Tube vs. Solid State

[Brooks]

spitting out hell and fury that will kill

newborn babies and call upon rain and

thunder and the true Gods of Valhalla to

send their wrath up on man.

have you been listening to manowar? ha.

solid state is good for pristine cleans

(ala avi bortnick's tight funk rhythms on the 2 john scofield band cd's),

or really heavy metal rhythms (pantera!!!).

i like tubes for everything inbetween.

like everyone else has said,

tubes respond differently and color the sound in a musical way.

[MCChris]

Ahhh, a Lead 100 MOSFET.

I have an Artist 3203. Same head box and chassis, only with a pair of EL34s in the power section. PI tube only in the preamp, so I guess I do have some partially solid state gear, but you'd never know this thing wasn't all tube just by listening to it.

[silentman]

Ahhh, a Lead 100 MOSFET.

I have an Artist 3203. Same head box and chassis, only with a pair of EL34s in the power section. PI tube only in the preamp, so I guess I do have some partially solid state gear, but you'd never know this thing wasn't all tube just by listening to it.

I used to have one of those 3203 heads also. It was my first 1/2 stack.

[Turdus]

I had a Roland Jazz Chorus, and once turned up past 2 or so, it started to distort in a pleasing kind of way.... A lot of guys here know who Tommy Keene is. I am not sure if he still uses them, but I am pretty sure he used JC120's for awhile. Tommy has always had a good rock tone.

[MCChris]

I used to have one of those 3203 heads also. It was my first 1/2 stack.

Mine too. Cab is long gone (gave it to my old singer to pay for the band house cable TV box that was in his name; I destroyed it for reasons I don't remember) but the head likely will accompany me as I'm slid into the crematorium oven.

[JohnnyB]

You can talk about odd and even order harmonics, about progressive harmonic distortion (in tubes) vs. the distortion wall you hit in SS, and try to explain the differences. But as some people point out, many of the distortion harmonics of tubes can be imitated in SS.

What I believe is fundamentally different, however, is how a tube circuit tracks amplitude changes vs. SS. Tube circuits inherently track the most minute subtleties in dynamics, imparting that "organic" sound some of the posters in this thread were talking about. Typical transistor circuits are not nearly as linear or granular at tracking low level signals and low amplitude changes. This is often where the music lives in expressiveness.

In high end audio, the solid state pre-amp circuit designs jump through all sorts of hoops just to come close to what the good tube preamps do naturally. In fact, at that level the SS preamps are at least as expensive as the tube ones because of what it takes to beat a transistor circuit into doing what a tube circuit does so naturally.

I'm more inclined to talk about how a tube amp "feels" than how it "sounds", vs. solid state. The tube ones by and large naturally track the nuances of your playing dynamics with a higher degree of resolution. The even order harmonics and progressive THD curve, as lovely as they are, are gravy by comparison.

[SteveB]

Tube circuits inherently track the most minute subtleties in dynamics, imparting that "organic" sound some of the posters in this thread were talking about. Typical transistor circuits are not nearly as linear or granular at tracking low level signals and low amplitude changes.

I'm not sure I buy that. Outside of the pro-audio and musician camps, is there any audio equipment out there using tubes because they are more "accurate"?

[JohnnyB]

I'm not sure I buy that.  Outside of the pro-audio and musician camps, is there any audio equipment out there using tubes because they are more "accurate"?

Depends on what you mean by "accurate". Which is more accurate--the one with the best numbers on a spec sheet, or the one that psycho-acoustically affects the listener's or player's emotional response the most like live, acoustical music? Which is more accurate, the one with a lower overall THD number, or the one that is more linear in tracking the subtlest playing nuances?

The standard list of specs was largely formulated in the '60s to push solid state gear over the tube stuff which was being obsolesced by the electronics industry. These specs were formulated, touted, and largely accepted as gospel even though there was no research to correlate the measurements and parameters with the human hearing process.

And yes, high end audio is rife with serious tube designs, most of which measure quite well:

• Lamm Industries
  
• Balanced Audio Technology
  
• Audio Research Corporation
  
• Cary Audio Design
  

... and many, many others. A high end audio dealer that specializes in tube electronics is Upscale Audio.

[SteveB]

I don't disagree that many people find tube amps more pleasing to their ear, but I have a hard time believing that a tube design can beat solid state at "linearity in tracking", ie, if you consider linearity to mean changes in output level corresponding directly to changes in input level. That at least should be measurable. I'm not saying that maybe tubes arn't somehow better, but it seems counter-intuitive. In fact, arn't Sag and Bloom basically describing NON-linear behavior in output verses input?

