• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

If "Tube Sound" Is a Myth, Why Tubes?

raindance

Major Contributor
Joined
Sep 25, 2019
Messages
1,037
Likes
968
I have repaired and built tube preamps and amps for years, but recently lost interest. Here is my take on tube sound:

1) Poorly designed tube preamp circuits definitely have a sound of their own. A good example is the Conrad Johnson PV12L - the circuit looks like it was designed around 12AX7 tubes and they simply changed to 12AU7 tubes to reduce the gain. As a result, the tubes run at a very low bias point on a very non-linear portion of their transconductance curve. This would color the sound for sure.
2) A lot of tube preamps oscillate when certain tubes are used (such as the one mentioned above) due to the manufacturer either not knowing better or saving a buck by not using grid stopper resistors - this would cause obvious differences in the sound signature of certain tubes.
3) I believe that MOST of the tube sound signature that people are chasing is from the interaction between the relatively high output impedance plus non-linearities of the transformer coupled output with the speakers. In fact, I don't think I've ever seen measurements where the use of the 4 ohm tap (lower impedance) does not improve the performance with 8 ohm speakers. For a lot of speakers, this interaction causes an increase in output of up to a few dB in the area where speaker impedance is highest, often in the "warmth" area of the bass region. I have only played with OTL amps for headphone use and they have similar issues, but in their case the output impedance is even higher, but without the transformer non-linearities. Oddly enough, or not, depending on your stance, feedback improves the situation in transformer-based amps - but then we have manufacturers offering "features" like a triode/ultralinear switch which does not optimize the feedback loop for one of the modes, causing weird frequency response anomolies as a result in one of the modes that seem to be pleasing to certain ears. Funny enough, it is the cost no object zero feedback designs that seem to be most popular, even though they suffer from limited bandwidth and high distortion, especially in the bass.
4) Well designed tube preamp circuits will sound very much like solid state with a slightly higher noise floor. Tube amplifiers, on the other hand, always sound like tube amplifiers due to all the compromises in trying to couple a high impedance device (a tube) to a low impedance transducer (speaker).
5) Cathode followers or buffers, which have a gain of slightly less than 1, do not exhibit "tube sound" as they do not provide the harmonic distortion profile that tube gain stages exhibit. I'd go so far as to say that any significant change they make to the sound is due to impedance matching issues or worn out tubes. Adding them to solid state devices, like DAC's, simply adds hum and noise.

There ya go :)
 

Island_Kenny

Member
Joined
Oct 2, 2019
Messages
40
Likes
25
There will be many recordings played back on any given audio system. The only kinds of recordings that have any chance of "faithful reproduction" are those recorded in real time [everybody playing at once] in a venue with its own acoustic, like Carnegie Hall. But most recordings are a mosaic of different sources with different sonic treatments assembled into a mix where the only real determiner of balances are the ears of a given engineer/producer and the playback gear they used. Every recording will be different, there are no "standards" for this sort of thing. "The Absolute Sound" can only be heard without a recording. If you spend any time with microphones you will note that they are all colored, all leave their footprint on the sound. Like speakers, like phono cartridges, microphones are transducers, unavoidably leaving their own "sound" on the recording. And plenty of music is generated from electronic sources directly fed into the mix. Seriously, all that really counts is what the auditor wants to hear. There is no way one can know if the sound faithfully reproduces the producer/engineer/artist's intent unless one is sitting beside those people when the final mix is being played back.

The performance of record/playback gear, the sorts of measurements Amir makes, can be scientifically calibrated. But the artistic decisions of producers/engineers/artists cannot. And recording is primarily an art.

IMO, "absolute sound" is the biggest myth. Musicality is all about subjectiveness, in the end, it is not the microphone, master tapes, vinyl, CD, codecs, DAC, amp, speaker/headphone, or even your ear, it is the brain that interprets the sound. No two brains hear the same sound, even the same brain does not hear the same sound twice.
 
