December 2013 Mid-Level DAC Comparison
Nov 15, 2013 at 5:41 PM Post #76 of 1,331
If you're opening that door, arguments can be made for many other similarly lower priced dacs. Heck, open the door to the pro audio sector and you've got several dozen more candidates. Focusrite, Apogee, Native Instruments, MOTU, Lavry, Antelope, RME, etc...
 
Nov 16, 2013 at 2:56 AM Post #77 of 1,331
I match using a cell phone app. I'm sure it doesn't give accurate absolute measurements, but I believe it gives accurate relative measurements, and that's all we really care about.


Volume matching $3,000 DACs with a 50c microphone doesn't make any sense. Please, at least use a multimeter to measure the voltage on the DACs line-outs. They should output about 2 volts into your headphone amps. A range of 1.9V to 2.1V means a difference of about 0.44dB below and above the reference level; a range of 1.95V to 2.05V means a difference of about 0.22dB both ways.
 
Nov 16, 2013 at 2:59 AM Post #78 of 1,331
   
 On a side note, you suggest that more expensive gear does generally offer worthwhile improvements [if not necessarily as good a value proposition]; which is reassuring. I think!

 
Anoobis' post leads me to think I need to give you all a bit of my philosophy so that you understand why I'm going to put up ~$10K (mostly refundable) to test a bunch of DACs over the holidays.  If you are one of those folks who thinks anything related to philosophy is BS (and you engineers know who you are), just skip this post, I'll get to the reviewing in a few weeks, and then you might be interested in reading my "stuff" again.  Or not.  My pay is the same either way.  So here goes:
 
In 40 years of chasing "high-fidelity audio" I have concluded that some more expensive gear offers worthwhile improvements -- sometimes those improvements are even significant enough to provide a good value in my view.  The trick is figuring out which hardware that is, at least to my brain via my ears.  As I have often pontificated with excessive verbosity on these boards, this is all personal choice.  While there are certain biological commonalities in all humans, there are enough differences in physiology and experience that what I think is a good value is not necessarily what others think is a good value.  E.g., I chose to spend the extra $1K to get LCD-3s over LCD-2s.  Others think that's a waste of money.  They are likely correct, for their brains and their money.  My brain (what little there is of it), my money (what little there is of it), my choice.
 
In the old days, I spent many hours in audio salons listening to all of the hardware, whether I could afford it or not. Over all of those years I established a personal scale for performance vs. price that I use to make decisions about what I want to own, and what I don't want to own.  Of course sometimes I screw up anyway and make purchases I end up regretting.  Life sucks that way.  But I prefer screwing up on my own to just taking other people's advice and buying what's popular, or well-marketed. 
 
Once again:  there are no unbiased opinions in audio.  Everybody's opinion is biased by something: the limits of the audio spectrum they can detect, how well they can process that detected signal (some people really are tone deaf), how much experience they have that allows them to match reproductions with originals, etc., etc.  Professional reviewers are biased.  I am biased.  All of you are biased.  That's a good thing:  if we all received and processed audio signals from 20-20khz exactly the same, there would be no Head-fi.  There would probably only be one website with fully calibrated test results, and we would all just buy the hardware able to produce the flattest frequency response within the audio spectrum, and that would be it.  Of course some folks do that anyway... but I've learned over the years that the hardware with the flattest frequency response doesn't sound the best to me.  My personal in-body receivers and processor are thus flawed, dooming me to spend my nights and weekends (and some days, if truth be told) seeking out that hardware whose output flaws most closely match up with my input/processing flaws to provide the most pleasing total experience -- to me. 
 
