Does It Really Sound The Same?

[xnor]

Quote:

The short answer to the OP, in my opinion, is you can't over generalize such things. Words like "all" or "always" are usually pretty easy to dispute as it's not hard to come up with exceptions.
 
But if the entire design has been engineered to some reasonable point of diminishing returns, and everything is literally "good enough" (which doesn't mean expensive), then it's very likely in a blind test you won't be able to tell one DAC from a similarly well engineered different DAC.

 
Spot on, and what some of us have tried to tell upstateguy as well.
 
 

 

[bigshot]

I made the mistake of buying a very good SACD player based on the comments in this forum about "night and day" differences. At first I really didn't know what the difference was because the design of the deck inserted a ten second gap when I tried to switch layers. The volume level between layers was noticeably different too. Then I started noticing differences in the mix between the CD and SACD layers... Things that the format couldn't be responsible for, like different balances on vocals and different reverbs. Finally, I went to great lengths to locate a recording that was only available as a hybrid SACD, a highly regarded recording by Pentatone. I arranged for it to be played back side by side in both an SACD player and a CD player. I balanced the line levels and determined that there was absolutely no difference between the two.

I understand why people go with subjective reactions. There are people out there who are bound and determined to muddy the waters. It's a he'll of a lot of work to chase out all the bugs in a test. It took me a month to set up a fair comparison of CD and SACD. But there are some things that I've done to improve the sound of my system that are clear and obvious improvements. I still think of adjustments i can make. Whenever I hear people talk about how their system is almost perfect and expensive cables or a fancy DAC are giving them the last .001% of improvement, I know that they are on the wrong track. I'm positive there are obvious areas of improvement that they're neglecting and just want to throw money at the problem instead of strategy and analysis.

But once you chase down one lie and prove it to yourself that it's false, it makes it easier to question all the other naked emperors.

 

[khaos974]

In the defense of SACD*, some masterings are only available on SACDs and those masterings are objectively better than what's available on CD.

*Actually, it means that it could be a good move to buy an SACD player, the format itself has next to no purpose.

 

[Uncle Erik]

nwavguy, I agree. You touched on the status symbol argument, too.

What's sad is how people get upset at how inexpensive great audio is today. I grew up with LPs, 8-tracks and cassettes. All quite flawed. But if you look at today's digital from a perspective 30 years ago, digital seems like a miracle. Not only are most of the source problems solved, digital is incredibly cheap. Great players under $50, you can get used discs for $1 or $2, and you can put thousands of albums onto a five pound laptop you can take anywhere.

Great audio has never cost less and you'd think this would be cause for celebration. But no. People are angry because their status symbol works the same as everything else.

 

[Prog Rock Man]

Quote:

nwavguy, I agree. You touched on the status symbol argument, too.

What's sad is how people get upset at how inexpensive great audio is today. I grew up with LPs, 8-tracks and cassettes. All quite flawed. But if you look at today's digital from a perspective 30 years ago, digital seems like a miracle. Not only are most of the source problems solved, digital is incredibly cheap. Great players under $50, you can get used discs for $1 or $2, and you can put thousands of albums onto a five pound laptop you can take anywhere.

Great audio has never cost less and you'd think this would be cause for celebration. But no. People are angry because their status symbol works the same as everything else.

Having worked my way through the formats since the 1970s it was the much maligned 'is really hifi' ipod that got me into the digital age. When you put an ipod up as your source, plugged into the back of an amp and state it sounds brilliant, ouch, the reaction is painful.

 

[upstateguy]

Quote:

The short answer to the OP, in my opinion, is you can't over generalize such things. Words like "all" or "always" are usually pretty easy to dispute as it's not hard to come up with exceptions.
 
But if the entire design has been engineered to some reasonable point of diminishing returns, and everything is literally "good enough" (which doesn't mean expensive), then it's very likely in a blind test you won't be able to tell one DAC from a similarly well engineered different DAC.
 
