[Toe Tag]

What I meant by brain burn-in was (and I'm not asking Gary to take this on) to take a DAC and listen to it for 40 hours of actual logged listening time. Not just turn it on or run music through it and into the closet for 100 hours. This is going off the theory that you're burning in your brain, not the capacitors. Then rinse and repeat, with a different DAC, for another 40 hours. And so forth. I believe that you'd then hear differences. Further, this test could be performed with less expensive DACs (since probably takes longer than the returnable period). Say, Emo vs. DACport LX vs....

 

[Gary in MD]

I understood what you meant, but that in fact does not produce reliable comparative results.  In fact, people have really bad memories when it comes to comparative listening.  If you really want to compare equipment using humans (as opposed to calibrated lab equipment), A/B testing is really the only reliable way to do it. 
 
On the other hand, reliability in the results doesn't matter if you are happy with whatever hardware you selected.  Unfortunately for me, I am the type of person that wants to A/B everything... Head-Fi, yard equipment, cars, girlfriends... My methodology worked great with cans, amps, cars, and tractors... not so great with girlfriends, with DACs also currently proving a bit problematical. 

 

[PXSS]

I understood what you meant, but that in fact does not produce reliable comparative results.  In fact, people have really bad memories when it comes to comparative listening.  If you really want to compare equipment using humans (as opposed to calibrated lab equipment), A/B testing is really the only reliable way to do it. 

On the other hand, reliability in the results doesn't matter if you are happy with whatever hardware you selected.  Unfortunately for me, I am the type of person that wants to A/B everything... Head-Fi, yard equipment, cars, girlfriends... My methodology worked great with cans, amps, cars, and tractors... not so great with girlfriends, with DACs also currently proving a bit problematical. 

Gotta teach me how to get the girlfriends to agree to an A/B comparison. I just keep on getting slapped across the room everytime I bring it up!

 

[Gary in MD]

Honestly I find it strange that all DAC sound like the Benchmark. Wasn't this guy described as bright and lean? I hear nothing of that with the Dangerous Source.

About the Ciunas: it's most likely the most neutral DAC of the bunch. Thanks to its battery, it should have the blackest background of the bunch which should bring out details. Weird that you can't hear that honestly.

Then come the forceful DACs like the Chord. Typical British PRaT that only crave for rock songs being played. These bring out details by forcing on the transients a bit. The NFB-27 is among those two.

I'm trying to spoil you here

The Benchmark DAC 1 was supposedly bright and lean... they claimed to have fixed that with the DAC 2.  I don't know that for sure since I don't have a DAC 1 here.  All I know is that several of these DACs exactly like the Benchmark DAC 2 to me.
 
Since I don't actually know what the absolute reference standard actually is, it could be that the Ciunas is the neutral one, and that the Indistinguishables are in fact lean.  I can only tell you that the Indistinguishables are on one side of the Ciunas and the Metrum is on the other side.
 
I might be able to tell you where the Chord is relative to the others if you folks would stop inciting me and let me listen to the DAC in peace.  This has been a completely wasted night for testing, though I have been listening, sort of, and it sounds good... I just have no idea how it compares to the others, because that takes my full concentration.

 

[Gary in MD]

Gotta teach me how to get the girlfriends to agree to an A/B comparison. I just keep on getting slapped across the room everytime I bring it up!

Yeah, me too... I told you it didn't work out that well for me...

 

[Argo Duck]

Glad you mentioned girlfriends Gary, because (hopefully) the best girlfriend for me isn't the best for you, and vice versa. Works a lot better when we don't have everyone chasing the same girl/boy.

Head-fi gear is probably a lot like this. There simply isn't an 'ultimate' DAC, amp, headphone etc. In fact we know there isn't. Some people swear by the LCD3, others the HD800 for instance.

On another matter, I'm interested in the statement 'all have the same detail'. Broadly speaking, I found this was true in my 3-dac review.

However, being conscious of how memory works (as currently understood), an issue I debated and didn't resolve during that review was that once you hear a detail in a track and know it's there, your brain tends to reproduce it on subsequent occasions. Once our brain has learned something, it constantly adjusts its inputs to be consistent with what we now know.

For example, I would listen to one track with one dac, then notice something new with the next dac. Fascinated, I would switch back to the first dac. Oh, I hear it there too now...

Or did I just miss it the first time through? Because another factor with our brains is 'the focus of attention'. When there's a lot of information coming in, we tend to use selective attention. Problem is this can change moment by moment - the same track on the same gear doesn't necessarily sound the same on a repeat listening.

