What makes a good DAC?

[Halcyon]

So I just helped a friend of mine move house, he handed me what was a $6000 CD player. Nothing more, just played standard audio discs. Granted it did have balanced audio out along with the standard RCA analog outputs.

It got me thinking... WHY!?! His response was "Oh it has an awesome DAC in it, you'd be amazed". What makes this DAC any better than whats in a regular CD player or a PC sound card? Why would that CD player be any better than playing a disc through my computer or playing a wave file through decent speakers and an amp?

Just another audiophool or is there something in this?

[NiHaoMike]

You'll be amazed how expensive pro audio gear gets. Some of that is very good build quality, some of that is premium parts, but it's mostly (relatively) low demand and the pros willing to pay high prices that keeps prices high.

That said, I'd be surprised if the analog signal path component cost even add up to $50. I designed a headphone DAC/amp with an analog signal path that is under $20 if produced in volume and you'd be hard pressed to do very much better.

[retrolefty]

So I just helped a friend of mine move house, he handed me what was a $6000 CD player. Nothing more, just played standard audio discs. Granted it did have balanced audio out along with the standard RCA analog outputs.

It got me thinking... WHY!?! His response was "Oh it has an awesome DAC in it, you'd be amazed". What makes this DAC any better than whats in a regular CD player or a PC sound card? Why would that CD player be any better than playing a disc through my computer or playing a wave file through decent speakers and an amp?

 Music listening is a very subjective and emotional experiance. Typical speakers and room acustics will have the most influence on audio 'quality'. Asking others their opinion will again only get you other subjective and emotional experience.

 So does a $6K CD player sound better then any other CD player? Only a well designed double blind test will tell the facts, but I suspect not.

 There is a whole high-end golden-ear audio equipment industry selling equipment for ridicules prices too people who have the disposable income and perceive that electronic music reproduction is like wine testing, where the opinion of others 'experts' count more then measurements beyond human perception. Tell the golden-ears that and they will just say you aren't measuring the right thing(s). Ask them to participate in a transparent double blind test and they will have some other excuse to invalidate such testing.

 

[DrGeoff]

Depends on when it was made. Around 1985, CD players were about $600-$800. The technology was different to what is used today. 16 bits resolution was a big deal at 44.1kHz stereo along with anti-aliasing filters. Delta-sigma converters started being used a bit later I think, with oversampling used to allow for less brutal filtering for aliasing components. The CD player you mention may have used very expensive components (at the time), along with better power supplies and balnced outputs. The market would also have been very small for such a player, so to recoup the development and tooling costs the prices tended to be high. This has beed the case with a lot of pro level gear in the past. Nowdays the components are better performing and much cheaper, so prices are a lot less for similar quality.

[Rasz]

$6000 CD player.
What makes this DAC any better

-placebo effect
-confirmation bias
-post purchase rationalization
and few others, Im sure it sounded AMAZING to him after he forked over $6000!

[Richard Crowley]

I hope you got the $2000 power cord for it! It will sound horrible unless you use the $2000 power cord. Oh, and the special audio cables. Only $3000 for a matched pair!  And be sure to observe the arrows on the magic cable. You don't want to install it backwards or the audio won't flow properly.

Why do you think they call them audiophools?

[BradC]

For $6k with balanced outputs it sounds like a professional grade unit that was originally designed for studio use. Not built down to a price, and not all about the DAC.

You want a unit that is guaranteed to go the distance in a 24/7 professional environment for years, you need to pay for it.

[Richard Crowley]

Even the kind of CD players made for heavy-duty broadcasting applications didn't carry price tags like that.
$6K puts it squarely into audiophool territory.

[helius]

Prices used to be higher when the technology was newer. But broadcasting CD players are still expensive.
http://www.bhphotovideo.com/bnh/controller/home?O=&sku=885003
($4K)

[PChi]

I along with a friend did compared some CD players many years ago. If the test is carried out carefully by matching the levels between the CD players to a fraction of a dB I couldn't hear any differences. My friend could hear a slight difference which turned out to be the filter that Mission had added to the output of a Philips designed CD player attenuating the bass slightly. This was testing with large floor standing loudspeakers with good bass extension. The difference probably wouldn't have been audible with typical 2 way stand mounted loudspeakers.
However loudspeakers are all different and possibly spening $6000 might be worth it.

[German_EE]

The balanced outputs indicate that whoever designed and built this unit may have taken some care. Six thousand dollars? probably a little too high but I would be interested to see inside.

So, what makes a good DAC?

1) Low noise

2) Linear Output, double the number going in and you should see double the voltage.

3) Close matching between the left and right outputs

4) No switching spikes, some DAC chips have a voltage spike in the middle of their range

5) Perfectly flat frequency response

Now all you need to do is make sure all the circuitry surrounding the DAC is built to a similar specification, including the power supply.

[Halcyon]

Depends on when it was made.

It's a modern unit, within the last 5 years or so.

For $6k with balanced outputs it sounds like a professional grade unit that was originally designed for studio use. Not built down to a price, and not all about the DAC.

