Tube Vs. Solid State

Well, I don't have a Jotun or a Crack, or else I might haha.

Wouldn't that run the risk of double amping?

 

um, it IS double amping. That being said you run into the same problem with almost any active preamp attached to the to the Jotunheim. Try to only use passive preamps with Jotunheim. Or better yet, don't use any preamps.

If using the crack as a preamp you should really only use it with speaker amps.

 

Mm, gotcha. Thanks kindly, I do appreciate it!

 

So I have a question.
Are tube amps objectively better? Do they perform electrically better than their solid state counterparts?
Or is it more of a subjective timbre/plankton type deal?

 

Painting with strokes so broad that I'm basically using a paintgun to apply Bozo's post mortem makeup... tubes have higher magnitude but lower order distortion, solid state has lower magnitude but higher order distortion. Varies hugely on design/topology/feedback/voodoo/blah blah blah. Again... paintgun meets Bozo, with a touch of drunken Dali.

 

Tubes seem to have the potential for more dimensionality than solid state.

 

And @OJneg will no doubt mention resolution:

 

If directly comparing a tube to a transistor the tube measures much better then a transistor does.

http://spectrum.ieee.org/consumer-electronics/audiovideo/the-cool-sound-of-tubes/3/distortion

Yet SS amps usually have better measurements than tube amps. Why? It's because transistors are cheap enough where you can have a lot of them in a circuit and then apply tons of negative feedback. You can conceivably do the exact same thing with vacuum tubes but it becomes rather unwieldy and really expensive.

 

It's not about the number of transistors, it's about open loop gain. You use that that to calculate how much NFB you can afford to apply. Output device parallelisation is usually done to get more current or to use small signal transistors.

I'd blame the small signal performance for most of the tube magic. The smaller the signal the better the tube is at amplifying it.

 

I'm not a fan of blinding LEDs either. That's why I spent quite a bit of time finding a subtle blue LED for the HP-1 and mention it in the feature list.

A fully transparent amp would deliver Vout = Vin * A, where A is the amplification factor (real number). This is a completely linear function. There's nothing added. No THD. No IMD. Nothing. Just gain.

With its ultra-low THD and IMD, the HP-1 is as close to Vout = Vin * A as I can get with currently available parts.

The DG300B does have some 2nd, 3rd, and 4th order harmonic distortion. This means if you input a pure 1 kHz tone, you will get 1 kHz, 2 kHz, 3 kHz, and 4 kHz out. The 1 kHz will be the tallest spectral component and measure about 3*Vin in amplitude. The 2,3,and 4 kHz components will add up to about 0.2 % of the amplitude of the 1 kHz signal (assuming 1 W out anyway - back of envelope math). These 2,3, and 4 kHz components were *NOT* in the input signal. They were added by the amp due to its nonlinearities. That's how distortion in amplifiers works.

Exactly.

I own an Audio Precision APx525 audio analyzer. Audio Precision has been the industry standard for audio measurements since the 1980ies. They're the gold standard. The APx525 is their second-best analyzer. The only reason I don't have the APx555 (the top of the line) is that the base model starts at over $32k. A fully loaded model is over $50k. The APx555 had just been released when I bought my APx525.

With the APx525, I can present a (nearly) pure sine wave to an amp and monitor the amp output. It is very easy to see and quantify how much distortion is added by the amp.
For those measurements where an even purer sine wave is needed, I have a precision oscillator. Its THD is better than -140 dB. That's as close to a pure sine wave as physics allow these days.

My cool toys are calibrated, so they won't be wrong. I have full confidence in the data I post.