Last edited:

rwortman

Addicted to Fun and Learning
Forum Donor
Joined
Jan 29, 2019
Messages
740
Likes
683
Tube sound is both actual added distortion and an effect of preconceptions. I have personally heard "tube warmth" while gazing at the dim glow of tube filaments and then realized that the thing I was looking at wasn't actually playing at the time. On the subtopic: The recording was mixed and then mastered in a certain room with certain equipment. That sound is what the recording and mastering engineer wanted. Unless you build a replica of that, you are hearing something else. Putting together a reproduction chain that minimizes distortion is a reasonable goal but in the end this is about enjoying the experience of listening to music and we should all feel free to assemble a system based entirely on liking the way it sounds. If you want to use some EQ to pump up the bass, have at it. If you want some tube created harmonics to sweeten the pie, go ahead. Ditto for record players and anti-deluvian ladder DAC's. What any reader of this site should have learned, though, is that these distortions, however pleasing or not to the ear, don't have to cost a fortune to produce and are not higher fidelity or truer to the event or source.
 

Island_Kenny

Member
Joined
Oct 2, 2019
Messages
40
Likes
25
You are again talking about producing the record, not reproducing it.
Speaking of record, you can easily make a perfect copy of a digital file, it probably does not take too much trouble to get a perfect copy of any tapes. If you are talking about sound, record!=sound
 

rwortman

Addicted to Fun and Learning
Forum Donor
Joined
Jan 29, 2019
Messages
740
Likes
683
Speaking of record, you can easily make a perfect copy of a digital file, it probably does not take too much trouble to get a perfect copy of any tapes. If you are talking about sound, record!=sound!
You can't make a perfect analog copy of an analog source. Every generation adds distortion and noise. This one of the many reasons why the vast majority of recording and sound processing today is done in the digital domain.
 

Island_Kenny

Member
Joined
Oct 2, 2019
Messages
40
Likes
25
You can't make a perfect analog copy of an analog source. Every generation adds distortion and noise. This one of the many reasons why the vast majority of recording and sound processing today is done in the digital domain.

True. My point is that you cannot hear a record without property equipment, reproducing record is easy, objectively reproducing sound is somewhat pointless.
 

Julf

Major Contributor
Forum Donor
Joined
Mar 1, 2016
Messages
3,008
Likes
3,999
Location
Amsterdam, The Netherlands
True. My point is that you cannot hear a record without property equipment, reproducing record is easy, objectively reproducing sound is somewhat pointless.

But what is the point of that point? Hifi is all about (accurately) reproducing recordings, not sounds.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,836
Likes
16,501
Location
Monument, CO
IMO, "absolute sound" is the biggest myth. Musicality is all about subjectiveness, in the end, it is not the microphone, master tapes, vinyl, CD, codecs, DAC, amp, speaker/headphone, or even your ear, it is the brain that interprets the sound. No two brains hear the same sound, even the same brain does not hear the same sound twice.

But research from Toole and others show most of us prefer neutral speakers given a chance to hear and choose. The brain does its own thing, but to me that does not obviate the goal of obtaining an accurate, faithful reproduction of the source. Deviating from that just adds one more variable. I do not know how you or anyone else defines "absolute sound" but I think reproducing at the ears exactly what went in from the source seems a reasonable goal for an audio playback system. Source clearly being in this context the recording one purchases, not the original performance, since we have no control over the process from mic to (e.g.) CD.

IMO - Don
 

Island_Kenny

Member
Joined
Oct 2, 2019
Messages
40
Likes
25
But research from Toole and others show most of us prefer neutral speakers given a chance to hear and choose. The brain does its own thing, but to me that does not obviate the goal of obtaining an accurate, faithful reproduction of the source. Deviating from that just adds one more variable. I do not know how you or anyone else defines "absolute sound" but I think reproducing at the ears exactly what went in from the source seems a reasonable goal for an audio playback system. Source clearly being in this context the recording one purchases, not the original performance, since we have no control over the process from mic to (e.g.) CD.