Unfortunately, these days the number of true audio salons where I might seek such imperfect hardware to audition is down to a sparse few.  So the point of this comparison test is for me to try, in my home with the rest of my system, a number of DACs, some of which are quite low cost (My Emotiva XDA-1 cost me $200) some of which are bordering on high end ($2K is a significant amount of money to most people) and the rest are in between.  Marketing and fanboy claims make it hard to determine which units in this range are most likely to match my flaws and which aren't, so I'm asking the community to suggest entrants, and I'm using my best judgment to pick a representative field that will offer the best chance of finding that perfectly flawed hardware at the right price to provide me a good value, or even great value, DAC.
 
Why do I even ask you folks for inputs, if our biases are probably so different?  Because, of course, I'm too lazy to search every mention of every DAC on every website in the world.  But I know that many of you are even more addicted to this stuff than I am, and have heard every piece of hardware out there, or have read every piece of information or marketing hype available on every website.  Biases or not, you know what's out there so I'm using you as scouts.  I have promised to keep as open a mind as I possibly can about what should be in the test, and I've mostly done that, except for the poor ODAC (sorry Anoobis), which all of the data available says is not the right match for my imperfect personal audio appreciation system.  Why?  Because I'm the guy who paid $1K extra to get the added realism that the LCD-3 offers over the LCD-2, even though I went into that particular comparison test hoping that the HE-500 would be my favorite so I could save money.  So the chances of me liking the ODAC over 9 well-regarded mid-range DACS is pretty close to zero.  As I said in my earlier post, that isn't fair to the ODAC.
 
My mind is mostly open about what I audition, as long as it has to potential to be competitive.  I believe my mind is also as open as possible about picking a winner among those that do make the field.  I owe no allegiance to any manufacturer or supplier.  As I proved with my prior tests of both cans and amps, I'll pick whichever hardware I like best, price be damned.  If my current XDA-1 wins, fine.  If the most expensive new DAC wins, that's fine too.  If I'm stupid enough to fall in love with 2 units, I'll probably buy both, as I did with the amps.  I like spaghetti, and I'll eat it every day of the week if that's what it takes to afford what I want.  Within reason, of course:  price limit for this comparison test is $2500, unless somebody can make a VERY compelling case for something slightly above that range.  But I'm not spending $10K or even $5K on a DAC.  My insanity has some limits.
 
And finally, why should you help me in this?  First, because I'm such a nice person.  Okay, maybe not.  How about because I write so well?  No?  How about because I write so much?  Hah!  How about I break in your new Yulong DAC for you (Hey, that's apparently enough for one of you, thankfully)?  Maybe because you'll feel better about your own DAC choice if I pick the one you already own? 
 
Or maybe because in return for your help I will provide as exact and detailed a set of reviews as I can regarding how I think each DAC sounds to me.  My reviews will be biased, but I hope that I will be able to provide some service to some of you:  not a definitive answer, but a few clues or snippets of descriptive information that might help you narrow or broaden your own choices of what products to audition for yourself.  And maybe a little entertainment value, whether in reading my verbose posts, or in arguing with me about them.  That is why we come here, after all... right?
 
Nov 16, 2013 at 3:22 AM Post #79 of 1,331
Gary...you rock man! Long live the individual!
beerchug.gif

 
Nov 16, 2013 at 3:38 AM Post #80 of 1,331
Volume matching $3,000 DACs with a 50c microphone doesn't make any sense. Please, at least use a multimeter to measure the voltage on the DACs line-outs. They should output about 2 volts into your headphone amps. A range of 1.9V to 2.1V means a difference of about 0.5dB below and above the reference level.


I could do a multimeter check.  Most of these DACs do output 2V into SE, but I can double check it. 
 
Note that I've found that my relative matching system is pretty effective.  The microphone in the phone is not that much worse than the ones in the cheap SPL meters that I would buy for this kind of testing, and in any case it is equally bad for all of the hardware, so as long as it is consistent, that is more important than absolute accuracy. 
 
