DAC + tube amp showdown!

[PaulyT]

FINALLY getting around to doing some evaluation of Walls's gear that he has generously loaned me. Results to come shortly.

First the loadout, from left to right: Beyer T1 headphones (mine), Schiit Bifrost DAC (Walls's), Bottlehead Crack amp (Walls's), Woo WA6 (mine), Yulong DA8 DAC/amp (mine, in "pure DAC" mode with the internal amp turned off), Squeezebox classic (mine, connected to the DACs by optical/toslink).

Sorry about the cluttered table... but well, that's my house in a nutshell. :?

 

[PaulyT]

Using my usual evaluation track list (all ripped as flac, served via the squeezebox) :

:text-link:

 

[heeman]

I pray every night to protect me from the Headphone Beast that lives in our world...............

:happy-smileygiantred:

 

[CMonster]

Nice, I almost started a thread the other day to ask about this...

 

[PaulyT]

I pray every night to protect me from the Headphone Beast that lives in our world...............

:happy-smileygiantred:

Don't worry, there's still time to be bitten. I'll make sure to set aside a little time for you to listen to my rig in June. :evil:

 

[Flint]

Intriguing! I strongly recommend you calibrate the levels using a Volt meter with the amp's output under load to ensure you are comparing apples to apples. It only takes a very small change in SPL to perceive a preference for an amp over another.

 

[PaulyT]

Yeah... well... I understand what you mean, but the problem is that my demo tracks very WIDELY in overall volume level, so I have to constantly tweak the volume knob. It's not really practical to do such careful calibration as I'd have to redo it for every track individually. And with tube amps you just can't switch so fast, as I don't want to do a lot of plugging and unplugging when they're turned on; a tube amp powered on without a load (headphones) is not a good thing, and I only have one set of headphones.

I'm not doing instant a/b switching, either, where I think the issue of volume level is most critical. I set up a rig, and go through my entire track list. I realize there's no perfect objectivity here, but, as I see it, the goal of a careful evaluation of a series of tracks - taking written notes the entire time, something I learned from you BTW - is to try to form a reasonably fair opinion averaged over time. It's about as objective as I think I can be, and I admit it's not perfect.

 

[Flint]

And that is why these tests mean very little in terms of repeatable, accurate, scientific results of what you are actually hearing. It would be VERY easy to decide you like one of the amps merely because you inadvertently turned it up about 0.7dB higher than the others on a given track - and we all know how SPL greatly affects the overall balance of bass, midrange and treble due to the way our auditory system works as explained through the Equal Loudness Curve.

I am sure all of the amps sound amazing, but what you will report as one amp being a little more bright could be nothing more than the level was 1dB lower than the amp you were comparing it to and not the fact the amp was actually any brighter.

 

[PaulyT]

Well you'll just have to take my reports as non-scientific then.

 

[PaulyT]

And I don't understand, then, how you could have had a speaker rating system with 100 discrete levels, when all your evaluations of those speakers could not possibly have been precisely calibrated for SPL.

I'm not trying to write a scientific journal article here. I'm not trying to sell a product, or even to recommend to anyone one product over another. I'm chronicling my own experience, that's all. I think after spending a reasonable amount of time with each piece of gear, I can hear differences that are more than just SPL. But I have no way to prove that, so ... not even gonna try.

 

[PaulyT]

Sorry, I'm getting defensive, don't mean to be. My point is, I believe - but cannot prove scientifically - that my listening/evaluation method averages out these small SPL differences at least to some degree, and that I can arrive at some semi-objective preference for one piece of gear over another; that the slight differences I hear are real and not just SPL. And personally, what I'm evaluating isn't frequency response so much anyway, more of imaging accuracy and sense of space or three-dimensionality of the aural presentation. That doesn't change so much when I twiddle the volume knob, even within a given track.

 

[PaulyT]

Ok, pushing on, at the risk of digging myself into a deeper hole... Oh well, WTF, you guys already know I'm insane about headphones.

Spent several hours swapping around components and listening to various combinations of amp + dac. (Yeah I took the day off...) All using the same headphones (T1) and demo tracks.

First, the bottlehead crack amp. Right off, I had some issues with this amp. Mainly, the volume control is, from what I can tell, defective. There's a point below which if the volume is turned down, there's a major hum that suddenly kicks in in the right channel; not gradually, but an abrupt change at that point. I don't think it's a tube problem, because it only happens at lower volume settings; above that point, it's fine. It's very weird, but I'm guessing there must be some electrical short or something in the potentiometer in the lower portion of the dial. I tried the usual thing of turning the knob back and forth for a while which sometimes cleans up a slightly noisy volume knob, but no change.

This amp has some noticeable hum, especially if turned all the way up. (By contrast, the WA6 has no perceptible hum at any volume setting.) Even at the ~halfway point, it's audible, though very slight. Not enough to interfere with actual music listening at any volume I would normally have it at, but it's there.

