DACs

[maximoiglesias]

Has anybody out there ever used a DAC (Digital to Analog Converter)??
Is it worth it? Can you really tell the difference when playing a CD??

 

[JerryLove]

I asssume you are referring to external DAC stereo components. Anything that converts digital-to-analog is a DAC, so there's one in the CD-player, one in any modern receiver, one in the iPod, etc.

I've not used a dedicated DAC at home. I'm not convinced that there's audiable problems in the built-in DACs in good equipment.

 

[Josuah]

Dedicated DACs get more coverage at some other audio forums. I'm using a Neko Audio D100 in two of my setups. There are lots of different DACs out there, and they do sound different.

Some differences, which can impact the sound, include upsampling, oversampling versus non-oversampling, design and use of analog filters, use of feedback.

Other popular ones I can think of off the top of my head include:

Benchmark
Lavry
Apogee
Cambridge Audio
Parasound
MHDT
Stello
PS Audio
Wadia

 

[oppman99]

I use an external DAC connected to my computer to send music to my preamp. I've also tried using it with an older Sony DVD player and didn't notice much of a difference. I've only had a chance to use the Cambridge DACmagic and PS Audio DIII. Which CD player are you planning on using with one?

 

[Seth=L]

There are lots of different DACs out there, and they do sound different.

Oh really?

I just love when it's stated like it's a fact. I suppose there's a test that proves it, since it's a fact.

 

[JerryLove]

Actually: I don't care if they sound "different" or not. I'd want to see how they sound "better".

 

[maximoiglesias]

Thanks to all who have answered.
I read a lot of different Audio and HiFi magazines and like Someone said dedicated DACs are becoming very popular, either to connect a PC to your reciever or to upscale CD players to 32 bits/192 kHZ which is very close to analog (vinyl) sound.

I wanted to connect my old and reliable Sony disc player to my vintage NEC integrated amp.

If anybody else has had an experience with DACs please let me know.

 

[Seth=L]

Thanks to all who have answered.
I read a lot of different Audio and HiFi magazines and like Someone said dedicated DACs are becoming very popular, either to connect a PC to your reciever or to upscale CD players to 32 bits/192 kHZ which is very close to analog (vinyl) sound.

I wanted to connect my old and reliable Sony disc player to my vintage NEC integrated amp.

If anybody else has had an experience with DACs please let me know.

I've heard and read that Vinyl is supposed to be a superior format for playing back music and that CDs can not produce as good of sound as an LP. This is untrue. Vinyl today almost always sounds better than the same material on CD, and it's only because the record company ruins the CD. Well mastered CDs are harder to come by these days because of this iPod generation.

Having owned and used many CD players I can honestly say the differences are so minute it's not worth spending huge sums of money on unless it's just a pride in ownership thing.

 

[Josuah]

Oh really?

I just love when it's stated like it's a fact. I suppose there's a test that proves it, since it's a fact.

I often state things that are facts without going into deep explanations or providing a proof. Most people I know do that. But I believe you would like to conduct a test to prove it to yourself.

If you only want to talk about measurements, then yes, they will measure differently. You could take a voltage vs. time measurement of some complicated music into a high load (e.g. 1 Mohm), normalize for the difference in voltage levels, see if they match exactly or not.

That's somewhat complicated to implement or eyeball, so you could instead use an oscilloscope and hook one up to channel A through a 1:1 transformer, the other to channel B through the same transformer but reverse the secondaries, set the trace option to difference, and see if the line deviates from zero.

If you don't want to do that, buy a couple of DAC chips and implement the reference circuits listed in the white papers and see how they end up different for music. The reference circuits might employ different analog output stages or filters though. So try swapping one chip into the other analog output circuit and see how they differ (if that swap is possible).

If you really want to go farther, pick up some of the older NOS chips which will inherently result in a different result when passed through the reconstruction filter (do the math).

I've heard and read that Vinyl is supposed to be a superior format for playing back music and that CDs can not produce as good of sound as an LP. This is untrue. Vinyl today almost always sounds better than the same material on CD, and it's only because the record company ruins the CD.

This is also in part because of how the most common DAC implementations are done (in CDPs, receivers, etc.). As you've implied, lots of people refer to this as the digital sound. But you can find DACs that don't do this. Amps and pre-amps can inject the same problem. You won't see this manifest on sine/sweep measurements.

A good place to start is zero-feedback solid-state designs.

 

[Knucklehead90]

Much of that 'digital sound' is from mastering CDs that are 'hot' - dynamics are compressed and sound like crap. Most $10 DACs will play a well mastered CD just like the $1000 DAC.