You want a unit that is guaranteed to go the distance in a 24/7 professional environment for years, you need to pay for it.

Definitely a "consumer" unit. The only thing professional about it are the XLR connectors on the back. It's not rack mountable and is clearly designed to be shown off in ones entertainment unit. Only very basic features too in terms of functionality.

I've seen actual professional units cheaper than this that do a whole lot more.

[GoldSrc]

Audiophoolery at its best.
But you know, the placebo effect is a powerful thing, and you will probably are never going to make him accept that.
I have an old CD player from 1988-89 and it sounds good, and if you take into account that DACs are better than what they were back then, you don't need to spend much to have good sound.

[kosmonooit]

"I think therefore I am"

René Descartes, 17th Century

[Rasz]

"I think therefore I am"

how would that fit here? full quote is dubito, ergo cogito, ergo sum, Show me an audiophool that questions ze truth of the Lamp?

[Mechanical Menace]

What makes this DAC any better than whats in a regular CD player or a PC sound card? Why would that CD player be any better than playing a disc through my computer or playing a wave file through decent speakers and an amp?

If said PC has a decent PSU and you're not using the cheapest leads between the PC and the amp almost certainly nothing, and in any kind of objective measure probably worse. My main test of a sound source is how well it recreates the recorded sound, not how subjectively 'better' it sounds, and the well overpriced stuff tends to do terribly at this. To make it sound 'better' they're almost always too bassy compared to the real world source. Now if your personal preference is lots of bass I'm not saying that's a bad thing. But that's what tone control and/or equalisers are for*.


*Well maybe not what equalisers are meant to be for but what they tend to get used for lol.

[BradC]

If said PC has a decent PSU and you're not using the cheapest leads between the PC and the amp almost certainly nothing, and in any kind of objective measure probably worse. My main test of a sound source is how well it recreates the recorded sound, not how subjectively 'better' it sounds, and the well overpriced stuff tends to do terribly at this. To make it sound 'better' they're almost always too bassy compared to the real world source. Now if your personal preference is lots of bass I'm not saying that's a bad thing. But that's what tone control and/or equalisers are for*.

Except most PC sound has a very high noise floor, despite however many million dB dynamic range is stamped on the box. My PC hooks into my receiver/amp with a toslink cable. No additional noise

I'm with you on sound subjectivity. I like my EQ flat. My wife hates the sound in my car, which has been tuned to be as flat as is practical across the board. She calls it "lifeless and tinny". If I wind the treble down to nothing and wind up the bass and sub, she thinks it sounds great.

A bit like my CRT TV's. I loved Blaupunkt because it has the best accurate colour rendition, but all my friends told me it looked dull and boring. Coming from an exposure to broadcast equipment from a very young age, I grew to appreciate accuracy over sensationalism.

[coppice]

Except most PC sound has a very high noise floor, despite however many million dB dynamic range is stamped on the box. My PC hooks into my receiver/amp with a toslink cable. No additional noise

Most PC sound used to have a horrible noise floor, but that hasn't been true for several years. Its unusual to find a bad example these days.

[DrGeoff]

I'm with you on sound subjectivity. I like my EQ flat. My wife hates the sound in my car, which has been tuned to be as flat as is practical across the board. She calls it "lifeless and tinny". If I wind the treble down to nothing and wind up the bass and sub, she thinks it sounds great.

Except, of course, it isn't "flat". The space into which the drivers project contributes significantly to the response of the drivers. You will find horrible LF peaks and troughs (maybe your wife is sitting where the LF null is) and HF flutter echoes in most listening environments that have not been appropriately treated. Perform a response plot for the room at the listening position and you will see what I mean. The dips and troughs of an untreated space can easily be +/-20dB. Only acoustic treatment can fix this, eq is useless for fixing room acoustic problems.

It always amazed me when audiophools would spend huge amounts of money on ridiculously specified audio components and ignore the single most significant factor in the audio reporoduction environment.

[BradC]

I'm with you on sound subjectivity. I like my EQ flat. My wife hates the sound in my car, which has been tuned to be as flat as is practical across the board. She calls it "lifeless and tinny". If I wind the treble down to nothing and wind up the bass and sub, she thinks it sounds great.

Except, of course, it isn't "flat". The space into which the drivers project contributes significantly to the response of the drivers. You will find horrible LF peaks and troughs (maybe your wife is sitting where the LF null is) and HF flutter echoes in most listening environments that have not been appropriately treated. Perform a response plot for the room at the listening position and you will see what I mean. The dips and troughs of an untreated space can easily be +/-20dB. Only acoustic treatment can fix this, eq is useless for fixing room acoustic problems.

It always amazed me when audiophools would spend huge amounts of money on ridiculously specified audio components and ignore the single most significant factor in the audio reporoduction environment.