In fact, the true dB level is actually irrelevant.  I''m going to listen to my reference tracks at a pleasing level using one of the DACs (probably my current one), then play white or pink noise through the system at that volume setting, and measure that SPL level with the phone held in the cans in a particular way.  Then I will set all of the other DACs to reproduce that noise level with the phone held the same way.  The process isn't laboratory perfect (trust me, it's very hard even to control the background noise in my house to +/- 0.5dB), but it doesn't need to be for this activity. 
 
Also note, level matching has its limits in the kind of comparisons I'm going to be doing.  I'm not a professional audio reviewer or doing blind A/B testing.  I am trying to discern which of these DACs is most pleasing to my ear, and under what circumstances.  In normal listening, we adjust the volume to the level that pleases us most at that particular time for that particular piece of music.  There is an argument that, to do a true, real-world test of these DACs -- as opposed to a perfectly analytical, scientific laboratory test -- I should adjust volumes during the test to allow each DAC to be at its best.
 
In the end, I intend to do both kinds of testing.  While I will do one set of comparisons with the levels matched, I will give those that don't sound best at that level additional opportunities at whatever levels I can find that do make them sound as pleasing as possible.  I don't want to miss out on a particular competitor's best performance just because I happen to set the level for these tests higher or lower than its sweet spot.  And conversely, I don't want to over-value a competitor just because I happened to test it in its sweet spot, but it actually gets harsh or muddy at higher or lower volumes. 
 
Make sense?
 
Nov 16, 2013 at 3:48 AM Post #81 of 1,331
The microphone in the phone is not that much worse than the ones in the cheap SPL meters that I would buy for this kind of testing, and in any case it is equally bad for all of the hardware, so as long as it is consistent, that is more important than absolute accuracy.


A multimeter is all of the above: cheap, consistent and accurate. 

There is an argument that, to do a true, real-world test of these DACs -- as opposed to a perfectly analytical, scientific laboratory test -- I should adjust volumes during the test to allow each DAC to be at its best.


Then don't bother!
 
Nov 16, 2013 at 3:54 AM Post #82 of 1,331
Keep the BMC in if you can. I've had it for some two months. Remarkable (I won't say why as it's for you to decide if you agree or not).

Good luck. Took me two weeks of intense work to compare three dacs. You'll be plenty tired at the end of this one.
 
Nov 16, 2013 at 10:34 AM Post #84 of 1,331
Well said Gary... I'd love to know what the best price/performance dac is within your comparison, so that I'd know what to upgrade to when I actually manage to get some money in the future. Based on what I've read, I feel like the Gungnir is going to perform really well despite being in the lower price range of this comparison.,
 
Nov 16, 2013 at 11:26 AM Post #85 of 1,331
A multimeter is all of the above: cheap, consistent and accurate. 
And I have one that I will use in this test.
Then don't bother!

I think both ways of comparing have merit.  I will level match to the best of my ability, but I'm not going to spend hours calibrating the test setup each day.  I want a fair competition.  However, once I've done those comparisons, I will work to find the sweet spot for every DAC.  I'll then note what it sounds like to me at that level.  There's a possibility that the results of the test will differ based on the two methods.  Or not.  But I will let you folks know exactly what I've done in each case, and exactly what I've found.  It's not that much extra work, and there's really no penalty other than extra time, as long as the moderators on this site don't put a word/character limit on m
 
Nov 16, 2013 at 11:39 AM Post #86 of 1,331
Keep the BMC in if you can. I've had it for some two months. Remarkable (I won't say why as it's for you to decide if you agree or not).

Good luck. Took me two weeks of intense work to compare three dacs. You'll be plenty tired at the end of this one.


I'm already tired... but it's a good tired... at least until I fall asleep while mowing the lawn and run into a tree.
 