The most bizarre thing was that I happened to have my home phone (cordless but not a cell phone) on the table next to me. When the phone came on for an incoming call, or even if I just opened the line to a dialtone, there was significant interference with the amp. If I moved the phone around, especially if I put it close to the left side of the amp (where the volume control is), it would hum and whine like I was playing a theramin. Yeah it was that noticeable. If the phone was turned off, nothing. Never had that experience with any other gear before. May be partly due to the fact that the Bottlehead is set in a simple wooden box, no completely surrounding metal box like the WA6's to minimize interference.


Anyway, as far as the sound of this amp - well it's a step up from my old Audiotailor Jade (from what I remember, I didn't actually go back and try that today, too many other tests to run), but it doesn't quite compare to the WA6. The precision of the imaging just isn't quite there, especially in the HF, like tambourines or delicate cymbal ticks. It's not that the HF is missing, just not as clean or precise. There were a few tracks that I thought sounded quite good, but on the majority, I felt that the breadth and depth of soundstage was somewhat less than the WA6.

Again, this is only with my T1 headphones, which are very high impedance. I may go back and try with my low-impedance Denons, which suffer markedly with the WA6; I got the qualitative sense that maybe the Bottlehead is a little more powerful than the WA6.

 

[PaulyT]

Next, the Schiit Bifrost DAC. This was a much closer call. I compared it to my Yulong DA8, using the WA6 amp and T1 headphones, switching out only the DAC.

First, I noticed that the optical port on the Bifrost is rather flaky; any movement of the cable would cause the audio to drop out. It's not the cable itself, as I have no such issue with the DA8 using the same cable. So I switched to using the coax port on the Bifrost, which worked just fine. I'm assuming of course that the digital transport doesn't affect the sound.

I can say that I think there's an audible difference between the Bifrost and the DA8. But it's very slight. And I fully admit that I could be imagining it because I WANT the DA8 to be better 1) because it's mine and 2) because it's significantly more expensive (though partly because it also includes an internal amp). With the DA8, there's an improved clarity of imaging, and depth of soundstage. Yeah I know that's an imprecise description, but I don't know how else to put it into words - the usual problem when attempting to describe an audio experience. Instruments are a bit more individually discernible, especially on something really thick and crowded like Rodrigo y Gabriela's Hanuman. Again the small, subtle HF sounds are sharper with with the DA8, and there's a greater sense of some instruments being located behind or in front of others.

Note that this is, from what I can tell (Walls may correct me if I'm wrong), NOT the "Uber" version of the Bifrost that Schiit offers as an upgrade.

 

[Flint]

And I don't understand, then, how you could have had a speaker rating system with 100 discrete levels, when all your evaluations of those speakers could not possibly have been precisely calibrated for SPL.

I'm not trying to write a scientific journal article here. I'm not trying to sell a product, or even to recommend to anyone one product over another. I'm chronicling my own experience, that's all. I think after spending a reasonable amount of time with each piece of gear, I can hear differences that are more than just SPL. But I have no way to prove that, so ... not even gonna try.

I realize that you aren't trying to write a scientific paper, but I was just offering advice I thought you'd be interested to prove to yourself that you are indeed hearing differences in the performance of a given amplifier rather than a difference caused solely by small differences in SPL. It is VERY common for people to compare electronic gear without calibrating levels. Back when I was hardcore on this stuff I would corner people I knew doing it and almost always could prove that what the listener heard as a difference - or at least the primary attributes they used to make a decision of preference - was tied directly to SPL levels being different between the two electronic devices being tested.

When I used to do tons of speaker auditions I would, indeed, calibrate the levels using an SPL meter held in front of my face. So, in fact, I was calibrating the output before attempting to generate an opinion of a speaker. I know a few salesmen who HATED that about me. I started doing the SPL calibration NOT because of the Equal Loudness Curve, but because my hearing would slowly get desensitized and I would inherently start turning up the levels as the day progressed. I was shocked to find that by the second hour of auditioning I thought the SPL was the same even though I had turned it up several dB (SPL) than when I was first listening to the first set of speakers. I honestly thought I had the levels the same, but my perception was flawed and a SPL meter was absolutely necessary to get the levels the same. In a conversation with Paul Klipsch on the subject, he pointed out the implications of the Equal Loudness Curve as being even more critical to matching SPLs versus mere sound pressure on the ear alone. Also, in my speaker rating scale I made it abundantly clear that is wasn't scientific as it was impossible to claim I experience with all those speakers in the same acoustic environment, but I did try to only include speakers which I had extensive experience with. I also point out that the difference between any two models of speakers is going to be gargantuan when compared the extremely minute differences between two pieces of electronics where a small difference of voltage will be vastly more audible than the tiny difference in frequency response, phase, or distortion (unless the device is not functioning properly).

So... that said, I know you will have fun making comparisons of these interesting amps. Please don't let me slow you down. I only make these recommendations (and then defend them) with the belief that you want to eliminate variables which can create a vast difference in perception while seeming to be unimportant. The audiophile community has too many flawed reports and tests where no consideration of these things are included.

Have fun!