Sorry, when I said flat I meant by actually using an Audessy EQ DSP that compensates for quite a bit of that using multi-point EQ and time correction. I have 2 curves. One is just for me and compensates pretty closely for my head position and the second is a compromise between my head position and the passengers head position. It's actually pretty good, and far better than just a multi-point EQ. I spent a great deal of time with a reference mic setting it up to make the best job of it both I and the software would allow.

Coupled with material changes in the vehicle, and extra treatment behind the door skins, it's not perfect, but it's a crap load better than an average vehicle sound environment.

[DrGeoff]

Sorry, when I said flat I meant by actually using an Audessy EQ DSP that compensates for quite a bit of that using multi-point EQ and time correction. I have 2 curves. One is just for me and compensates pretty closely for my head position and the second is a compromise between my head position and the passengers head position. It's actually pretty good, and far better than just a multi-point EQ. I spent a great deal of time with a reference mic setting it up to make the best job of it both I and the software would allow.

Coupled with material changes in the vehicle, and extra treatment behind the door skins, it's not perfect, but it's a crap load better than an average vehicle sound environment.

You sound like you have gone to a lot of trouble. Stock vehicle environments have always been too boomy for me, but make a great check for translation of material after mastering. You can certainly tell when the HPF is not enabled!

The problem with EQ compensation is that it only requires head movement of a few centimeters to be out of the sweet spot for which the compensation was performed. Probably the best you can hope for in a car, but definitely not much use for room compensation.

[BradC]

You sound like you have gone to a lot of trouble. Stock vehicle environments have always been too boomy for me, but make a great check for translation of material after mastering. You can certainly tell when the HPF is not enabled!

I spent about $600 on the whole stereo and considerably more than that on the vehicle interior. Cost/benefit says interior mods give better returns, and I'd rather spend a couple of hundred $$ on new head lining material than a bigger sub or extra amp. Head position is important, but it turns out I don't tend to move my head that much. Between the sculpted head-rest and the mirror position I'd be lucky to vary more than 50mm in any direction and then not often.

The problem with EQ compensation is that it only requires head movement of a few centimeters to be out of the sweet spot for which the compensation was performed. Probably the best you can hope for in a car, but definitely not much use for room compensation.

Yes, I have a mate with the Audessy system in his home theater receiver, and it sounds great in his seat but not so crash hot elsewhere in the room.

Back to the topic at hand, I've listened to some pretty impressive systems over the years and aside from laptops and motherboards I've been hard pressed to tell a quality DAC from a quality DAC (where quality implies the DAC itself cost more than about $4 qty/1). I've heard some considerable difference between CD players, but none of it is related to the digital signal path and all of it is related to cheap (and I mean *cheap*) consumer gear with the cheapest of cheap analog paths.

Unfortunately, with the advent of compressed audio (in both senses of the word) any pretence in accuracy has gone out the window and most people are no longer interested in what the recording engineer intended it to sound like, because the mastering engineer has ruined the dynamics anyway.

[Mechanical Menace]

Unfortunately, with the advent of compressed audio (in both senses of the word) any pretence in accuracy has gone out the window and most people are no longer interested in what the recording engineer intended it to sound like, because the mastering engineer has ruined the dynamics anyway.

Crafted albums dying and going back to the bad old days of them just being a way to resell singles probably has a lot more to do that than compressed audio, especially with how much better (sizewise) lossless codecs are getting and how popular they're becoming. And at least DRM on them where relevant doesn't mangle the audio like it did with CDs.

And when it comes to the mixing recording and mastering engineers, or any sort of sound tech when live, have never and will never listen to what the performers want if they can avoid it.

[Bassman59]

And when it comes to the mixing recording and mastering engineers, or any sort of sound tech when live, have never and will never listen to what the performers want if they can avoid it.

Speaking as a professional live sound engineer, I can tell you that we ALWAYS listen to what the performers want. Because if we didn't, then we wouldn't be on the gig. If the band wants it loud (My Bloody Valentine, anyone?) that's what they get.

Now, also speaking in that capacity, all too often the mix person is at the mercy of the system du jour. I can approve or disapprove of a system in advance, but until I get to the venue, I have no idea whether it is installed properly. There is only so much one can do with equalization. If the boxes aren't aimed at the audience or set up correctly, it's game over before it even begins. All one can do at that point is to make it sound "as good as possible," which is as frustrating for the mix-person as it is for someone in the audience in a bad location. (Did I mention that sometimes mix position is in the worst spot in the house? Under a balcony! Please, just kill me now.)

And there are some rooms that should not be hosting shows at all (Wilbur Theatre in Boston, Aragon Ballroom in Chicago, Terminal 5 in NYC, Wellmont Theatre in Montclair, NJ, Electric Factory in Philly, etc etc). Big empty barns with no acoustic control, absolute nightmares.

Note I haven't even mentioned what happens when the band's stage volume is out of control.

And my friends who mix and master recordings professionally? They all, without exception, hate the loudness wars, but they do what is asked of them by the client.  Because there's always someone willing to do the work.

[kosmonooit]

"I think therefore I am"

how would that fit here? full quote is dubito, ergo cogito, ergo sum ...

Someone mentioned the placebo effect ...