There are many examples that demonstrate this. Meyer & Moran, for example, published a landmark AES study involving over 500 listening trials over a year including members of an audiophile club, recording engineers, and students with especially acute hearing. They played high resolution (DSD) SACDs on a high end system and sometimes, unknown to the listeners, switched an extra 16 bit 44 Khz A/D and 16 bit 44 Khz D/A into the signal path. At any realistic volume setting (to not expose the higher 16 bit noise floor during silent passages) NOBODY COULD TELL THE DIFFERENCE!
 
The above study says a lot. It proves you can put a whole bunch of extra hardware in the signal path, and as I said above, if it's sufficiently well engineered, it doesn't change the sound enough for anyone to detect it's even there let alone what it "sounds" like. The A/D and D/A had nearly everything on your list but nobody could tell when all that was added to the signal path.
 
While those who like their expensive DACs, SACD players, etc. naturally tried to dispute Meyer & Moran. The critics, as is often the case, didn't provide their own study to prove Meyer & Moran wrong (even though they could have easily funded one from all the commercial SACD interests watching their entire market being debunked). They mostly just tried to take pot shots at them and throw rocks. Meyer & Moran even responded to the critics in a follow up AES paper and proved many of their concerns wrong. There are links to all this on my site in the subjective debate article.
 
So can all those things on your list sound the same, even for very different designs using different parts, yes they can if you define "sound the same" as "nobody can hear a difference in a blind test".
 
 

I readily agree with most of what you've said, but I'm still having trouble getting on board with a small part of what you said.
 
Everyone will agree that over generalizations can be defeated by finding exceptions.
 
But "sufficiently well engineered" becomes a bothersome statement as well, because it seeks to exclude exceptions.
 
Here's one of the problems I'm still having:
 
Take 2 Benchmark DACs, like yours, for instance.  Would it be fair to say that if I plug my headphones into the Benchmark headphone jacks, the two units will sound the same?
 
Now suppose I remove the power supply from one of them and replace it with a inexpensive wall wart switching PS.  Will  the 2 Benchmarks still sound the same?
 
Suppose I take the amp section out one of the Benchmarks and replace it with the amp section from my old Zhaolu?  Will  the 2 Benchmarks still sound the same?
 
Suppose I take the DAC section out of one of the Benchmarks and replace it with the DAC section from my Zhaolu?  Will  the 2 Benchmarks still sound the same?
 
Would all those variations measure the same as the untouched control unit?
 
 
USG
 
 
 
 
 
 
 
 

 

[bigshot]

When you start mixing and matching parts from different components, the component isn't designed any more. They're designed to perform to certain specifications. That's why they sound the same.

 

[nwavguy]

 
Quote:

 
But "sufficiently well engineered" becomes a bothersome statement as well, because it seeks to exclude exceptions.
 
Here's one of the problems I'm still having:
 
Take 2 Benchmark DACs, like yours, for instance.  Would it be fair to say that if I plug my headphones into the Benchmark headphone jacks, the two units will sound the same?
 
Now suppose I remove the power supply from one of them and replace it with a inexpensive wall wart switching PS.  Will  the 2 Benchmarks still sound the same?
 
Suppose I take the amp section out one of the Benchmarks and replace it with the amp section from my old Zhaolu?  Will  the 2 Benchmarks still sound the same?
 
Suppose I take the DAC section out of one of the Benchmarks and replace it with the DAC section from my Zhaolu?  Will  the 2 Benchmarks still sound the same?
 
Would all those variations measure the same as the untouched control unit?
 

Of course it's possible to degrade the DAC1 to the point it sounds different, but so what? John Siau at Benchmark designed it properly just as the engineer(s) at Grace Designs did with the m903. If you do a blind ABX with the two products, it's a safe bet they'd be hard to tell apart despite the fact they're very different designs.
 