I have made no decisions yet about what's in or out other than the ones that are definitely in, cited above.  And at this point, even those aren't in for absolutely sure yet, because there is no guarantee until they actually show up at my door.  So stay tuned, and please provide any additional information you might have that would help me determine which of the 9018 DACs should represent that chip in the competition.  Just having multiple representatives using the same chip should be really interesting and informative.  I just want to make sure we get enough of the different flavors to provide good information to folks.  To put it another way, I don't want to include 3 Sabre DACs that sound alike... what's the point there?  I want at least one (better if there are more) that is significantly different from the others, to see what the potential range is with that chip, since it is currently the dominant decoder among upper-mid-grade DACs.
 
Nov 16, 2013 at 2:22 PM Post #87 of 1,331
Gary, if it's any consolation, I can't think of a finer more broad list that represents overall cream of the crop in this category. Due diligence has been served!
 
Nov 16, 2013 at 2:42 PM Post #88 of 1,331
  I think both ways of comparing have merit.  I will level match to the best of my ability, but I'm not going to spend hours calibrating the test setup each day.  I want a fair competition.  However, once I've done those comparisons, I will work to find the sweet spot for every DAC.  I'll then note what it sounds like to me at that level.  There's a possibility that the results of the test will differ based on the two methods.  Or not.  But I will let you folks know exactly what I've done in each case, and exactly what I've found.  It's not that much extra work, and there's really no penalty other than extra time, as long as the moderators on this site don't put a word/character limit on m

Level matching is such a thorny subject!  I have had a few instances where I had extreme difficulty level matching two components.  Turns out, they were horribly mismatched in frequency response.  It really is impossible to level match in that circumstance.
I've also had the occasion where I had difficulty level matching the headphone output with my SPL meter and fell back on my volt meter.  My meter doesn't like pink noise or white noise.  The averaging is just not long enough for what I consider a decent result.  My scope does a much better job because I can control the averaging period WOOHOO!  CONTROL!  Oh, that's the control freak in me coming out...
 
Back to level matching with a multi-meter:  The compromise I ended up with was using a multi-point comparison.  Use a tone generator to generate 3 separate tones.  One between 100 and 500 Hz, one between 1 kHz and 5 kHz and one between 5 kHz and 20 kHz.  You can slide them around as you wish.  Measure the RMS output voltage at the speaker tap outputs (NOT the DAC outputs).  Then set the competitor to match at the middle frequency and note offsets at the two ends.
 
Its a bit more work, but very revealing.
 
Burn in that Yulong!  I read one report of the USB input not "getting there" until somewhere around 180 hours....  A week of continuous run... Hmmmm.... Still don't understand the whole "burn in" phenomenon.  Not that I haven't personally experienced it, I just don't understand it!  Hypotheses welcome!
 
Nov 17, 2013 at 8:03 AM Post #89 of 1,331
 
I could do a multimeter check.  Most of these DACs do output 2V into SE, but I can double check it. 
 
Note that I've found that my relative matching system is pretty effective.  The microphone in the phone is not that much worse than the ones in the cheap SPL meters that I would buy for this kind of testing, and in any case it is equally bad for all of the hardware, so as long as it is consistent, that is more important than absolute accuracy. 

 
I agree with this part.
 
But I do think that in fairness, you can only properly compare one DAC with another DAC only if they are SPL matched.
 
Just my 2 cents worth......................
wink_face.gif

 
Nov 17, 2013 at 8:21 AM Post #90 of 1,331
   
I agree with this part.
 
But I do think that in fairness, you can only properly compare one DAC with another DAC only if they are SPL matched.
 
Just my 2 cents worth......................
wink_face.gif

Absolutely - a difference of as little 1db will confer an advantage - but using the phone app to confirm levels is sufficient - it's enough work to examine several DAC's without adding to that workload.
 
I should also add that even if the DAC specs SAY they are at the standard 2v output - most manufacturers know how most "lax" comparisons in showrooms work (just keep the volume level the same between DACS specced at 2v output), and tend to "goose" their outputs by 1-2db (just enough to not be perceived as "louder", but enough to be perceived as "better").  Certainly, something to take into account, lest all the hard work be invalid.
 

Users who are viewing this thread

Back
Top