In double blind tests a $100 CD player could not be singled out from a $1000 CD player on a given setup.

That should be enough to tell you something about a lot of the high priced gear in this hobby. There is a diminishing return with any hobby and the audio hobby is no exception.

 

[Josuah]

Much of that 'digital sound' is from mastering CDs that are 'hot' - dynamics are compressed and sound like crap.

The digital sound I'm referring to is not due to compression or sound engineer preferences. I've certainly heard vinyl that was mastered differently than the CD, but there is also coloration that can be added by the DAC implementation and I definitely disagree with a majority generalization that a $10 DAC sounds the same as a $1000 DAC.

In double blind tests a $100 CD player could not be singled out from a $1000 CD player on a given setup.

That should be enough to tell you something about a lot of the high priced gear in this hobby. There is a diminishing return with any hobby and the audio hobby is no exception.

Double-blind tests are useful to an extent. Any debate that boils down to a DBT on one side isn't going to get very far. And price is a pretty arbitrary choice by which to differentiate DACs. From a scientific process point of view, it would be much better to differentiate between two units based on their components, circuit, and build.

I've also conducted DBTs with inexpensive and expensive DACs with a fair number of people (audiophiles and not) and differences were audible. Whose DBT wins now? just kidding

Regardless, I am presenting my opinion on audible differences and also scientific methods by which audible differences can be matched with hard measurements. I know just because I say something and provide some random evidence the vast majority of people on Audioholics won't believe me. So I invite people to try and test it out themselves.

(If I said something that matched the accepted wisdom of the Audioholics crowd and didn't provide evidence, there wouldn't be any debate.)

 

[JerryLove]

The digital sound I'm referring to is not due to compression or sound engineer preferences. I've certainly heard vinyl that was mastered differently than the CD, but there is also coloration that can be added by the DAC implementation and I definitely disagree with a majority generalization that a $10 DAC sounds the same as a $1000 DAC.

A testable theory.

Double-blind tests are useful to an extent. Any debate that boils down to a DBT on one side isn't going to get very far. And price is a pretty arbitrary choice by which to differentiate DACs. From a scientific process point of view, it would be much better to differentiate between two units based on their components, circuit, and build.

[...]

Regardless, I am presenting my opinion on audible differences and also scientific methods by which audible differences can be matched with hard measurements. I know just because I say something and provide some random evidence the vast majority of people on Audioholics won't believe me. So I invite people to try and test it out themselves.

Scientific method is pretty simple. You create a model of something capabale of making predictions that can be tested, and the predictions always test true. If predictions test false, it's a bad model.

So your hypothesis is that different DACs result in different sound. The opposing side is that different DACs (malfunctions and deliberate distortions not withstanding) do not sound different.

The DBT is clearly the indicated test for this prediction: though meter-testing of the electrical waveform would also establish differences or lack thereof (but not conclusively the quality of those differences: can they be perceived).

A study of the components is useless for testing the model you propose. It's only usefult to make more predictions. For experiences of perception *nothing* trumps DBT.

So far, I've never noticed a difference; but I've not tested it either. Simply showing me the flaws in the analog wave off, say, my SoundBlaster would be a good start.

(If I said something that matched the accepted wisdom of the Audioholics crowd and didn't provide evidence, there wouldn't be any debate.)

This is true: if you had said something I already believed to be true before you said it, I would not question it's validity just because you said it.

Watch me do it.
1) The sky looks blue-ish.
2) The sky is full of frogs.

Notice you would likely demand proof of one and not the other... because one was proven to you long ago.

 

[Josuah]

So your hypothesis is that different DACs result in different sound. The opposing side is that different DACs (malfunctions and deliberate distortions not withstanding) do not sound different.

The DBT is clearly the indicated test for this prediction: though meter-testing of the electrical waveform would also establish differences or lack thereof (but not conclusively the quality of those differences: can they be perceived).

A study of the components is useless for testing the model you propose. It's only usefult to make more predictions. For experiences of perception *nothing* trumps DBT.

So far, I've never noticed a difference; but I've not tested it either. Simply showing me the flaws in the analog wave off, say, my SoundBlaster would be a good start.

There are also people who can differentiate between more frequencies in the visual spectrum than others. Yet if you were to perform a DBT among a sample of the general population, you would conclude that there is no way to differentiate between those frequencies.

I referenced a research paper in a different thread that examined specifically the audible perception of distortion between trained musicians and non-musicians which found a significant statistical correlation between the ability to detect that distortion and whether or not the person was a trained musician.

The reason I take issue with the claim of DBTs being the final answer is that ultimately each individual should be purchasing the right equipment for maximizing their personal enjoyment. Different mountain bike designs are unlikely to change my biking experience but can make an enormous difference to a serious biker.

