Wow. I know this is an old thread, and I'm not trying to reopen old wounds, but there are a lot of misunderstandings and misconceptions flying around here and it seems worthwhile to set the record straight, if not for this discussion that is long over, then at least for those people who happen across this thread in the future. And not that it's relevant to the OP at this point, but I do have information specific to PrimaLuna, which was the original subject of the thread, though that fact seems to have gotten lost along the way.
"Musta got lost, musta got lost, musta got lost, somewhere down the line..."I'm trying to avoid an argument - but you keep begging the question.... If they sound different, and the difference
ISN'T distortion, then what
IS it?
"Sound stage" and "imaging" aren't actually separate things - they are simply ways in which your brain interprets the sounds it hears. And so both could be "removed" from a recording if something alters the cues in that recording your brain needs to "hear" them. Likewise, both could be added to a recording if some device or equipment adds phony versions of the cues your brain is expecting to find. (Much as I can sit down with a pen and paper and draw a pair of pictures that your brain will "see" as a single 3D image... if I do it correctly.) And both could be altered if some device alters cues that are there in the recording.
Bear in mind that there really is no such thing as magic... and nothing exists which inherently cannot be measured (only things that so far we have failed to measure).
If two things both sound perfect, then they
MUST sound the same - by definition.
And, if you play the same recording on two systems, and the results sound different, then one or both must technically be incorrect.
(Although the converse isn't necessarily true; two things could sound the same because they're wrong in exactly the same way.)
Therefore, by logic....
If tube equipment and solid state equipment sound different, then either....
1) tube equipment adds some sort of audible distortion that solid state equipment does not
2) solid state equipment adds some sort of audible distortion that tube equipment does not
3) both tube and solid state equipment add different forms of audible distortion that sound different
Those are the ONLY possibilities that logic allows.
I'll admit to not knowing what "cues" are in this context, unless this is shorthand for the differences in a recording between the left and right channels, which results, not just in a stereo recording, but in the ability for a recording to portray an aural 3D image, with specific placement of musicians from left to right, front to back, and top to bottom.
This, however, bothers me a bit:
"If two things both sound perfect, then they MUST sound the same - by definition.Although this statement seems logical enough on its surface, there are some major problems. First and foremost is the word "perfect". I'm not sure there is any such thing, any physical object on Earth that is perfect. I think "Perfect" is an idea, something we can imagine, but has no material counterpart. The most perfect circle (not the band) will reveal flaws upon microscopic inspection. It may be "the most perfect" compared to others, but isn't actually perfect. I think "Perfect" belongs to the realm of the divine, and not to us mere humans.
The other problem is that the sentence implies that there is one sound which is correct and all deviations from that ideal are incorrect - the more the deviation, the more incorrect. Again, it makes sense on its face, but I do not believe that it is a reality. All components in a sound system influence the sound, your speakers, turntable, CD player, amp, preamp, cables, etc. So the arguement would be, "the sound that most accurately represents the original recording". Well, who knows what the heck the original recording is "supposed" to sound like? At best, the people who were at the concert or in the recording studio, but that' about it. Go find me someone who was at Sinatra's famous recording at the Sands (but be nice to blue-haired people, and dead people).
Then there's the problem with those very people present at the time of the recording. Here's a news flash: everyone hears things differently! There are people who are literally tone-deaf, and there are those with perfect pitch, and there's everyone in-between. Some people have reduced hearing in the mid-range frequencies. It's probable that they will like a system that to me would sound very bright and harsh in the mid-range. At one point, I had a hearing deficit in one of my ears. I had to sit a bit to the right of center to hear a vocalist placed solidly in the middle of the soundstage. The variations are endless, and what's "perfect" for me might not be "perfect" for you.
So then people will say, "the closest thing to a straight wire with gain". Well, this comparison is about as useful "as tits on a boar hog", as they say down south. There is no such thing as a straight wire with gain. There's not even a
bent wire with gain. And if that is the analogy, tube amps, generally speaking, are much simpler circuits than solid state designs. Furthermore, I can
see the inside of a glass valve, but I have no idea what's happening inside a particular transistor.
Of the three "if --> then" logic statements quoted above, the third one, "both tube and solid state equipment add different forms of audible distortion that sound different", is the closest. But it's not
distortion that's being added. A more accurate term would be "sonic character". My speakers have their own sonic character that they impart on the sound, but they're not distorting. I have to play them really, really loud before they start to distort. Remember, even the worst examples of both formats will play at least play at moderate levels without distortion, though at these levels it is difficult to discern the dirrerences between the two for all except those with the most highly trained ears (such as reviewers), but the sonic characteristics of each still remains.