The blind tests provide verification of "sufficiently well engineered". There are some generally accepted guidelines as to what's "good enough" and us engineers try to design to those goals. When we're done ABX provides significant validation of our work. There are lots of AES papers on the topic of "sufficiently well engineered" and I plan a future article to try and summarize some of it (it's a big topic).
 
Much of the paranoia over what's good enough should be put to rest by solid evidence like the Meyer and Moran study. Obviously it's not hard to hit that "sufficiently well engineered" goal when you can do something as dramatic as they did and nobody can hear any difference.
 
It seems to me you're mainly interested in finding some small chink in the substantial armor of solid engineering and blind testing? It's like going to the Porsche dealer and saying "yeah the 911 is a nice car but would it still perform as well if I put an old VW beetle engine in it?" Of course, not, but so what? It won't tow a big trailer either. And there's a bug splattered on the windshield from the last test drive. You can go on that way all day, but why? It's a great well engineered car for its intended purpose. Why not just enjoy the music?
 

 

[leeperry]

Originally Posted by nwavguy /img/forum/go_quote.gif
 
Meyer & Moran, for example, published a landmark AES study involving over 500 listening trials over a year including members of an audiophile club, recording engineers, and students with especially acute hearing. They played high resolution (DSD) SACDs on a high end system and sometimes, unknown to the listeners, switched an extra 16 bit 44 Khz A/D and 16 bit 44 Khz D/A into the signal path. At any realistic volume setting (to not expose the higher 16 bit noise floor during silent passages) NOBODY COULD TELL THE DIFFERENCE!

 
TBH, It's a well known fact that SACD sounds better than CDDA due to a much more careful mastering...DSD doesn't kill PCM per se, but the mastering is where the difference lies. I'm very fund of SACD rips, some of them were done digitally using an Oppo standalone player converting internally to 24/88.2, some were done in analog....they still sound much superior to their CDDA counterparts in both cases. I can provide short samples too. Steve Hoffman said that the Thriller SACD was the closest he had ever heard to the original master tapes, and the SQ is outstanding...even when converted to PCM.
 
CDDA can provide 96dB of dynamics, the medium is not at fault...the customers are, coz they want their CD's to sound as loud as possible. So we get garbage on CDDA, and audiophile quality on SACD/downloadable HD audio.
 
I won't get into the DAC DBT debate, as I also strongly disagree.

 

[nwavguy]

Quote:

 
TBH, It's a well known fact that SACD sounds better than CDDA due to a much more careful mastering...DSD doesn't kill PCM per se, but the mastering is where the difference lies. I'm very fund of SACD rips, some of them were done digitally using an Oppo standalone player converting internally to 24/88.2, some were done in analog....they still sound much superior to their CDDA counterparts in both cases. I can provide short samples too. Steve Hoffman said that the Thriller SACD was the closest he had ever heard to the original master tapes, and the SQ is outstanding...even when converted to PCM.
 
CDDA can provide 96dB of dynamics, the medium is not at fault...the customers are, coz they want their CD's to sound as loud as possible. So we get garbage on CDDA, and audiophile quality on SACD/downloadable HD audio.
 
I won't get into the DAC DBT debate, as I also strongly disagree.

 
 
I agree most SACDs are mastered differently and that accounts for their better sound. And I agree lots of music is over compressed but that's not the CD format's fault. It's been demonstrated over and over as a playback medium 16/44 CD audio really is good enough for music of any dynamic range you can reasonably listen to. That's what Meyer and Moran did.
 
They also, indirectly, demonstrated that not only do good DACs sound the same, they don't have any detectable "sound" at all. Essentially the same test has also been done with vinyl. You can put an A/D and D/A in the signal path of a high-end all analog system playing vinyl and, when the switching is blind, people can't hear all that extra hardware. There are references in my article.

 

[leeperry]

Originally Posted by nwavguy /img/forum/go_quote.gif
 
They also, indirectly, demonstrated that not only do good DACs sound the same, they don't have any detectable "sound" at all. Essentially the same test has also been done with vinyl. You can put an A/D and D/A in the signal path of a high-end all analog system playing vinyl and, when the switching is blind, people can't hear all that extra hardware. There are references in my article.