Some people like the tube distortion because it's mostly lower order. It adds a little warmth (higher THD at LF) and harmonic extension. These effects are well documented (see any Sensing & Perception psychology text). Also, I believe there's some research behind the addition of low-order THD leading to the perception of sound stage. These effects are, however, added by the amp. I can measure that easily on my DG300B using my APx525. Anybody with a DG300B and an audio analyzer will be able to reproduce my measurements. This isn't heresy or speculation. It's hard facts.
Now, whether you like the "DG300B effect" or not is a matter of personal opinion. I'll be just as happy to sell you a DG300B board set as I will be selling you an HP-1 board.

Tom

 

I have very little experience with tube amps I got to say. I am now convince that tube amps bring a specific harmonic content which humans tend to like. I can't tell exactly how this content is shaped though and I would be glad reading some good papers on it. So far I read only superficial papers, they didn't satisfied me.
On the other hand, for solid state amp, dude, I like the cold they bring. I like when my trebles are not even warmth, just tight and there. It makes me think about aluminium for some reasons (this sentence is debatable ^^).
I tried sometime ago PIEGA Premium 5.2 loudpseakers with an AYRE amplifier (the AX-5) and I really enjoyed it. On the other hand when I tried TRIANGLE loudpseakers (which are known to be rather mid-treble oriented) with a tube amp, I found the sound very nice but the warmth was too much for me. I must say that the two systems were not playing in the same room.

 

TL;DR
At the end of the day, it's about the connection for most peeps. I'm sure there are a select few who take solace in knowing, through measurements, that an amplifier is doing little more than turning a small signal in to a bigger signal. I'm sure there are folks who prefer the utmost in precision to even small amounts of euphoria. But I think the majority are seeking that connection: musicality, emotion, inner warmth, whatever it is. And while these are objectively meaningless terms, most folks do know them when they hear them. So we use these terms because they provide a palpable way to compare A to B audibly whereas distortion and noise specs on paper do not. On the objectivist <----> subjectivist continuum, I'm a moderate. I appreciate thorough engineering, and I'm impressed by the technical achievements of the HP-1. But opamp designs are a plenty, and THD @0.000032% vs something larger does not translate directly to a better subjective listening experience. I believe this topic has been the subject of research. Close to three orders of magnitude higher THD in the HP-1 would probably not be detected blindly by our best SBAF ears, except maybe (?) in the midrange. Two orders, definitely not. And really, nobody can point to any THD+N measurement and exclaim "That's the sound I'm looking for!". So I think most popular amps have a significant element of "designed to taste" subjectivity in their schematics. The HP-1 is just not one of them. But because I think most people react positively, for example, to organic harmonics - those akin to what is produced in nature by physical phenomena - it's going to be most common to see the HP-1 judged through that lens. Tom's design goals were different though, and I actually do think it's fair that he refer to other designs as "effect boxes" - at least to a certain degree. Enter tubes and we find, as a random example, that high frequency AC heater current which produces non-audible IMD products to sound subjectively better at soundstage [re]production. It seems right to describe that as an effect because after all, we are deliberately altering the audio signal as opposed to simply using an arguably cleaner DC circuit which would in turn yield less of the desired subjective result. Right? Incidentally the DG300B uses DC. ;-) So perhaps it'd be interesting to learn more about acoustic science as it pertains to audio signal reproduction. How can the objectivists inject a bit of subjectivity in to their thinking, and maybe vice versa. Oh and, BTW, there's probably a parallel between this and, say, one of Luis' headphones. I appreciate a tuning that can conform to a well-accepted ideal, but at the same time I don't always find it the most engaging either. Imagine Yggdrasil -> HP-1 -> Slants. But I digress. I'm interested in hearing the HP-1 cleanliness and any other positive traits it offers, but I'm not expecting those traits to be necessarily comparable to my ZDS or T3. Qualifications should be integral to feedback, and a little variety can be a good thing.....

 

@ultrabike - didn't you take measurements on changstar comparing the HD600 on the Schiit Valhalla and an SS amp? I thought those were very interesting.

In general I also think more measurements should be taken with headphone loads. Resistors are great but don't match the complex impedance of dynamic drivers. Maybe planars.