Overall, however, I think most will agree that tube amps/preamps sound different than solid state amps - at least as a general statement. Though compared to vintage tube gear, as good quality modern tube equipment has gotten far better at reproducing the entire audio spectrum, producing transients more quickly, with improved micro-dynamics, and better bass control, the differences between tubes and solid state can be harder to discern, especially for those not practiced in very critical listening (such as reviewers).
This is the critical question:
If they sound different, and the difference ISN'T distortion, then what IS it?There are definitely still differences. The question is why? The most direct cause for the differences in sound is NOT distortion. The cause lies in the devices used, i.e. tubes vs transistors. Each device has its own sonic signature. Tubes do not impart distortion, and transistors are not inherently more "accurate". This is a common misconception and is simply wrong. For the former to be true would be to imply that a tube amplifier is constantly producing distortion at all volume and output levels. For the latter to be true would be to imply that transistors are a near-perfect device, which they certainly are not.
If the above is correct, why do tube amps and solid state amps sound different? In many cases very different. The answer; harmonic distortion and possibly the significant variation in frequency response due to the high output impedance of tube amps.
As many like to say here, the proof is in the listening. And you can certainly hear it IMHO.
Russ
No. Just no. This is not why tube amps and solid state amps sound different. What you hear when you hear differences between tube amps and solid state amps is the difference between the valves and the transistors themselves. There are audible ramifications for the circuit design of each, of course, but to keep it simple, the most direct difference lies between these two devices. That's why you can change the sound of a tube amp by changing the tubes. Tubes have their own unique sonic signature. Again, it's not distortion, its just a sonic signature. It's not "correct" or "incorrect". Likewise, transistors have their own sonic signature. Whereas tubes tend to sound more liquid and smooth, transistors tend to sound more dry and hard. That sound is not "correct" or "incorrect". It is not distortion. It is simply the sound of transistors. That's why there was a big push in the early 90's toward MOSFET transistors, because audio engineers thought that they sounded "better". Not more "accurate". Just better. The idea that tube sound is somehow "colored" is fair enough, but unfortunately the same is true of transistors. Transistors impart their sonic signature on the audio signal just as much as valves do. A person might prefer the sound of transistors over the sound of tubes, but that doesn't make them better or make valves worse. It's just what you like. The argument that solid state is better than tubes is like arguing that Tuesday is better than Wednesday.
There are a number of people here that are claiming that the difference in sound between tubes and solid state is due to distortion. First of all, it should be understood that at normal listening levels, neither format is generally operating in a range that produces audible distortion, assuming we are talking about reasonable quality samples of each. However, when these formats are pushed to the point of audible distortion, both designs impart distortion into the audio signal. Neither format is free from blame. Also, the way in which distortion is handled by each format is quite different, and in this difference there are definite advantages to tubes vs solid state designs. When distortion does come into play, at the operating extremes of an amp, what is unique to both is the TYPE of distortion and the AMOUNT.
Before we talk about distortion itself, we should first understand what distortion is comprised of so we can understand and even predict what it will do to the sound. In audio, distortion enters the audible domain in the form of
Harmonics. Harmonics are a special type of overtone. Harmonics are a resonant tone that is a whole-number multiple of the original note being played - the
Fundamental. Let's use as an example a 60Hz tone. If I recall correctly (feel free to correct me, it's been years since my music theory classes), the Fundamental is also considered the 1st Harmonic. 60Hz divided by 60 = 1, therefore the 1st Harmonic. The 2nd Harmonic would be at 120Hz, though at a lower volume (amplitude) than the Fundamental. The 3rd Harmonic would occur at 180Hz, though at a lower amplitude than the 2nd.
Most, if not all instruments produce harmonics. If there is a bell tuned to ring at the specific frequency of 1000Hz, it will naturally create a descending slope of harmonics like those explained above. So will the string of a piano, or an oboe or a human voice. These harmonics are natural to our ears. In particular, the dominant harmonic in nature is the
SECOND Harmonic. This specific harmonic is what resonates in unison with virtually ALL music and instruments that exist in nature, it's what our ears
expect to hear. If a bell rang at 1000Hz
without the resonant 2nd Harmonic, it would sound
WEIRD. So the presence of the 2nd Harmonic is very pleasing to our ears as it corresponds to our natural experience of music. When there are resonance's that are NOT comprised primarily of the 2nd Harmonic, the resulting sound is what we describe as
Dissonance.
