HOW DOES A PERFECT DAC ANALOG SIGNAL LOOK DIFFERENT THAN A CHEAP DAC

[beerandmusic]

To remind people (as if you haven't been reminded too much already), I am the person that caught a lot of slack for stating I have heard very little differences in DACs and it wasn't until I upgraded my amplifier that I could say that is a substantial upgrade in sound.  I have tried cheap ifi's to schiits to Marantz to Teac to Sony to Chords to Korg to even using embedded dacs on recievers.  I have done DLNA, DSD, and PCM.  Maybe I am not a "true audiophile" but even though i may have heard differences, they were minimal, not substantial, and i couldn't say one sounded better than another. 

 

I sold most of my dacs (still have 3), but i do plan on buying one more, but I will be patient in my next purchase after spending over a year of trying different dacs only to be less than impressed.

I have been more impressed with software improvements, speaker improvemts, amp improvements, eq improvements, but dac improvements have been next to nil.

My next dac will have galvanic isolation and a volume control and good choice of inputs, and will support DSD and will be less than $1500...I might have to wait, but i haven't heard anything dramatic worth the investment imho.

 

Anyways on to my question....i am going to try and to make it as simple as possible.

 

Let's assume my binary bits for my song looks like this

1010101011100011100110101010101 (obviously it would be much more complex)

and we want to convert that into a sine wave to be amplfied and output to the speakers.

Assuming i am happy with my amplfier and speakers, the only task left is to

perfectly transform those bits into an accurate wave form without any noise.

Assuming galvanic isolation takes care of the usb noise, then the only task is to

accurately convert the bits into the accurate waveform.

 

my question is how much different will the waveform be between a 50K DAC and a 1K DAC?

And why can't it be measured and compared?

 

Right now i am thinking the LKS mh-da004 is best bang for the buck.

 

 

 

 

 

[fas42]

3 hours ago, beerandmusic said:

Assuming galvanic isolation takes care of the usb noise, then the only task is to

accurately convert the bits into the accurate waveform.

 

my question is how much different will the waveform be between a 50K DAC and a 1K DAC?

And why can't it be measured and compared?

 

 

The answers are, a little bit, just enough to be audible in the right context; and, also because it's a little bit! IOW, the variations are subtle and hard to pick up in a measurement sense, but mean everything in the subjective experience sense.

 

DACs probably more than another component will behave differently depending upon the rest of the system and environment they're in. At the moment, to fully measure every aspect of their performance that counts is essentially an impossible task, so the usual practice of just trying various units until a good compromise is found is probably as good a way as any.

[wushuliu]

Why don't you ask the engineers of different chips like akm and ess who firmly believe and design accordingly to what they consider good sound. You can find a video presentation for ESS on YouTube and online interviews and discussions with akm engineers. 

 

I look forward to this being ignored.

[gmgraves]

3 hours ago, beerandmusic said:

To remind people (as if you haven't been reminded too much already), I am the person that caught a lot of slack for stating I have heard very little differences in DACs and it wasn't until I upgraded my amplifier that I could say that is a substantial upgrade in sound.  I have tried cheap ifi's to schiits to Marantz to Teac to Sony to Chords to Korg to even using embedded dacs on recievers.  I have done DLNA, DSD, and PCM.  Maybe I am not a "true audiophile" but even though i may have heard differences, they were minimal, not substantial, and i couldn't say one sounded better than another. 

 

I sold most of my dacs (still have 3), but i do plan on buying one more, but I will be patient in my next purchase after spending over a year of trying different dacs only to be less than impressed.

I have been more impressed with software improvements, speaker improvemts, amp improvements, eq improvements, but dac improvements have been next to nil.

My next dac will have galvanic isolation and a volume control and good choice of inputs, and will support DSD and will be less than $1500...I might have to wait, but i haven't heard anything dramatic worth the investment imho.

 

Anyways on to my question....i am going to try and to make it as simple as possible.

 

Let's assume my binary bits for my song looks like this

1010101011100011100110101010101 (obviously it would be much more complex)

and we want to convert that into a sine wave to be amplfied and output to the speakers.

Assuming i am happy with my amplfier and speakers, the only task left is to

perfectly transform those bits into an accurate wave form without any noise.

Assuming galvanic isolation takes care of the usb noise, then the only task is to

accurately convert the bits into the accurate waveform.

