Are tubes more musical?

[Henry53]

Noise is not musicality. But it is relevant. Tube experts may comment on soft compression pre-generating harmonic distortion and musicality. With tubes the broadband noise may be perceptible vs low noise preamps and speaker amps.

I saw a study of audio tube noise figures. I can't say the study uses the same method of transistor or op amp noise. Maybe other ASR experts can comment.

Tube measurements: https://tavishdesign.com/pages/audio-tube-noise-measurments

The tube measurements are in the µV range. Transistors and op amps are in the nV range, 10-100x less. You can see that in the ASR reviews of tube amp noise and harmonics measurements.

I have also heard that measuring noise in tubes used in microphones is very hard and that Schoeps was the leader in the test equipment for that.

The point is can you actually hear distortion measured in microvolts?

 

[Killingbeans]

Have you ever wondered why most electric guitar players insist on using tube amps? As a player, I can assure you it is precisely because of how nice a slightly or heavily overdriven tube amp sounds. All that rich sustain you hear in a soaring guitar solo is nothing but tube distortion. Solid state amps are notoriously poor at this (leaving aside modeling amps, which are nothing but digital simulations of the real thing). Mind you, a guitar amp is not meant to merely amplify the signal from the guitar. Quite to the contrary, it is a musical instrument in itself and a sound shaper. The goal is not the same as a high fidelity amp, but it is obvious to any guitar player that tubes and solid state do not sound the same.

Cheers,

Antonio

Are you absolutely sure?:

 

[Blumlein 88]

Are you absolutely sure?:

Love this guys various videos.

 

[mhardy6647]

The point is can you actually hear distortion measured in microvolts?

I suspect that one could if one uses high-sensitivity loudspeakers. Some ones of us, here and there in the world, still do, you see?
Thus measuring the ratio of distortion to signal voltage (% or dB) is probably more meaningful in most cases.

Full disclosure, I haven't measured the average output power (much less voltage) of my amplifier when driving the ca. 104-dB SPL/watt @ 1 meter sensitive, Altec driver-loaded loudspeakers I usually listen to, at the (low-ish) level at which I typically listen, but I reckon the absolute values of power and voltage are pretty low.

Fermi estimate: if 1 watt (2.83 VAC into 8 ohms) gives 104 db SPL at 1 meter, and I might typically listen at 80 dB (i.e., 24 dB less, requiring ca. 251-fold less input power), we're talkin' (and, by all means, check my arithimetic!) on the order of 4 mW amplifier power (11 mVAC into 8 ohms). 1% distortion would be 110 uV.
Truth be told, at the sub-mV level, noise is more of a problem (in my own amplifier's case) than harmonic distortion, though.

Are you absolutely sure?:

Dude. The tone comes from the speaker.


Sometimes, the laminations on the transformers may sing a little, too, though -- in full disclosure.

 

[SIY]

@atmasphere Have you downloaded the files that @pkane prepared for you?

Obviously, anyone else is welcome to give those a try.

Bumping this for @atmasphere .

 

[Multicore]

Have you ever wondered why most electric guitar players insist on using tube amps? As a player, I can assure you it is precisely because of how nice a slightly or heavily overdriven tube amp sounds.

I don't think so. I believe players who insist on using a tube amp do so because it's what they are used to. I prefer a modeler, linear amps, and FRFR boxes. Because convenience. I can get the results I want into headphones, into the DAW, into a Zoom meeting, or into the air at any SPL level. In comparison, tube amps are a gigantic pita.

At the same time I do not at all discount the importance of sticking with what you're used to as a musician. But there's nothing intrinsically "more musical" about the tubes, it's all just effects and different tech for implementing these effects.

What's way more interesting here is how "the sound" of electric guitar, i.e. the collective cultural understanding of what electric guitars sound like, or should sound like, has these technical historical contingencies. And if you're in the amp modelling business, you have to figure out the features of the old tech that are important. For example, some modelers simulate different guitar cable lengths, hahaha.

 

[MAB]

Have you ever wondered why most electric guitar players insist on using tube amps? As a player, I can assure you it is precisely because of how nice a slightly or heavily overdriven tube amp sounds. All that rich sustain you hear in a soaring guitar solo is nothing but tube distortion. Solid state amps are notoriously poor at this (leaving aside modeling amps, which are nothing but digital simulations of the real thing). Mind you, a guitar amp is not meant to merely amplify the signal from the guitar. Quite to the contrary, it is a musical instrument in itself and a sound shaper. The goal is not the same as a high fidelity amp, but it is obvious to any guitar player that tubes and solid state do not sound the same.

Cheers,

Antonio

Hi Antonio.
I do wonder.
Except that the sustain has nothing to do with tubes, soaring in a guitar solo or otherwise.
Also, the sound is in the EQ and the distortion, you even answered your own question. Go do a blind test of distortion, try to see if you can actually tell the difference between overdriven tube and a plugin of the same.

Putting aside the inaudibility of the things you mention... The part that really makes no sense is applying these fanciful ideas to reproduction. If I listen to Deep Purple, do you propose I roll out a Leslie speaker to properly reproduce John Lord's Hammond organ? I'm just wondering how far you intend to take this analogy. And hope you see it makes zero sense. These creation cum reproduction analogies don't work. And you misstate what makes an electric guitar sound as it does.

 

[MAB]

Are you absolutely sure?:

Thank you so much. Musicians are no different than audiophiles, no different than anybody else. Oblivious about the source of their preferences.