Getting back to these two dacs, another possibility is both contain the same resolution of detail, but the presention is different. In the example above, the second dac presented a particular detail more 'forwardly' so that I noticed it (in perception terms, it crossed the threshold to reach conscious attention). Of course it's present in the first dac too, but - what? - a few percent less forward? So naturally, once I've learned it's there, I hear it every time in both dacs. And by now, probably in any other properly-functioning dac I ever listen to in future.

This does all come back to problems Gary, I and others discussed earlier in this thread about what kind of methodology would 'work' for this shoot-out. On one hand, it's an advantage to use a "within-subjects" design such as here, where the 'measuring instrument' (Gary) is the same one for all 'conditions' (dacs). OTOH, learning is a well-recognised issue with this design. The alternative "between-subjects" design (different subjects are the measuring instruments) is approximated when different owners of different dacs compare notes, such as we often do on head-fi and have done here too in recent pages. Of course there are many missing controls from these casual comparisons. Their usefulness is limited though I personally think they have some positive value.

I do believe we have to think about problems and limitations of the present work - if any - and at the same time acknowledge Gary (and Barry too) for having done a fine job with the process. Gary, I sense you need a good rest mate but be assured it has definitely been in a good cause. It stands alongside other careful dac comparisons as a serious piece of work that - along with these others - enhances our knowledge and understanding of dacs and their differences, or lack of them.

And as has already been pointed out, it provides a great basis for the next comparison...

 

[Toe Tag]

+1 AiDee. And IMHO you have provided the explanation for burn-in. It is real, as many will staunchly defend. But it happens in the brain (neurons) not in the capacitors (tantalum). 
 
Also, here is the Purrin thread http://www.head-fi.org/t/693798/ranking-of-17-dacs-and-dac-configurations
 
Gary, apologies if thing got stirred up. It only underlines the importance of the work you have taken on for us. It seems you are moving closer to conclusions, so I will try to stand by for that. 

 

[Gary in MD]

Aidee:
 
All excellent points, as always.  I am definitely aware of the learning issue, and the focus of attention issue.  In fact I try to use the latter to try to remedy the former to some extent.  I will hear a detail in a song on a particular DAC (often knowing the detail is there from knowing the song well to start with), then focus on that particular detail several times through the snippet of song, comparing how that detail is rendered by each DAC in turn.  Play, rewind while switching.  Play, rewind while switching.  I hate to tell you how many times I've listened to the first kick drums in Hotel CA, or the chair creaks at the start of Dave Matthews' The Riff, or the sounds of fingers on the fret boards in Mumford's "Broken Crown" along with similar individual details in each song.  What is really nice about some songs is that they repeat a  particular sound multiple times -- rhythmic cymbal strikes or percussion, bass lines, repeated bars in blues-based jazz.  Repeated sounds are the comparison tester's friends!  Then all I have to do is switch!  Does that detail sound the same on each DAC (attack, reverb, sustain, decay)?  Is it equally loud?  Is it equally long?  Etc., etc., etc.  For hours and hours. 
 
Of course I am using my brain to measure each of these characteristics, but I really try to focus on differences with each of these details.  When I hear any differences, I note them.   I am totally focused on finding differences, and in fact worry more about bias in finding differences that aren't there than in missing detail that is. 
 
Is what I'm doing perfect?  Of course not.  But this is for fun, and at no time have I ever claimed that it was scientific research for the betterment of all mankind.  It is my ears, my brain, my setup, trying to find out whether I can pick a winner -- for me.  I appreciate all of the help I have gotten, particularly the hardware folks have sent, and I am doing the best I can to be objective and open and respond to every valid suggestion for improvement of the process.  But nobody ever pretended that this was a full-fledged double-blind test with large numbers of test subjects and fully calibrated equipment. That would take lots of money and time, and nobody is funding that study folks... so you'll just have to make do with me and my flawed ears and my flawed brain and my flawed process and my flawed LCD-3s... or do as I always recommend and don't take my word for it -- compare this hardware for yourselves.  All it takes is time and money...

 

[boatheelmusic]

All - 
 
I think we all agree we appreciate Gary's long hours and diligence on this project.  And in so doing, many of us are gaining learning and very possibly saving $ in the process.
 
I want to propose starting a "Gary DAC Fund" whereby each of us chips in $5 or $10 to help Gary fund his next DAC of choice.
 