 

my question is how much different will the waveform be between a 50K DAC and a 1K DAC?

And why can't it be measured and compared?

 

Right now i am thinking the LKS mh-da004 is best bang for the buck.

 

 

 

 

 

 

Not enough to be noticed on an oscilloscope directly. You would be able to see factors which would/could/should affect the sound by running a full set of tests on the DAC in question, and even those wouldn't necessarily tell you what the DAC actually sounds like. For Instance John Atkinson tested the Schiit Yggdrasil DAC (US$2300) and found the measurements not altogether to his liking. Yet his subjective reviewer loved it's sound, as did The Absolute Sound's Robert Harley, who literally gushed all over the page with superlatives. He couldn't say enough good about the piece (and I concur. It is by far the best DAC I have ever heard!).  So even if you could measure different DACs exhaustingly, it would tell you little about the sonic differences between different examples. In this case, let your ears tell you what you need to know. This is one case where, since you would, ostensibly, have no preconceived expectational bias for any of the DACs being considered, your ear/brain is unlikely to lie to you. 

[beerandmusic]

11 minutes ago, wushuliu said:

Why don't you ask the engineers of different chips like akm and ess who firmly believe and design accordingly to what they consider good sound. You can find a video presentation for ESS on YouTube and online interviews and discussions with akm engineers. 

 

I look forward to this being ignored.

 

my guess is that one chip manufacturer or the other does rely more on measurements than a subjective sound comparison.  I also believe one chip or the other believes they are inferior and is trying to engineer better.

 

[beerandmusic]

3 minutes ago, gmgraves said:

 

Not enough to be noticed on an oscilloscope directly. You would be able to see factors which would/could/should affect the sound by running a full set of tests on the DAC in question, and even those wouldn't necessarily tell you what the DAC actually sounds like. For Instance John Atkinson tested the Schiit Yggdrasil DAC (US$2300) and found the measurements not altogether to his liking. Yet his subjective reviewer loved it's sound, as did The Absolute Sound's Robert Harley, who literally gushed all over the page with superlatives. He couldn't say enough good about the piece (and I concur. It is by far the best DAC I have ever heard!).  So even if you could measure different DACs exhaustingly, it would tell you little about the sonic differences between different examples. In this case, let your ears tell you what you need to know. This is one case where, since you would, ostensibly, have no preconceived expectational bias for any of the DACs being considered, your ear/brain is unlikely to lie to you. 

i believe that it is highly likely that one dac may "subjectively" sound better even though it's measurements are worse...example, some people also like more bass, but that doesn't mean it is closer to original.  I also know that i typically prefer warm than clinical, but that it is very likely that the more clinical sound is closer to the digital input.

 

[cjf]

There are many measured tests of high dollar dacs and cheap dacs on the web that one could compare against in an objective manor. Stereophile's website would probably be the best place to look first assuming the person comparing knows what the various measurements mean.

 

I think some of the biggest objective differences lie in the measurements that show bits of resolution, channel separation, jitter, bandwidth and noise measurements between the various price points.

 

Then you need to consider what some say are the most important differences which are all subjective in nature between the various products. I'm a firm believer that one wont hear the big differences between the cheap and good products until they have all their other supporting gear at a level where such differences can be noticed. It will all sound more or less the same until then.

 

A disappointing outcome should be of no supprise when placing a $1K DAC into a $50K system and vise versa when placing a $50K DAC into a $1K system.

[beerandmusic]

19 minutes ago, cjf said:

I think some of the biggest objective differences lie in the measurements that show bits of resolution, channel separation, jitter, bandwidth and noise measurements between the various price points.

 

 

 

 

This is more in line what i was thinking about....my curiosity is how much different will a $5K dac measure than a $1500 Dac (i use $1500 arbitrarily because that is the price of a current contender I am contemplating...MH-DA004).  Since I don't really consider myself an audiophile, because I can't say one DAC sounds better than the other in the same way i can say a better amp makes a huge difference....my ears must not be that good....so I am going with wanting a DAC that has features I want and is objectively on par or better..   I can hear subtle differences, but nothing i consider worth paying more for.  I have good "space" and "depth" with a good amp, and it wasn't until i got a good amp until i recognized that, but swapping out dacs has made little difference.