I have tested both electrically and aurally. And my results (including the ones where I have conducted DBTs) provide support for my "hypothesis". Why aren't my DBTs as valid? Maybe the other guys used headphones where I used speakers. Or they used a typical living room where I used a heavily treated room. Maybe they ran everything through a receiver that performed ADC and DAC washing out the differences of the CD players whereas I didn't.

Since I am making claims with respect to individuals instead of groups, I suppose it makes sense to differentiate myself. I have a B.S. in Electrical Engineering so I have the background to understand how different DACs can reproduce music differently. I learned to play the piano, violin, and trumpet. My parents played violin and piano. One brother plays piano and violin. The other plays piano, guitar, saxophone, drums, and has done sound mixing. So I'm not tone deaf and very familiar with musical instruments and composition. I happen to know my hearing sensitivity is greater than those of most other people, even for someone of much younger age.

I recognize the skepticism. That's why I am trying to provide methods by which others can validate my claims. But I also propose that you should be skeptical of "existing claims" such as "DBTs prove no audible difference between DACs" and be open to the idea that those claims are wrong. That's also part of the scientific process.

This is true: if you had said something I already believed to be true before you said it, I would not question it's validity just because you said it.

Watch me do it.
1) The sky looks blue-ish.
2) The sky is full of frogs.

Notice you would likely demand proof of one and not the other... because one was proven to you long ago.

Another statement commonly believed on this forum is that there aren't differences between amplifiers with respect to sound quality, and so there is no reason to switch to something unless it provides significantly more power to avoid clipping and provide a dB gain. At least I believe that to be a commonly accepted fact on this forum from what I've read.

Yet I also know this to be false. ::shrug::

FWIW, the Wyred 4 Sound ST-500 outperforms the Parasound Halo A23 with respect to THD and smearing for music with at most a Vrms of 1V which is well within the operating range of both amplifiers.

Likewise, the Marantz AV8003 DAC is able to reproduce music with less smearing than the Emotiva DMC-1 DAC (the AV8003 also has a better design as its noise before gain is at least 20dB better than the DMC-1).

 

[mtrycrafts]

Thanks to all who have answered.
I read a lot of different Audio and HiFi magazines and like Someone said dedicated DACs are becoming very popular,

Yep, there are a lot of 'popular' fad diets out there too but that doesn't mean they work, right
Same applies for dacs, just a fad, unnecessary fad.

... to upscale CD players to 32 bits/192 kHZ which is very close to analog (vinyl) sound.

There is no evidence for this claim. Yet more of the 'popular' anecdotes, urban legends, etc.
Vinyl is a preference, not a technically better system, period, end of the story.

 

[mtrycrafts]

There are also people who can differentiate between more frequencies in the visual spectrum than others. Yet if you were to perform a DBT among a sample of the general population, you would conclude that there is no way to differentiate between those frequencies.

I bet you would be wrong in that hypothesis of yours, and would see that some/many would differentiate it. Just a matter of numbers who can and cannot, not that no one can.

I referenced a research paper in a different thread that examined specifically the audible perception of distortion between trained musicians and non-musicians which found a significant statistical correlation between the ability to detect that distortion and whether or not the person was a trained musician.

Would you re-post that paper here, please.

The reason I take issue with the claim of DBTs being the final answer is that ultimately each individual should be purchasing the right equipment for maximizing their personal enjoyment.

Well, DBTs are the arbiter of the facts. Has no bearing what some might prefer or not different from the facts. One only has to look at any consumer area for this to be the case. Many go to psychics for answers

I have tested both electrically and aurally. And my results (including the ones where I have conducted DBTs) provide support for my "hypothesis".

Perhaps your testing was flawed? How about other credible supporting evidence?

Why aren't my DBTs as valid?

Replication by others?

Since I am making claims with respect to individuals instead of groups, I suppose it makes sense to differentiate myself. I have a B.S. in Electrical Engineering so I have the background to understand how different DACs can reproduce music differently. I learned to play the piano, violin, and trumpet. My parents played violin and piano. One brother plays piano and violin. The other plays piano, guitar, saxophone, drums, and has done sound mixing. So I'm not tone deaf and very familiar with musical instruments and composition. I happen to know my hearing sensitivity is greater than those of most other people, even for someone of much younger age.

Wonderful. Yet, other studies show that even highly trained conductors cannot differentiate their own discarded works nor perform better than musicians of ordinary folks. Harman has shown that highly trained people get the right answers quicker and more reliable but untrained folks can get the answers as well.