Surely this work is worth the cost of a stereo magazine, but without ads?
 
I'm not sure how this could work, perhaps through Paypal, but I'm up for it.
 
Bill

 

[Argo Duck]

^ Great idea Bill. You can count me in.

 

[nicko7i]

^ A 'donate' button is a fine idea.  I'd contribute.

 

[goaliedad39]

This whole thread has been extremely informative for me and I would most definitely be willing to contribute.

 

[boatheelmusic]

Hearing some interest, I've reached out to headfi management to see how we might set up a "contribute" button or other means for folks to help without exposing Gary's email address or other personal information.
 
Stay Tuned!

 

[zilch0md]

Wow!  
 
A lot of good stuff was posted in the past 12 hours.
 
And speaking of time, Gary's last post was made at around 3:20 AM Eastern.   

 
Quote:

  Purrin's 17 DAC comparison and the spreadsheet you can find if you google compares some DACs ability to resolve detail.
 
What has been said repeatedly is that the HD 800 would be better for these tests.  While I agree with Gary's assessment that the LCD-3 is what he uses and what he's testing for, the HD 800 is better at resolving these details.  High end speakers, too, I'd imagine.  One of the main points drawing me to the M7 is it's reported detail resolution ability.  I haven't heard it, and I'm not even sure my 38 year old ears would be capable of hearing the difference, but it's out there, and so I want to try it.  It's a curse.
 
Gary, I'm convinced that you're right.  Yet, I'm sure that if someone spent a whole day listening to one song with all of these DACs with HD 800s or good speakers, they'd hear some detail differences.  It's totally not worth it.  I only posted that to address what Toe Tag was saying.
 
I'm anxiously awaiting your results.

 
@cizx - Not desiring in the least to ruffle Gary's feathers, nor yours, I actually agree that for the explicit purpose of comparing these DACs' ability to resolve detail, the HD800 would likely be the ideal headphone (if not some choice of speakers).  That said, if it takes an HD800 to hear the difference, is everyone going to buy an HD800 to go with the DAC of choice?  As a newcomer to listening with HD800's, I am really glad he used the LCD-3 - a headphone that, as I've said before, is among several candidates in its ability to perform well with a wide range of DACs and amps, unlike the HD800.
 
But, yes...  if Gary were to extend the testing to include critical analysis of detail rendering, the HD800 would be a good choice, in my opinion, but how far behind is the LCD-3 in this regard?  And would that effort be of any value for people who have lower-resolving headphones?  
 

  Keep up the great work Gary, we all appreciate you

 
Roger that!  There are no true antagonists here - only supporters - with a lot of ideas being tossed on the table and a great exchange of opinions.
 

  [snip]
 
The DM Source has thus been my "reference" DAC representing the Indistinguishables in testing against other DACs for the past several days.  I am comparing each of the others (Metrum, Ciunas, Benchmark, Chord, Yulong) to the DM Source.  Those that don't sound different will be considered Indistinguishable (the Benchmark).  If an Indistinguishable offers good value, it will be evaluated again at the end.  If not, it goes back (the Benchmark -- big feature set, but pretty much the same as the DM and Emo for twice the price).  Those with audible differences (Ciunas and Metrum for sure, Chordette maybe, Yulong I dunno, I haven't listened to it) will end up facing off against the DM and Emo in the final showdown (the latter two will be facing off in features, particularly their headphone outs). 
 
In the end there can be only one.  Or maybe two, if they are cheap enough. 
 
[snip]

 
I'm good with this - it's a valid strategy.
 

  All - 
 
I think we all agree we appreciate Gary's long hours and diligence on this project.  And in so doing, many of us are gaining learning and very possibly saving $ in the process.
 
I want to propose starting a "Gary DAC Fund" whereby each of us chips in $5 or $10 to help Gary fund his next DAC of choice.
 
Surely this work is worth the cost of a stereo magazine, but without ads?
 
I'm not sure how this could work, perhaps through Paypal, but I'm up for it.
 
Bill

 
When Gary reads this he's going to roll his eyes, but the gesture is genuine, I'm sure - and I'd like to participate, too. 
 
If I were in his shoes, I'd consider my time to have been a gift to the community, but I'd welcome some assistance on all of the costs.  
 
What do you think, Gary?  (Just say, "OK.")
 
If you're lucky, you'll receive enough to buy a sandwich at your favorite deli.
 

 
Mike

 

[Gary in MD]

No reimbursement is necessary or desired.  Everybody just be quiet about the HD800s already.