[beerandmusic]

13 minutes ago, cjf said:

A disappointing outcome should be of no supprise when placing a $1K DAC into a $50K system and vise versa when placing a $50K DAC into a $1K system.

I am not sure i buy this...i have gone to a few audio shows listening to 20K to 300K systems, and the dac seemed little do with any of the systems overall sound.  I heard some great systems where they used a media server with embedded dac, some using cheap dacs where the speakers and amp made all the difference....i was like, really you are using that cheap dac and getting that sound.... (btw i really love the sound of focal, ess, and atc speakers).

[jabbr]

This is very debatable and there is no single agreed upon answer. One of the problems is that some measurements are hard to do, take a lot of time and require what might be expensive equipment, so nothing even close to all possible measurements have been done.

[cjf]

4 minutes ago, beerandmusic said:

I am not sure i buy this...i have gone to a few audio shows listening to 20K to 300K systems, and the dac seemed little do with any of the systems overall sound.  I heard some great systems where they used a media server with embedded dac, some using cheap dacs where the speakers and amp made all the difference....i was like, really you are using that cheap dac and getting that sound.... (btw i really love the sound of focal, ess, and atc speakers).

The only way to really tell is using your own gear that you are already intimately familiar with. Borrow a $50K DAC for a month and plop it inline. Assuming the existing gear already in place is of sufficient capability you may be impressed or maybe you wont but that decision cant be made until after you plop your own DAC back inline and listen again for another month to find out if anything is missing.

[fas42]

1 hour ago, beerandmusic said:

I am not sure i buy this...i have gone to a few audio shows listening to 20K to 300K systems, and the dac seemed little do with any of the systems overall sound.  I heard some great systems where they used a media server with embedded dac, some using cheap dacs where the speakers and amp made all the difference....i was like, really you are using that cheap dac and getting that sound.... (btw i really love the sound of focal, ess, and atc speakers).

 

Which is part of the real answer ... a system always has the "sound" of the worst, poorest parts of the chain - which could be anywhere. A perfect component just replicates what is being fed to it, perfectly - a system of perfect parts always sounds like the contents of the recording, imparts zero personality upon each one - two systems made up of radically different components, which are all perfect, will always sound identical ... there's no other way it can be.

[esldude]

What you are describing seems to fit pretty well with the idea modern DACs are audibly transparent and therefore equivalent or very close to being equivalent.  If you are having such difficulty hearing the difference it must necessarily be small assuming your gear otherwise is pretty decent.  People so badly want to believe.  I wonder what it would take to convince them there isn't anything going on of great significance.  Does everyone truly believe there is simply no limit to what humans can hear?  Do they not believe at least in principle it would be possible to make something perform with enough fidelity it can't sound more accurate?  It can sound different, but not more accurate. 

[beerandmusic]

20 minutes ago, fas42 said:

 

Which is part of the real answer ... a system always has the "sound" of the worst, poorest parts of the chain - which could be anywhere. A perfect component just replicates what is being fed to it, perfectly - a system of perfect parts always sounds like the contents of the recording, imparts zero personality upon each one - two systems made up of radically different components, which are all perfect, will always sound identical ... there's no other way it can be.

I was saying they had great sound with cheap dac.

[beerandmusic]

19 minutes ago, esldude said:

What you are describing seems to fit pretty well with the idea modern DACs are audibly transparent and therefore equivalent or very close to being equivalent.  If you are having such difficulty hearing the difference it must necessarily be small assuming your gear otherwise is pretty decent.  People so badly want to believe.  I wonder what it would take to convince them there isn't anything going on of great significance.  Does everyone truly believe there is simply no limit to what humans can hear?  Do they not believe at least in principle it would be possible to make something perform with enough fidelity it can't sound more accurate?  It can sound different, but not more accurate. 

yes, i can hear differences but not where i would say one is better...just very subtle, and sometimes i think i like one song better on one dac and another song better on another dac....but most songs i hear very little difference.  one example, the schiit i have may sound better for rock, but a cheap dsd dac will sound better for jazz playing dsd files. 

[Ralf11]

fairly subtle would not surprise me at all; not so sure about very subtle

 

can you tell us what system components are involved?

 

and what characteristics you are listening for?

 

and... what source material?

[barrows]

2 hours ago, beerandmusic said:

 

my guess is that one chip manufacturer or the other does rely more on measurements than a subjective sound comparison.  I also believe one chip or the other believes they are inferior and is trying to engineer better.