More often than not, those tests are run using loudspeakers in non-acoustically treated rooms...and most ppl will tell you that >75% of the sound of speakers *is* the room. Something that is very real using transparent headphones can easily end up totally irrelevant in such a careless set up.
 
All those fancy "studies"(for lack of a better word) about jitter being inaudible(when they actually used crappy headphones or better: mono signals recorded on tape...and I don't even make that up) or loudspeakers set ups trying to hear subtle differences between cables/DAC's just don't fit the bill. I call this pseudo-science, the same way some ppl will RMAA cables/opamps, end up w/ identical results and come up w/ the conclusion that they all sound perfectly identical.
 
You stated on your website that there's nothing worse than irrelevant measurements to prove a point, and that's pretty much all the "everything sounds the same" preachers have to show us to try convincing us...when OUR real world experience proves us the exact opposite. It seems that the strong believers of team "everything sounds the same" have a genuine need to naysay in order to feel happy...so they'd better show us real figures, maybe you could help them? You just said that all the DAC's sound the same, prove it?
 
I can find a lot of comparisons between DAC's, here's one I really like as it matches exactly what I think when I read the list of components that were used in those DAC's(ugly sounding 5532/4562 opamps, high ripple SMPS PSU's, mandatory ASRC, high jitter coax input chips, laughable "NOS" legacy DAC chips w/ ridiculous post-filtering/oversampling, etc etc): http://ravenda.wordpress.com/2009/08/14/audiogddac19/

1. PS Audio Digital Link III w/ Cullen Stage IV mod ($1290 ~ fairly warm, fairly transparent, but lacks the dynamics, tonal balance, and timbre of the DAC-19Mk3. I’d have picked the DL3 if I hadn’t already bought the DAC-19Mk3)
2. Benchmark DAC1 ($995 ~ harsh, flat out boring)
3. Cambridge DACMagic ($429 ~ does not sound any better than a $99 NuForce uDAC, a disappointment)
4. Red Wine Audio Isabellina ($2000 ~ typical NOS DAC sound signature, sounds warm and pleasant like a $260 Valab DAC and MHDT Havana. It is slow at everything it does. It does nothing right. Runs off the grid, very very expensive or should I say very overpriced for what you get in the unit)
5. Pico DAC ($349 ~ small, err… tiny, there is nothing wrong with the SQ considering its’ size, excellent value for the money and space
   
   )
6. AMB Gamma 2 ($230-parts cost only ~ superb DAC for its’ small size, my second best DAC, it sounds neutral like the Benchmark DAC1 but less harsh, one can also drive a headphone from one of the analog outputs)

 

[bigshot]

More often than not, those tests are run using loudspeakers in non-acoustically treated rooms...

That isn't true. I've read about studies conducted in the preferred listening rooms of audiophiles and sound engineers.

 

[Head Injury]

Quote:

I can find a lot of comparisons between DAC's, here's one I really like as it matches exactly what I think when I read the list of components that were used in those DAC's(ugly sounding 5532/4562 opamps, high ripple SMPS PSU's, mandatory ASRC, high jitter coax input chips, laughable "NOS" legacy DAC chips w/ ridiculous post-filtering/oversampling, etc etc): http://ravenda.wordpress.com/2009/08/14/audiogddac19/

Too bad they're all sighted and subjective 

 

[nwavguy]

There's a lot in leeperry's post that's questionable and subject to the exact involuntary bias I tried to document. And yes, many blind tests have been done in the person's own home, with their own high end system, listening to their own music. Everything is as familiar and consistent as possible. And they still fail miserably. The cable test referenced in my article was just such a test in 3 different homes.