I recognize the skepticism. That's why I am trying to provide methods by which others can validate my claims. But I also propose that you should be skeptical of "existing claims" such as "DBTs prove no audible difference between DACs" and be open to the idea that those claims are wrong. That's also part of the scientific process.

Why would they be wrong? Because they are different from your results? Perhaps your methods are flawed? Unless you have trained people see, follow your protocol, no one will know for sure.

Another statement commonly believed on this forum is that there aren't differences between amplifiers with respect to sound quality, and so there is no reason to switch to something unless it provides significantly more power to avoid clipping and provide a dB gain. At least I believe that to be a commonly accepted fact on this forum from what I've read.

Yet I also know this to be false. ::shrug::

In that case, the probability of your protocol being flawed is even higher. Perhaps you should consider that to be more likely than you being right.

 

[ParadigmDawg]

How about using an outboard DAC when you are using a computer as a source?

Yep, there are a lot of 'popular' fad diets out there too but that doesn't mean they work, right
Same applies for dacs, just a fad, unnecessary fad.

.

 

[JerryLove]

There are also people who can differentiate between more frequencies in the visual spectrum than others. Yet if you were to perform a DBT among a sample of the general population, you would conclude that there is no way to differentiate between those frequencies.

Ahh yes, the magic "golden ear" people.

The first problem is that these differences would show up on a scope.

The second is that general population tests is how we found people with extended hearing ranges in the first place. You don't seem to understand statistics.

The third problem is that one of these golden-ear people long ago should have been proving his golden ear in DBTs. Many have claimed to be able to and been tested. None have been able to demonstrate this ability.

I referenced a research paper in a different thread that examined specifically the audible perception of distortion between trained musicians and non-musicians which found a significant statistical correlation between the ability to detect that distortion and whether or not the person was a trained musician.

No trained musician has been able to identify level-matched equipment in a DBT either. This is a red-herring.

The reason I take issue with the claim of DBTs being the final answer is that ultimately each individual should be purchasing the right equipment for maximizing their personal enjoyment. Different mountain bike designs are unlikely to change my biking experience but can make an enormous difference to a serious biker.

Irellevent. If you cannot tell in a DBT, then there is no audiable difference to you that is not the power of suggestion.

I have tested both electrically and aurally. And my results (including the ones where I have conducted DBTs) provide support for my "hypothesis". Why aren't my DBTs as valid? Maybe the other guys used headphones where I used speakers. Or they used a typical living room where I used a heavily treated room. Maybe they ran everything through a receiver that performed ADC and DAC washing out the differences of the CD players whereas I didn't.

No, the DBTs I've looked at were done with high end gear in treated envyronments (though I've certainly seen others). Assuming you are being honest, I would say your DBT was tainted.

Since I am making claims with respect to individuals instead of groups, I suppose it makes sense to differentiate myself.

Since you have classed large groups (those with HF hearing and trained musicians) in your group: It's irellevent.

You might as well claim to have the power to move things with your mind.

I recognize the skepticism. That's why I am trying to provide methods by which others can validate my claims. But I also propose that you should be skeptical of "existing claims" such as "DBTs prove no audible difference between DACs" and be open to the idea that those claims are wrong. That's also part of the scientific process.

You are too far away for me to test your claims. Perhaps another audioholic in CA would like to take you up on that.

Another statement commonly believed on this forum is that there aren't differences between amplifiers with respect to sound quality, and so there is no reason to switch to something unless it provides significantly more power to avoid clipping and provide a dB gain. At least I believe that to be a commonly accepted fact on this forum from what I've read.

Yet I also know this to be false. ::shrug::

My neighbor knows that UFOs visit this planet. At least I can't disprove his claim with a test.

Likewise, the Marantz AV8003 DAC is able to reproduce music with less smearing than the Emotiva DMC-1 DAC (the AV8003 also has a better design as its noise before gain is at least 20dB better than the DMC-1).

How many db of "smear" does the Emotive have when playing 110db sound?

If I had the gear to run the DBT here, I'd bet you the plane ticket. Sadly, I cannot validate that your favorite DAC isn't coloring the noise, which would indeed make it discernable... much like how I can identify an amp with "loudness" on vs one with it off when level matched.

 

[adwilk]

How about using an outboard DAC when you are using a computer as a source?

I know what device you are looking at, and I do think there is a difference between the external device and allowing the computer to do the processing. ESPECIALLY in a windows based machine. Its silly the amount of influence something like WMP can add when playing some files through something like a headphone jack on a laptop. Consider that outboard DAC comparable to an upgraded, external soundcard. That said, while there are some differences and I know of them, I probably couldn't discern one sound from the other and the headphone out would work just fine for me if I didn't know any better.