 

Instead of guessing, why not listen to Wushuliu's suggestion and seek out the videos of Mark Mallinson explaining how ESS engineer's their DAC's.  Of course ESS uses measurements and sound engineering principles, but they also imply listening panels to improve the performance of their DACs. 

[beerandmusic]

4 minutes ago, barrows said:

Instead of guessing, why not listen to Wushuliu's suggestion and seek out the videos of Mark Mallinson explaining how ESS engineer's their DAC's.  Of course ESS uses measurements and sound engineering principles, but they also imply listening panels to improve the performance of their DACs. 

i googled " videos of Mark Mallinson explaining how ESS engineer's their DAC's" didn't find anything...link?

it would be interesting to know which dac manufacturer (not chip manufacturer) they use for their listening panels.

 

[beerandmusic]

36 minutes ago, Ralf11 said:

fairly subtle would not surprise me at all; not so sure about very subtle

 

can you tell us what system components are involved?

 

and what characteristics you are listening for?

 

and... what source material?

i have tried many system components (as many as 100 different speakers in last few years and a dozen different amps)

various source material from rock, bass tests, jazz, piano, vocals, accoustics.

 

to me the most important characteristics is depth and soundstage, but of course listen for detail and warmth as well....

nothing even 1/10th the difference to upgrading the amp, and once i found amps and speakers that met provided depth and soundstage, dacs haven't made a difference in that area.

[barrows]

1 hour ago, esldude said:

What you are describing seems to fit pretty well with the idea modern DACs are audibly transparent and therefore equivalent or very close to being equivalent.  If you are having such difficulty hearing the difference it must necessarily be small assuming your gear otherwise is pretty decent.  People so badly want to believe.  I wonder what it would take to convince them there isn't anything going on of great significance.  Does everyone truly believe there is simply no limit to what humans can hear?  Do they not believe at least in principle it would be possible to make something perform with enough fidelity it can't sound more accurate?  It can sound different, but not more accurate. 

As long as I can hear a vast difference between live music and even the very best audiophile systems, I would suggest that the human ear/brain system is quite capable of discerning differences.

[barrows]

https://www.youtube.com/watch?v=JYjHKv2_OqQ

 

https://www.youtube.com/watch?v=8Mn5PrnZV-k

 

There are a couple of links for you beer.

[beerandmusic]

3 minutes ago, barrows said:

As long as I can hear a vast difference between live music and even the very best audiophile systems, I would suggest that the human ear/brain system is quite capable of discerning differences.

you can have the perfect seat with a system...with LIVE, you can see the instruments and artists performing (which will affect the brain) and the ambiance will be different...no one would ever be able to compare the 2, so it's kind of a mute point, and don't really see the correlation.

[barrows]

4 minutes ago, beerandmusic said:

you can have the perfect seat with a system...with LIVE, you can see the instruments and artists performing (which will affect the brain) and the ambiance will be different...no one would ever be able to compare the 2, so it's kind of a mute point, and don't really see the correlation.

The point is that elsdude seemed to be suggesting that audio systems are already so good that the human ear/brain system cannot discern any differences.  This is absurd as the delta of performance between even the best systems and live music is still quite large and easily to discern.

[pkane2001]

27 minutes ago, barrows said:

The point is that elsdude seemed to be suggesting that audio systems are already so good that the human ear/brain system cannot discern any differences.  This is absurd as the delta of performance between even the best systems and live music is still quite large and easily to discern.

 

True, but why do you think that the fault lies with the reproduction equipment and not with the recording? Or perhaps stereo is just not the right format? Certainly DAC doesn't seem to be the main culprit in destroying the sense of 'live'.

[opus101]

46 minutes ago, beerandmusic said:

to me the most important characteristics is depth and soundstage, but of course listen for detail and warmth as well....

nothing even 1/10th the difference to upgrading the amp, and once i found amps and speakers that met provided depth and soundstage, dacs haven't made a difference in that area.

 

I would agree with you, based on my own experience of engineering DACs. Its fairly pointless to sweat much over DACs (other than going for multibit rather than opamp-based S-D types) without establishing the amp you have is relatively transparent. I'd also caution against using price as a metric for quality.

 

BTW the videos in question are by Martin Mallinson, not Mark (his brother).