What set of measurements can equate two DACs?

[musicapristina]

As a manufacturer of audio components, including DAC’s, are you interested in this topic so you can improve your designs?

Very simply, yes. There's much more to it than that, though.

I was told in 2010 that linear regulated power supplies made digital audio equipment sound better. I didn't believe it. And then I wanted to prove it wrong. So I started making stuff, started looking at everything on a scope, and started attempting to AB test the stuff I made. I haven't stopped.

 

[pkane]

Could you possibly provide a more generic answer?
I’m not buying it. DAC linearity sounds like something measurable but not necessarily consequential.

Nonlinearity is what generates harmonic distortion. If you think HD can be consequential, then nonlinearity can be consequential. There’s no separating the two.

 

[musicapristina]

@Gregss and @Jimbob54 and @AdamG247 and @mansr, I'm sorry if I've offended you. My question(s) are entirely sincere. I'm not trying to hide anything. I'm not here to sell anyone anything. I'm not here to freeload (I made my donation).

There's a "Wanted" post on the home page, it applies to me, so I'm here.

 

[Jimbob54]

@Gregss and @Jimbob54 and @AdamG247 and @mansr, I'm sorry if I've offended you. My question(s) are entirely sincere. I'm not trying to hide anything. I'm not here to sell anyone anything. I'm not here to freeload (I made my donation).

There's a "Wanted" post on the home page, it applies to me, so I'm here.

No problem. Send @amirm one of your devices for testing.

 

[AdamG]

I will pony up for return shipping cost.

 

[BluesDaddy]

@musicapristina I'd love to hear, without any marketing BS or purely subjective poetic descriptions, what makes your company's Virtuoso PCM DAC worth $6,500. Also any controlled listening test (i.e. some form of DBT) that have been conducted to validate any subjective impressions.

 

[preload]

Nonlinearity is what generates harmonic distortion. If you think HD can be consequential, then nonlinearity can be consequential. There’s no separating the two.

Thanks for clarifying, @pkane. I think to be clear, if we already have direct measurements of HD, and HD is extremely low, why would we worry about linearity measurements? In other words, how is the linearity measurement relevant if THD is -130dB?

 

[preload]

@musicapristina I'd love to hear, without any marketing BS or purely subjective poetic descriptions, what makes your company's Virtuoso PCM DAC worth $6,500. Also any controlled listening test (i.e. some form of DBT) that have been conducted to validate any subjective impressions.

It appears that the Virtuoso DAC uses chassis-mounted RCA jacks, which suggests SOTA performance. Unfortunately, I can't tell from photos if the front panel is a thick slab of aluminum, or if it's a thin sheet folded over to look like a thick slab. Therefore, I cannot definitively determine the product's performance.

 

[BluesDaddy]

It appears that the Virtuoso DAC uses chassis-mounted RCA jacks, which suggests SOTA performance. Unfortunately, I can't tell from photos if the front panel is a thick slab of aluminum, or if it's a thin sheet folded over to look like a thick slab. Therefore, I cannot definitively determine the product's performance.

I can read the description on the website for myself. And since when does a "thick slab of aluminum" or even "chassis mounted RCA jacks" equate to that premium of a price?

 

[preload]

I can read the description on the website for myself. And since when does a "thick slab of aluminum" or even "chassis mounted RCA jacks" equate to that premium of a price?

Well, sir, would you pay $6,000 for a DAC built in a plastic enclosure with panel-mounted RCA jacks? I think not!

 

[BluesDaddy]

Well, sir, would you pay $6,000 for a DAC built in a plastic enclosure with panel-mounted RCA jacks? I think not!

Do you really think that anything (edit: Please note that I intended to type "everything" here - since there's a response quoting the original, I won't correct it) UNDER $6,500 is made with a plastic enclosure and panel mounted jacks? Seriously? There's a whole lot of room under there and you've just set up a false dichotomy. My question was directed at the manufacturer, but thanks for playing.

 

[ebslo]

I think to be clear, if we already have direct measurements of HD, and HD is extremely low, why would we worry about linearity measurements? In other words, how is the linearity measurement relevant if THD is -130dB?

I think a large linearity error in a very small number of codes could be obscured by the THD measurement. A 24-bit DAC has over 16 million possible codes. A dataset large enough for most codes to be represented will also be large enough that a large error in a single code may not significantly increase the THD of the entire dataset.

 

[preload]

Do you really think that anything UNDER $6,500 is made with a plastic enclosure and panel mounted jacks? Seriously?