IMO - Don
IMO, the best reproduction of any musical performance is achieved by Spiros system by Steinways, which records the hammer and pedal movements and replays on other Steinways pianos. This is as close as we can get, more of mechanical engineering than electrical engineering
 

PaulD

Senior Member
Joined
Jul 22, 2018
Messages
453
Likes
1,341
Location
Other
IMO, the best reproduction of any musical performance is achieved by Spiros system by Steinways, which records the hammer and pedal movements and replays on other Steinways pianos. This is as close as you can get.
Does not work with string quartets, vocals, guitars, etc etc. Too limited by far, but I did like it when I've done it with Yamaha Disklaviers for a piano piece.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,836
Likes
16,501
Location
Monument, CO
IMO, the best reproduction of any musical performance is achieved by Spiros system by Steinways, which records the hammer and pedal movements and replays on other Steinways pianos. This is as close as we can get, more of mechanical engineering than electrical engineering

OK, but only for solo piano, and nothing to do with a normal audio system. And again the goal of the system is perfect reproduction faithful to the original performance irrespective of how individual listeners' brains process the sound. Maybe I misunderstood your initial post; your premise seemed to be that accurate reproduction is irrelevant (stronger than you said) since we all have brains that process the sound differently.
 

Island_Kenny

Member
Joined
Oct 2, 2019
Messages
40
Likes
25
Does not work with string quartets, vocals, guitars, etc etc. Too limited by far, but I did like it when I've done it with Yamaha Disklaviers for a piano piece.
vocals will be very impossible, strings and guitar very difficult but not completely impossible with the advance of robotic technology.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,386
Likes
24,752
Location
Alfred, NY
When we had the great folk artist John Statz over for a show, he played a '57 Martin. Now, maybe a robot can imitate his fingering and plucking, but where am I going to get that '57 Martin?

 

Island_Kenny

Member
Joined
Oct 2, 2019
Messages
40
Likes
25
OK, but only for solo piano, and nothing to do with a normal audio system. And again the goal of the system is perfect reproduction faithful to the original performance irrespective of how individual listeners' brains process the sound. Maybe I misunderstood your initial post; your premise seemed to be that accurate reproduction is irrelevant (stronger than you said) since we all have brains that process the sound differently.

I am not arguing that the accuracy of the audio system is not important, it is, and that is why we come to this website and read Amir's reviews. but the ultimate goal should not be perfect reproducing of a sound recording.

There are two points that I would like to make from the example of the Spirio system and its alikes. One, with Spiro, we can all agree that music will not sound good if played on a piano out of tune, but it probably does not matter at all if played on two well-tuned pianos, each with a distinct sound profile. We can still enjoy the same near-perfect reproduction of a performance. Two, in this setting, the source that is a step further from what ears hear is clearly the instrument. One instrument feeds many ears and microphones, no two ears or microphones heard/recorded the same sound. The same can be said for all performances of acoustic instruments and vocals. And reproducing a secondary source, no matter how high fidelity it is, is still a secondary source, not as glorious and holy as some want to believe.
 

Island_Kenny

Member
Joined
Oct 2, 2019
Messages
40
Likes
25
When we had the great folk artist John Statz over for a show, he played a '57 Martin. Now, maybe a robot can imitate his fingering and plucking, but where am I going to get that '57 Martin?


Being Audiophile is an expensive hobby. :D:) but giving him a 61' Martin, the music coming out of it would've been equally beautiful.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,386
Likes
24,752
Location
Alfred, NY
I heard him play several times before he got that '57, and honestly, the guitar made a difference. What was wonderful was that he got snowed in at our place (this was before we moved to Arizona), and while I was cooking breakfast the next morning, he pulled out that Martin and played me a cover of Peter Mulvey's superb song "Just Before the War."