I'm seriously considering ordering one of those things you were looking at. I have it bookmarked on my other laptop. I'm curious what you think of it.

 

[PENG]

There are also people who can differentiate between more frequencies in the visual spectrum than others. Yet if you were to perform a DBT among a sample of the general population, you would conclude that there is no way to differentiate between those frequencies.

Sure, but is that still the case for frequencies outside of the 20 to 20,000 Hz spectrum?

I referenced a research paper in a different thread that examined specifically the audible perception of distortion between trained musicians and non-musicians which found a significant statistical correlation between the ability to detect that distortion and whether or not the person was a trained musician.

That means nothing without specifying the magnitude and types of distortions. Please provide more details if you don't mind.

The reason I take issue with the claim of DBTs being the final answer is that ultimately each individual should be purchasing the right equipment for maximizing their personal enjoyment. Different mountain bike designs are unlikely to change my biking experience but can make an enormous difference to a serious biker.

I happen to agree with you on this, as long as you we are talking about specs that are relevant. For example, let's consider THD, some people may have trouble with 0.1% but we are talking about 0.01% vs 0.02% then THD would be irrelevant.

Since I am making claims with respect to individuals instead of groups, I suppose it makes sense to differentiate myself. I have a B.S. in Electrical Engineering so I have the background to understand how different DACs can reproduce music differently.

Not necessarily ture, a B.S. in EE does not even guarantee (just more likely, but no guarantee) an understanding of Fourier analysis or even more basic electrical/mathematical theories. To differential yourself you might want to tell us more than just a B.S. degree.

I recognize the skepticism. That's why I am trying to provide methods by which others can validate my claims. But I also propose that you should be skeptical of "existing claims" such as "DBTs prove no audible difference between DACs" and be open to the idea that those claims are wrong. That's also part of the scientific process.

Talk about skepticism, I wonder if you believe in Monster cables too. I am not being sarcastic, just curious.
Edit: I do like the build/look of the top model Monster cables. I also do like the look and build of a few expensive high end separate components including amps and DAC's and I may even collect (hence support) them if I were rich. I have nothing against any of those nice components, but I wouldn't tell people to go for any of them only for the claimed 'better' SQ that they may or may not like, or even perceive.

Another statement commonly believed on this forum is that there aren't differences between amplifiers with respect to sound quality, and so there is no reason to switch to something unless it provides significantly more power to avoid clipping and provide a dB gain. At least I believe that to be a commonly accepted fact on this forum from what I've read.

Yet I also know this to be false. ::shrug::

This is a little over the top, 'yet I also know this to be false', how do you know? I guess we are supposed to take your words for it. Sorry, now I am being a little sarcastic.

Likewise, the Marantz AV8003 DAC is able to reproduce music with less smearing than the Emotiva DMC-1 DAC (the AV8003 also has a better design as its noise before gain is at least 20dB better than the DMC-1).

I am supposing to find this amazing, coming from someone with a BS degree in.............but as I said before, there is no guarantee so I know I should not be surprised.

 

[Josuah]

I'm hoping to avoid this degenerating....

I said visual spectrum, not audible. Because I was referring to Tetrachromacy.

I am claiming differences that show up on a scope. I am providing methods by which a person can measure those differences in their own home, rather than come to my home. Sometimes the audio quality is so large it's blatantly obvious. Sometimes it's harder to detect, but I've identified differences and then tried to prove them to myself by looking for (and finding) different traces on a scope.

I am stating that statistical generalizations due to DBTs of large groups does not necessarily apply to individual DBTs, and provide tetrachromacy as an example of why this may be so.

When I refer to smear, I am referring to the blending of one note into another in the time domain. This is separate from harmonic distortion and frequency response measurements.

I wish I could find the research paper again. I thought I downloaded it but I can't find it. Sorry. I know that automatically lessens my credibility (I might be lying) but it's just my word that I'm telling the truth the paper exists. It's just one paper, so it's not the end-all-be-all, but it's the only one I found that tried to differentiate between musicians and non-musicians. I don't particularly like straight out refusal to believe research A because of research B though.

No, I don't "believe" in Monster cables. Do I believe a cable can affect the sound? Yes. Why? Because you can easily model an analog filter in terms of R, C, L and the circuit created by your output jack, cable, and input jack can be reduced to such a circuit. What cable do I prefer? Thick, well-insulated, well-shielded copper with low electrical characteristics.

I'm not asking anyone to believe me just because I said so. I've made claims, just like everyone else. When asked to prove my claims, I've asked the skeptics to prove my claims to themselves. I'm being handed the burden of proof. I'm trying to hand it back.