[JohnnyB]

I don't disagree that many people find tube amps more pleasing to their ear, but I have a hard time believing that a tube design can beat solid state at "linearity in tracking", ie, if you consider linearity to mean changes in output level corresponding directly to changes in input level.  That at least should be measurable.  I'm not saying that maybe tubes arn't somehow better, but it seems counter-intuitive.  In fact, arn't Sag and Bloom basically describing NON-linear behavior in output verses input?

It's my understanding that one difference between most solid state and tube designs is that the distortion level in SS electronics is constant. The impressively low distortion rating for SS electronics applies only when the device is operating at full output, but the same distortion level remains when the output drops. That means that the percentage of distortion is significantly higher at low output levels.

That's what I mean about low level linearity. In tube circuits, I believe the distortion rises and falls with the output, giving a cleaner, more linear tracking of low level signals. SS circuits are capable of good low level tracking, but it takes a more complex design to achieve it.

Of course this difference should be measurable, but I don't know if anyone has formulated a measurement and spec for it. Many of the currently popular audio specs were formulated or manipulated to furnish evidence that transistors were superior to tubes, regardless of the listeners' psycho-acoustic response.

A great example of this is the THD spec, which stands for total harmonic distortion. Being "Total," all forms of harmonic distortion are lumped into one spec--low level, high level, even order, and odd order. Research in the last 10-15 years has suggested that the brain reacts more negatively to some kinds of harmonic distortion, and barely notices others. For example, one would probably notice odd order harmonic distortion before even order, and distortion in quiet passages before noticing it in loud ones, but the THD spec lumps them all together and fails to recognize the differences. In doing so it creates the perception that SS is superior.

Another thing about THD, in SS designs it's easy to lower the THD spec to unmeasurably low levels. All you have to do is increase the amount of negative feedback until you get the number you want. But a funny thing happened on the way to the .00001% THD spec: the resulting sound was brittle, sterile, and most of all, musically uninvolving. One MOSFET SS amp I have circumvented this dilemma by nesting the feedback loops so there was a lower number of overall feedback loops which functionally reduced the distortion while keeping a very liquid and musical signal intact.

Another example of spec manipulation is that class B SS amps have what's called "notch" distortion, where the two mirror images of the signal don't match up accurately. In the '70s I remember you could have an A/B amp with very low measurable distortion, but would sound edgy because it had a high level of notch distortion that didn't get measured. The notch had a lot of amplitude but was of such short duration it was hard to detect by standard measurements, but listeners could detect it. It took high-speed oscilloscope photos to capture it.

Both tubes and transistors have their advantages when it comes to audio reproduction. Transistors are faster, have wider bandwidth, etc. Tubes have more sonic continuity and the natural ability to track low level signals.

[Guest Mike Lee]

I'll never understand why people try to go to such lengths to find technical reasons why something that sounds better is inherently superior on other ways, especially when such claims are based on flawed reasoning.

Let's put this in simple terms: Well-designed and manufactured solid state circuits are more linear and more accurate than tube circuits. They do not have a constant distortion level regardless of amplitude. They are pretty linear up to the rated output and then clip hard.

Negative feedback is a way of increasing linearity and reducing distortion by electronically comparing the output wave with the input wave and correcting differences. This is how most highly-accurate control systems work, especially motion control systems used in CNC's and robots.

Tube circuits have fundamental non-linearites by nature. They always have some distortion, which gradually increases with output level, but there is no hard limit as with a solid state device. So the transition to distortion is less distinct. Many tube circuits lack negative feedback, which can actually allow the output dynamics to swing more than the input dynamics. The amplifier essentially overreacts to the input and actually increases the dynamic range. Couple that with the natural compression and distortion of a tube circuit (as well as voltage sag) and the effects counteract each other with some very interesting waveforms.

Bottom line - tube circuits sound better for guitar because the non-linearities and distortion they impart on an audio signal sound better to our ears than an accurate, linear, undistorted reproduction of the same sound. But solid state in most cases will be more linear and accurate when comparing input signal to output signal.

[Brownsound]

Again...you have to add the "Feel Factor"....tubes feel very different to me when playing.

[Imiss85]

I've found a happy medium - Pod XT Live with an atomic reactor. SS computer technology, ton of amp sims, and the atomic is tube driven so it adds the sag, and really moves the air with a ton of bottom end. No, it's not a JCM 800, or a Fender Twin, but I can get it close enough that 99% of an audience cannot tell ( as if they knew the difference anyway). And, who in an audience is going to say... Play them side be side to convince me.