 

[Bruce Morgen]

Using tube microphones or other non-transparent gear at the recording is perfectly legitimate to create the sound. Reproducing the sound is altogether another issue, and in my view, much less legitimate if the object is accuracy in reproduction. By all means colour the playback to your taste, but then don't claim it's better or that horrible word, musical.

S.

The term "high fidelity" means "great faithfulness" -- IOW, transparency to the intentions of the creative folks who produced the source material. If you want something other than or in addition to that, there are methods to accomplish such goals, but IMO it's best to start with "high fidelity" baseline performance and do additional processing built on that foundation. I do that myself, mostly with a vintage Yamaha processor stack, but my amps and speakers are as "high fidelity" as my modest budget allows.

 

[Bruce Morgen]

What does musical even mean?

Allow me to provide a rather extreme example (which may or may not be relevant here): The sound of the oboist in an orchestra providing an "A" note to which the other musicians can tune their instruments is "musical," replete with an harmonic spectrum that largely defines the timbre of that instrument. The 440 Hz output of a laboratory signal generator has the same fundamental pitch as that oboe note, but lacking any harmonic content it doesn't sound "musical" when amplified and transduced so as to be audible.

 

[Chr1]

The term "high fidelity" means "great faithfulness" -- IOW, transparency to the intentions of the creative folks who produced the source material. If you want something other than or in addition to that, there are methods to accomplish such goals, but IMO it's best to start with "high fidelity" baseline performance and do additional processing built on that foundation. I do that myself, mostly with a vintage Yamaha processor stack, but my amps and speakers are as "high fidelity" as my modest budget allows.

My take too. Baseline: high fidelity.
Then any additional personal subjective enhancements, be it preferred EQ, valves or whatever else should be easily switched in/out.

 

[Cbdb2]

Allow me to provide a rather extreme example (which may or may not be relevant here): The sound of the oboist in an orchestra providing an "A" note to which the other musicians can tune their instruments is "musical," replete with an harmonic spectrum that largely defines the timbre of that instrument. The 440 Hz output of a laboratory signal generator has the same fundamental pitch as that oboe note, but lacking any harmonic content it doesn't sound "musical" when amplified and transduced so as to be audible.

So a flute playing a high note (very close to a sine waveform) is not musical? A tone generator with 10% harmonic distortion is musical? Musical is not a useful term for gear. Its too much about preference. Some people think distortion is more musical, some dont, some people think adding 10db at 60hz is more musical, some people don't. Most people think turning up the volume is more musical, I dont. Its just one of the dumb reviewers terms used to fill a page with subjective BS, like prat.

 

[Bruce Morgen]

So a flute playing a high note (very close to a sine waveform) is not musical? A tone generator with 10% harmonic distortion is musical? Musical is not a useful term for gear. Its too much about preference. Some people think distortion is more musical, some dont, some people think adding 10db at 60hz is more musical, some people don't. Most people think turning up the volume is more musical, I dont. Its just one of the dumb reviewers terms used to fill a page with subjective BS, like prat.

The "waveform" of a flute note may be close to sinusoidal, but it does not lack for harmonic content -- those harmonics are just at a lower level than that of an oboe or a clarinet. A laboratory-grade signal generator does not produce any harmonic distortion to speak of, as contrasted to your hypothetical "tone generator." Of course, musicality is indeed a subjective -- and likely less than useful -- term, but it seems to imply accurately reproduced harmonic content rather than intentional distortion of such content, even if said distortion is experienced as pleasantly euphonic. I reiterate that I'm not confident that my description attempt is relevant here -- it's just something that came to mind.

 

[Cbdb2]

but it seems to imply accurately reproduced harmonic content

Than why do a lot of people say tube or vinyl distortion is more musical?

 

[SIY]

The harmonics are already in the recording.

 

[Blumlein 88]

Than why do a lot of people say tube or vinyl distortion is more musical?

Myth handed down over and over again.

If you wish, you can test this for yourself using Pkane's Distort software. You choose the type and level distortion of any musical file you give it.

It is rather strange how few people use Pkane's software to investigate this for themselves.

 

[majingotan]

The harmonics are already in the recording.

More artificial harmonics are created as well as adding "rectifier sag" to the signal which is perceived as "musical"

 

[Birdy]

I had a really good tube amp, a 2x 30 watt class A and 2 x 70 watt AB monster DC coupled with low impedance output and it sounded like a good Solid state amp with tight basses and clean output, 0,35% THD. The thing is that a real good tube amp sounds like a good solid state amp but i must admit that overall my Aiyima 07 sounds better then that 14 kg monster!

 

[Bruce Morgen]

The thing is that a real good tube amp sounds like a good solid state amp but i must admit that overall my Aiyima 07 sounds better then that 14 kg monster!

I was startled when my first little Aiyima A07 -- I own four now, two of them driving the ambience/surround channels of my 7.1 system -- performed every bit as well as the reviewer-adored Adcom GFA-535 I swore by back in the previous century.

 

[mhardy6647]

The "waveform" of a flute note may be close to sinusoidal, but it does not lack for harmonic content -- those harmonics are just at a lower level than that of an oboe or a clarinet. A laboratory-grade signal generator does not produce any harmonic distortion to speak of, as contrasted to your hypothetical "tone generator." Of course, musicality is indeed a subjective -- and likely less than useful -- term, but it seems to imply accurately reproduced harmonic content rather than intentional distortion of such content, even if said distortion is experienced as pleasantly euphonic. I reiterate that I'm not confident that my description attempt is relevant here -- it's just something that came to mind.

so we need "warmer" signal generators, then, right? Sounds like a job for Bob Carver!

(ahem... sorry... sorry!)