 

[Rope]

Next, the Schiit Bifrost DAC. This was a much closer call. I compared it to my Yulong DA8, using the WA6 amp and T1 headphones, switching out only the DAC.

First, I noticed that the optical port on the Bifrost is rather flaky; any movement of the cable would cause the audio to drop out. It's not the cable itself, as I have no such issue with the DA8 using the same cable. So I switched to using the coax port on the Bifrost, which worked just fine. I'm assuming of course that the digital transport doesn't affect the sound.

I can say that I think there's an audible difference between the Bifrost and the DA8. But it's very slight. And I fully admit that I could be imagining it because I WANT the DA8 to be better 1) because it's mine and 2) because it's significantly more expensive (though partly because it also includes an internal amp). With the DA8, there's an improved clarity of imaging, and depth of soundstage. Yeah I know that's an imprecise description, but I don't know how else to put it into words - the usual problem when attempting to describe an audio experience. Instruments are a bit more individually discernible, especially on something really thick and crowded like Rodrigo y Gabriela's Hanuman. Again the small, subtle HF sounds are sharper with with the DA8, and there's a greater sense of some instruments being located behind or in front of others.

Note that this is, from what I can tell (Walls may correct me if I'm wrong), NOT the "Uber" version of the Bifrost that Schiit offers as an upgrade.

The "Uber" option is a significant upgrade for a mere $70.00. I changed the Bifrost board out and was pleasantly surprised at the sonic improvement.

Rope

 

[PaulyT]

Flint: Fair enough. And I get it. I really do. It's just that careful SPL calibration isn't practical with headphones where I can't just hold up an SPL meter. (And I doubt most SPL meters really could discern the ~.5dB level all that accurately...?) So I attempt with my evaluation procedure to average out the differences of SPL, such that the "signal" of what the actual sonic differences are can still be perceived. Maybe I'm just delusional, maybe not; I don't think so, but as I said, I can't prove it. Any audio system preference ultimately is subjective, no matter how careful you are. But I still maintain that the differences between electronic components as evaluated in a headphone rig are much more apparent than they are with a loudspeaker system.

During my listening tests, I'm constantly adjusting the volume. Even within a track, sometimes I'll go back and forth several times over the same section listening at different volumes to see if/how that changes what I'm hearing. I really am making a conscious effort to smooth over that variable. But if your contention is that SPL is the majority factor in any listening experience and that it's impossible to make a valid comparison without controlling that variable precisely, well... guess there's nothing I can do about that. Ultimately nobody can convince another person that what they hear is real, as audio is inherently subjective. So I describe what I experience here, not hoping to convince anyone else of anything, just trying to put my own impressions down in words. Take it or leave it.

 

[CMonster]

Note that this is, from what I can tell (Walls may correct me if I'm wrong), NOT the "Uber" version of the Bifrost that Schiit offers as an upgrade.

The "Uber" option is a significant upgrade for a mere $70.00. I changed the Bifrost board out and was pleasantly surprised at the sonic improvement.
Rope

It does have the "Uber" upgrade - I just double-checked and I do have a PM from Walls stating that.

 

[walls]

Yep it's the upgraded version. Strange that optical input is all I use and it's never given me any grief. Nonetheless I will check it when you send it back and if something is amiss Schiit will repair it for free.

On to the Crack, again I have listened to this amp for hours on end and have never heard a hum of any kind. Did the hum stop after you moved your phone then? If so if you have a volt meter go on the Bottlehead website and get a chart and do a quick check, maybe something jarred loose in transit?? I have a horrid ground loop problem with my WA2 that I have yet to figure out. Damn thing is dead quiet on every outlet I try in the house EXCEPT the one by my nightstand. Lol
I actually really thought that you would love the Crack, but if there is an issue then maybe it needs to be addressed to do a proper test.
Another thing you can try would be a different tube in the crack. If you have a TS 5998 it makes a huge difference.

 

[walls]

Also I have stated that I as well like the crack better with my HD650's rather then my T1's. And from what I read I am not alone in this regard, the crack for whatever reason just kills with the 650's. But on my WOO I strongly prefer the T1's.
The 650's are 300ohms while the T1's are 600ohms so maybe that is the main difference.

 

[PaulyT]

Another thing you can try would be a different tube in the crack. If you have a TS 5998 it makes a huge difference.

Ha! Actually, I do, as a matter of fact. It was one of the tubes I rolled with my Jade. So that's a drop-in replacement for the one you sent with the Crack?

What about a 6AS7G? I have a Chatham 6AS7G in my Jade, which ultimately I preferred over the Tung-Sol 5998. I wonder if that would work as well?

What's the small tube in the Bottlehead? I may have replacements for that as well...



There were three different hum problems I encountered with the crack. 1) some hum continuously increasing as the volume knob was turned all the way up (common to many tube amps), 2) hum when my phone was near it, but that I consider just a weird interference anomaly, not present when the phone was off or far away, and 3) *loud* hum with the volume less than about 1/4, but which cut off quite suddenly when the volume was turned up higher than that specific point; it was very definitely connected with the volume potentiometer.