(fun trivia: the guitar John was playing was the same one Peter was playing in that video)
 

MattHooper

Master Contributor
Forum Donor
Joined
Jan 27, 2019
Messages
7,201
Likes
11,820
I have repaired and built tube preamps and amps for years, but recently lost interest. Here is my take on tube sound:

1) Poorly designed tube preamp circuits definitely have a sound of their own. A good example is the Conrad Johnson PV12L - the circuit looks like it was designed around 12AX7 tubes and they simply changed to 12AU7 tubes to reduce the gain. As a result, the tubes run at a very low bias point on a very non-linear portion of their transconductance curve. This would color the sound for sure.
2) A lot of tube preamps oscillate when certain tubes are used (such as the one mentioned above) due to the manufacturer either not knowing better or saving a buck by not using grid stopper resistors - this would cause obvious differences in the sound signature of certain tubes.
3) I believe that MOST of the tube sound signature that people are chasing is from the interaction between the relatively high output impedance plus non-linearities of the transformer coupled output with the speakers. In fact, I don't think I've ever seen measurements where the use of the 4 ohm tap (lower impedance) does not improve the performance with 8 ohm speakers. For a lot of speakers, this interaction causes an increase in output of up to a few dB in the area where speaker impedance is highest, often in the "warmth" area of the bass region. I have only played with OTL amps for headphone use and they have similar issues, but in their case the output impedance is even higher, but without the transformer non-linearities. Oddly enough, or not, depending on your stance, feedback improves the situation in transformer-based amps - but then we have manufacturers offering "features" like a triode/ultralinear switch which does not optimize the feedback loop for one of the modes, causing weird frequency response anomolies as a result in one of the modes that seem to be pleasing to certain ears. Funny enough, it is the cost no object zero feedback designs that seem to be most popular, even though they suffer from limited bandwidth and high distortion, especially in the bass.
4) Well designed tube preamp circuits will sound very much like solid state with a slightly higher noise floor. Tube amplifiers, on the other hand, always sound like tube amplifiers due to all the compromises in trying to couple a high impedance device (a tube) to a low impedance transducer (speaker).
5) Cathode followers or buffers, which have a gain of slightly less than 1, do not exhibit "tube sound" as they do not provide the harmonic distortion profile that tube gain stages exhibit. I'd go so far as to say that any significant change they make to the sound is due to impedance matching issues or worn out tubes. Adding them to solid state devices, like DAC's, simply adds hum and noise.

There ya go :)

And yet I luv ma darn tube amps!

*skulks away....*
 

Julf

Major Contributor
Forum Donor
Joined
Mar 1, 2016
Messages
3,008
Likes
3,999
Location
Amsterdam, The Netherlands
One instrument feeds many ears and microphones, no two ears or microphones heard/recorded the same sound. The same can be said for all performances of acoustic instruments and vocals. And reproducing a secondary source, no matter how high fidelity it is, is still a secondary source, not as glorious and holy as some want to believe.

So you consider a recording a secondary source? In the last 60 years or so, otsde the classical music world, the recording has become the primary piece of art and performance.
 

Robin L

Master Contributor
Joined
Sep 2, 2019
Messages
5,213
Likes
7,593
Location
1 mile east of Sleater Kinney Rd
OK, but only for solo piano, and nothing to do with a normal audio system. And again the goal of the system is perfect reproduction faithful to the original performance irrespective of how individual listeners' brains process the sound. Maybe I misunderstood your initial post; your premise seemed to be that accurate reproduction is irrelevant (stronger than you said) since we all have brains that process the sound differently.
Even with that "solo piano" there will be differences. By way of example, there's a "recreation" of Glenn Gould's 1955 recording of Bach's Goldberg Variations, using a species of "player piano" that runs off the old audio recording as a trigger. That 'recreation' has a more broadband frequency response, not to mention the elimination of Gould's notorious "hum". But somehow, the end result has the rhythm messed up, that "dancing on air" quality of the original goes out the window. Only X = X.
 

Robin L

Master Contributor
Joined
Sep 2, 2019
Messages
5,213
Likes
7,593
Location
1 mile east of Sleater Kinney Rd
So you consider a recording a secondary source? In the last 60 years or so, otsde the classical music world, the recording has become the primary piece of art and performance.
And one can easily argue that something essential is lost when the base-line for the artistic statement is an auto-tuned bit of over-production.
 
Top Bottom