I'm not saying that all DACs sound the same. Plenty do enough wrong they sound different. But among many of the well designed, well measuring DACs, yeah, I think they would be very hard to pick out in a blind test--even with your favorite headphones which takes room acoustics out of the picture.

The listening room used by Meyer and Moran is documented and photographed in the article. They tried hard, as they put it, "to give the listeners every opportunity". Again, it's just throwing rocks to try and say because there weren't 12 tube traps arranged just so the entire test is invalid. That's all anyone has managed to do with most of these tests: Throw rocks. Where are their blind tests showing the differences if it's just a matter of using the right system and room?

Think about the massive money behind all the bigger high-end manufactures. Some are even funded by giants like Harman. If these differences are so obvious, it would be pocket change for them to fund the studies that demonstrated people really can hear the difference in the expensive stuff. And it would be very good for their business. But do they? Nope. Everyone should ask why not?

The best thing would be to get leeperry in a chair with his favorite headphones, an ABX box, and two level matched DACs that measure very well but he thinks should sound different. And then see what happens!

 

[leeperry]

I'm not saying that all DACs sound the same. Plenty do enough wrong they sound different. But among many of the well designed, well measuring DACs, yeah, I think they would be very hard to pick out in a blind test--even with your favorite headphones which takes room acoustics out of the picture.

 
Here's an interesting quote: http://forum.rightmark.org/topic.cgi?id=4:504-3
 
"Everybody -in different locations, at different times, without knowing from each other- told the same story, that they found the differences between opamps more important than the differences in dac chips."
 
It is also many ppl's real world experience(including mine) that drastically different sounding opamps can completely change the SQ of the very same DAC...eg: AD797BRZ vs LT1028ACN8. They provide way different soundstage widths/depths, and way different "colors" as well. Have you done this kind of tests or are you basing your beliefs on what those "studies" showed you?
 
Here's what Ray Samuels has to say about those two opamps: Emmeline HR-2 Headphone Amp - Ray Samuels Audio

The HR-2 is a high resolution headphone amplifier. It incorporates two state of the art AD797 audio IC’s—the best ever made.

REVIEW: Ray Samuels Emmeline HR-2 - SGHeadphones

As Mr. Samuels once remarked, the LT1028 is like putting sugar all over your sound. It has a very liquid, sweet, and warm midrange that exhibits sharp bass and treble roll off. Unlike the Analog Devices op-amps, it is not dark. In fact, low level details flow through with greater clarity and definition.

These are also the same conclusions majkel came to: http://www.head-fi.org/forum/thread/397691/audio-gd-discrete-op-amps-reviewed-opa-earth-opa-moon-opa-sun-v-2
 
And that's also how I'm hearing those chips, and several of my friends as well(and many more ppl through google)...have you ever tried them personally?

Think about the massive money behind all the bigger high-end manufactures. Some are even funded by giants like Harman. If these differences are so obvious, it would be pocket change for them to fund the studies that demonstrated people really can hear the difference in the expensive stuff. And it would be very good for their business. But do they? Nope. Everyone should ask why not?

 
Coz their customers don't care, plain simple. They use non-acoustically treated rooms and have prolly never heard what a good audiophile DAC could do for them, and in good I mean using all the best components on the market: WM8804 on the coax input, top of the range opamps as previously stated, low ripple linear regulated PSU's, full galvanic isolation on all connections, etc etc. The audio world is quite nasty as there's always something better sounding out there. Harman have price points to reach, and you can see how the craptacular DacMagic does...so much for the "audiophile" bs Cambridge wants us to believe in. All the worst components together in one shiny box, impressive! NE5532/OP275 opamps, CM108 USB chip, mandatory ASRC, high ripple SMPS PSU...you would have a hard time doing any worse, or maybe you could by using the legacy CS8416 on the coax input =D

The best thing would be to get leeperry in a chair with his favorite headphones, an ABX box, and two level matched DACs that measure very well but he thinks should sound different. And then see what happens!

The real best thing would be to know how much you're willing to spend to see it happening!