500 new on the street.

At the end of the day even I have to ask: How much can the human ear with even the best hearing really pick up and discern once you get past lots of loaded marketing rhetoric and "higher numbers must automatically be better"?

[Dusty Chalk]

At the end of the day even I have to ask: How much can the human ear with even the best hearing really pick up and discern once you get past lots of loaded marketing rhetoric and "higher numbers must automatically be better"?

My stance on this is well known, but I'll repeat it since you asked this question out loud -- I don't know whether or not the actual frequencies above the known human cutoff (which is age-dependent, amongst other things) can be heard, but I, for one, believe that their effects can be heard, albeit very subtly. My guess (purely talking out of my ass at this point, but when the mind does not have enough information, it makes stuff up -- it abhors a vacuum of knowledge) is that it does something to the phase of the signal, which the ear is known to be pretty sensitive to. Or something like that. Another guess is that once most of the harmonics in a triangle, square, or clarinet wave are heard, the mind's ear can hear a triangle, square, or clarinet wave, respectively, and therefore any discrepancies between what it thinks it should be hearing, and what it's actually hearing (what I call "negative hearing" -- not necessarily being able to hear to the level of detail of being able to know what's missing, but just the binary, "it sounds right"/"it doesn't sound right").

In all reality, it's probably some typically human complex combination of those things as well as others I haven't even figured out myself yet (IMO, again).

That said, upsampling isn't at that level of technology that it's going to correct phase, if the original information isn't there, so it's not going to fix that, though it may fix other things. Which is why keeping that information from the source to the point of delivery is more important than upsampling is at this point.

That said (yes, another one, on top of the other one), how much information is really there? The answer is almost none. Which is why (a) so few people can hear the difference (on top of just plain not caring or not being able to hear it anyway) and ( it doesn't matter all that much.

Pretty much everyone else on this board disagrees with me, and I honor that, but every once in a while, my frustration at not being able to explain my side of things that I have to try and express myself again. I'll try not to get too heated about it this time.

[Eric5676]

My stance on this is well known, but I'll repeat it since you asked this question out loud -- I don't know whether or not the actual frequencies above the known human cutoff (which is age-dependent, amongst other things) can be heard, but I, for one, believe that their effects can be heard, albeit very subtly.

Wouldn't surprise me. I have excellent hearing myself.

My guess (purely talking out of my ass at this point, but when the mind does not have enough information, it makes stuff up -- it abhors a vacuum of knowledge)

You can't go wrong with this kind of disclaimer.

... is that it does something to the phase of the signal, which the ear is known to be pretty sensitive to. Or something like that. Another guess is that once most of the harmonics in a triangle, square, or clarinet wave are heard, the mind's ear can hear a triangle, square, or clarinet wave, respectively, and therefore any discrepancies between what it thinks it should be hearing, and what it's actually hearing (what I call "negative hearing" -- not necessarily being able to hear to the level of detail of being able to know what's missing, but just the binary, "it sounds right"/"it doesn't sound right").

Certainly plausible.

In all reality, it's probably some typically human complex combination of those things as well as others I haven't even figured out myself yet (IMO, again).

Someone has to say it: Placebo. I see that higher number and I expect it to sound better.

That said, upsampling isn't at that level of technology that it's going to correct phase, if the original information isn't there, so it's not going to fix that, though it may fix other things. Which is why keeping that information from the source to the point of delivery is more important than upsampling is at this point.

Sure, I can agree with that.

That said (yes, another one, on top of the other one), how much information is really there? The answer is almost none. Which is why (a) so few people can hear the difference (on top of just plain not caring or not being able to hear it anyway) and ( it doesn't matter all that much.

It's like talking about upscaling/upconverting regular DVDs. Sure it makes the image look "better" to a point, but if the source material is crap to begin with then it's "garbage in, garbage out." And it's still never going to look as good as real high definition.

Pretty much everyone else on this board disagrees with me, and I honor that, but every once in a while, my frustration at not being able to explain my side of things that I have to try and express myself again. I'll try not to get too heated about it this time.

I think I know what you're trying to say. Certainly plausible IMO. Since I'm not an ENT or some other high end audiologist, I'll be happy to salute your "pull out of my ass" disclaimer and think you did a pretty good job here.

[Eric5676]

Dusty and others: How do you feel about NOS DACs? Like something from Red Wine Audio — battery–powered purist hi–fi

This: Products > Isabellina - Red Wine Audio Comes in several variants. I've been reading up on this company and their products and I admit to being quite fascinated.

[Dusty Chalk]

Sure, why not. I mean, I'm a big fan of the old MSB Platinum DAC that Justin used to haul around, which is also a ladder DAC, IIRC.