Actually yes, there are quite a few DACs available under $6,500 that come in a plastic enclosure + PCB mounted jacks. I could provide you with a few examples, but as you stated, you are capable of "reading the description on the website for yourself."

There's a whole lot of room under there and you've just set up a false dichotomy. My question was directed at the manufacturer, but thanks for playing.

Fact: my response does not prevent the manufacturer from responding, and "you're welcome, any time!" There's also a private message feature here if you didn't want others to be part of your personal conversation with the manufacturer.

 

[preload]

I think a large linearity error in a very small number of codes could be obscured by the THD measurement. A 24-bit DAC has over 16 million possible codes. A dataset large enough for most codes to be represented will also be large enough that a large error in a single code may not significantly increase the THD of the entire dataset.

Interesting. Is such a situation common in commercially available, modern DACs? What would be the resulting linearity measurement in such a situation?

 

[BluesDaddy]

Actually yes, there are quite a few DACs available under $6,500 that come in a plastic enclosure + PCB mounted jacks. I could provide you with a few examples, but as you stated, you are capable of "reading the description on the website for yourself."



Fact: my response does not prevent the manufacturer from responding, and "you're welcome, any time!" There's also a private message feature here if you didn't want others to be part of your personal conversation with the manufacturer.

I actually mistyped and meant "EVERYTHING" rather than anything. I won't correct it now, but I'm pretty sure you probably knew what I really meant.

 

[pkane]

Thanks for clarifying, @pkane. I think to be clear, if we already have direct measurements of HD, and HD is extremely low, why would we worry about linearity measurements? In other words, how is the linearity measurement relevant if THD is -130dB?

Nonlinearity is the cause, HD is the effect. For me personally, this is critical, but probably not so for most. If anything, measuring both gives a confirmation that the individual measurements were not an accident or otherwise flawed since the methods are not the same.

 

[ebslo]

Interesting. Is such a situation common in commercially available, modern DACs?

No, not common, unheard of even (at least by me). Almost all DAC's these days use a chip from a major manufacturer such as Texas Instruments or Analog Devices. Since linearity is so fundamental to the operation of a DAC, these manufacturers test it well. I would be absolutely shocked if they fabricated a chip with such an error, and it is highly unlikely that one would be introduced by a user of these chips. Nevertheless, I feel it is important that it remain in the test suite. Mistakes happen, and I bet someone has or will do something crazy like roll their own DAC out of an FPGA and some discreet components,

 

[richard12511]

Not necessarily snake-oil, it just won't sound different/better.



They don't have to be equal, they just have to be beyond the threshold of transparency. Beyond that, it's not about SQ any longer.

Strict threshold might be anything over (under?) -115dB sinad. Realistic threshold is likely closer to -90dB or less with normal music in a normal room.

Beyond some point, the differences in the signals are far below what could be detected with ears.

I'd even say -90dB is still too quiet. 90 might be hearable for extremely well trained individuals with access to instant switching. I'd guess it's more like -60dB to -70dB under more typical conditions(not trained and trying to compare it to the component you had previously).

I'd be really interested in learning the SINAD of the cheap $400 Yamaha AVR I used in our blind listening test against a class D Nord Hypex and class A/B Emotiva. Given the typical AVR performance we've seen, I can't imagine it's better than -80dB, and likely much worse. Whatever it was, it was good enough to where none of us could hear a difference.

 

[ebslo]

I'd even say -90dB is still too quiet. 90 might be hearable for extremely well trained individuals with access to instant switching. I'd guess it's more like -60dB to -70dB under more typical conditions(not trained and trying to compare it to the component you had previously).

I'd be really interested in learning the SINAD of the cheap $400 Yamaha AVR I used in our blind listening test against a class D Nord Hypex and class A/B Emotiva. Given the typical AVR performance we've seen, I can't imagine it's better than -80dB, and likely much worse. Whatever it was, it was good enough to where none of us could hear a difference.

Depends if it's used before or after the volume control. If it feeds a pre-amp you're probably right. If it directly feeds a power amp and relies on digital volume control then it needs to cover the dynamic range of the music + volume control; in this case it should have at least as much dynamic range as the power amp.

 

[JSmith]

Given 2 DACs have Noise and Distortion measurements below -115 dB SINAD

It's not all about the SINAD figure and never has been... you do look at the reviews right?

one selling for $299 and one for $29,999

Well that's just silly... the chips are like $5 - $35.



JSmith