Here's the thing: I'd rather have a really good 16/44.1 DAC for 16/44.1 source material than a mediocre upsampling one, but if I have 24/96 material (or whatever) -- which I do -- I, personally, would rather hear it at its original bitrate and bitdepth.

That said, I have had good luck with (another MSB product, the Link DAC III) upsampling as well. Perhaps because you're starting with a signal that's closer to the analog signal before the analog stage, that makes it easier...somehow...? Again, making stuff up, because I don't know the real explanation. All I know is that I like the sound.

What really matters is that it's a well-designed DAC with a well-designed output stage more than anything. And that if you have other source material than CD (16/44.1), that it handles it well.

[morphsci]

Dan Lavry writes white papers and publishes them on his site. That is not a peer reviewed scientific paper so take it for what it is, an opinion, dressed up to look like a scientific explanation.

[grawk]

Dusty and others: How do you feel about NOS DACs? Like something from Red Wine Audio — battery–powered purist hi–fi

This: Products > Isabellina - Red Wine Audio Comes in several variants. I've been reading up on this company and their products and I admit to being quite fascinated.

red wine audio products are often very expensive for what they are, IMHO.

[Eric5676]

Here's the thing: I'd rather have a really good 16/44.1 DAC for 16/44.1 source material than a mediocre upsampling one, but if I have 24/96 material (or whatever) -- which I do -- I, personally, would rather hear it at its original bitrate and bitdepth.

That said, I have had good luck with (another MSB product, the Link DAC III) upsampling as well. Perhaps because you're starting with a signal that's closer to the analog signal before the analog stage, that makes it easier...somehow...? Again, making stuff up, because I don't know the real explanation. All I know is that I like the sound.

What really matters is that it's a well-designed DAC with a well-designed output stage more than anything. And that if you have other source material than CD (16/44.1), that it handles it well.

I basically agree with you. I bolded the part that I can't see anyone disagreeing with. It's all about implementation in the end. It's either a good piece of equipment or it isn't.

red wine audio products are often very expensive for what they are, IMHO.

Yup. A little more research was starting to reveal that to me a little bit.

Dan Lavry writes white papers and publishes them on his site. That is not a peer reviewed scientific paper so take it for what it is, an opinion, dressed up to look like a scientific explanation.

True although it's quite educational. Some of it does go over my head since I'm not an engineer.

Although I tell you: It takes some hotshots to try and really be able to swim through it and pick some of it apart and attempt to refute it. That is when you can get the aforementioned soap opera on his forums and the gearslutz formus from the past. Shoot, it might have even crossed over onto a couple of other audio forums, too. Man, that was some epic stuff. 90 percent of the engineering and techno jargon was flying over my head but it was still entertaining as all hell.

[Dusty Chalk]

I basically agree with you. I bolded the part that I can't see anyone disagreeing with. It's all about implementation in the end. It's either a good piece of equipment or it isn't.

Yeah, sorry if that was condescending -- it wasn't meant to be. It just sometimes bears repeating when we get too carried away discussing minutiae such as what chips are used, upsampling, etc. -- just to bring the focus back to what really matters.

[Sherwood]

The problem comes with how difficult it is to describe the fitness of an output stage in a bullet point. One has to either be Colin (or Marc, or Naaman, or anyone but me), or just buy the thing and listen to it. And if there's one thing I hate about music playback equipment, it's listening to music.

[The Monkey]

Music is for suckers.

[Eric5676]

Yeah, sorry if that was condescending -- it wasn't meant to be. It just sometimes bears repeating when we get too carried away discussing minutiae such as what chips are used, upsampling, etc. -- just to bring the focus back to what really matters.

Agreed. And I didn't think you were condescending.

At some point it comes down to what a wise man told me in another thread here not too long ago: "Or, you could just buy stuff."

That's the only way to really be sure.

[Dusty Chalk]

electronic beeps and whistles, randomly generated, is where it's at.

So...Autechre?

[Sherwood]

One could call Brian Eno.

[Dusty Chalk]

I'm surprised larry fast's pink notes isn't mentioned on that page.

Sent from mah Droid

[jp11801]

I respect Dan Lavry but think he is off base at least as it pertains to my experience with the Sonic Studio 305 at 192 rate it bests the 96 rate when used as a phono pre amp.

[nnotis]

One thing that makes this discussion all the more confusing is that different converters probably perform better at different rates. Even if Lavry is right, I don’t doubt there are DACs out there that give their best audible performance at 192 kHz.

There is one thing that would turn me away from a DAC that only handles 16 bits though. If you run a 24 bit file into it, truncation will occur. And I can say confidently from my own experience, truncated files sound noticeable grainier and more congested than the 24 bit originals. This is due to quantization distortion.

I wonder if anyone has built a DAC that automatically dithers in real time if receiving a 24 bit signal?

[Dusty Chalk]

No, but if you can track down a GW Labs DSP, it'll downconvert.