With tubes, due to the nature of the valves themselves, when an amplifier is pushed to distortion, the TYPE of distortion created is EVEN ORDER HARMONIC DISTORTION and is comprised primarily of the 2nd Harmonic. Higher order harmonics generally are much lower in volume/amplitude compared to the 2nd Harmonic, and are often inaudible - even when the amp is producing audible 2nd Harmonic distortion. Since the presence of the 2nd Harmonic sounds very natural to us already, when a tube amp distorts, it usually still sounds pretty good. Where you likely will hear the result of distortion is in a compression of the soundstage, a reduction in dynamics and a lack of the amp's ability to play any louder. Of course, if you keep pushing the amp further into distortion (keep cranking the volume), the presence of higher order harmonics will become audible and the sound quality will be degraded.
Solid state amplification works entirely differently. While a solid state amp may have a THD spec. of 0.001% compared to a tube amp's spec. of 0.1%, a solid state amp emphasizes ODD ORDER HARMONICS. So instead of even intervals being introduced, odd intervals are imparted on the signal, which is NOT how music behaves in nature. Also, when a solid state amp is pushed to distortion, although the distortion spec. is very, very low, the Odd Order Harmonics are NOT reduced in amplitude as you go up the harmonic ladder. So while a tube amp introduces the 2nd Harmonic and higher order harmonics are reduced in amplitude, solid state amps introduce Odd Order Harmonics - all at the same amplitude! So the 5th Harmonic and the 7th Harmonic are as present in the audio signal as the 3rd, and our ears find that particularly unpleasant, which is why when solid state amps are pushed to distortion they immediately sound
horrible. Anyone with experience with solid state amplification, whether you're a fan or not, knows that once a solid state amp is pushed to clipping, it quickly sounds like a mush of noise - dissonance. It's
unlistenable and you run to turn the volume down just to make the awful noise stop.
Granted, it can be comparatively more expensive to build a tube amp that reproduces the entire audio bandwidth with good bass control as opposed to solid state. Aside from a good circuit topology, an expensive to manufacture, quality power transformer is an absolute must for powerful bass and good frequency response. Transformers are not devices that lend themselves to mass production - many of the best are wound by hand.
However, in the world of high end audio, anyone looking at a significant investment in amplification is already in the financial territory of quality tube amps, and therefore should not rule them out. And if your speakers just work better with solid state watts, which is not uncommon, then a good tube preamp is a fantastic way to go. You can achieve a majority of the benefits of tubes with just the preamp and still have the control required for your speakers with a solid state amp. Many audiophiles find this to be the ultimate combination.
Bill- we're not arguing whether or not tube amps sound good. In some cases I think they do and Keith would probably agree. The point I brought up early on is that you're not listening to an accurate reproduction of the source media due to the high levels of distortion generated by the tube amp, not whether this sounds good or bad. That's the point I was making and I don't see how anyone can argue with that as tube amps do sound different than solid state amps.
I started this hobby in the early 60s; had plenty of tube amps but none since the late 80s.
Russ
This is a common misconception. Lower distortion specs does not guarantee more "accurate" - not by a long shot. Transistors are not inherently more "accurate" than valves. That's a fallacy. A person might
prefer the sound of transistors over the sound of tubes, but again, that doesn't make transistors more "accurate", it just means that's what you like. Both formats influence the sound of a recording. It is an inescapable fact of audio engineering. In the 70's we were sold a bill of goods based on the new transistorized technology that a device's specifications were the final arbiter of good sound. Both American and Japanese manufacturers competed in a game of one-upmanship, particularly in the areas of channel separation, signal-to-noise ratio, and, of course, distortion. But a transistor has its own influence over the sound just as glass bottles do. And having a lower distortion spec. does not guarantee a more faithful reproduction of an instrument's tone (which I think is what's meant by "accurate", though "accurate" is an arbitrary and arguable word), or a more accurate placement of musicians within the soundstage, or better micro or macro dynamics, or any of the other factors we use by which to judge a piece of equipment's performance.
As discussed above, the idea of "best" or "accurate" or "correct" or "perfect" are arbitrary assessments. Unless you were physically present in the recording studio or at a live performance, who is to say what a recording is "supposed" to sound like? Even if you were there, how good is your aural memory? Really, the only thing we can directly compare is an instrument's tone. Here, again, I think a survey of professional reviewers would side in favor of a good quality tube product - most professional reviewers consider tube audio to produce the most "realistic" tone when compared to the actual instruments themselves, though certainly there are exceptions. Also, you have to consider that people hear the same thing differently. As a point of comparison, I've had a blue chair in my home for about 20 years. It's really an aqua-blue, but to me it's always been more blue than anything else. Predictably, my wife calls it the green chair. So does her offspring. For her and her bloodline, the green that helps make aqua is more dominant than the blue. Just as two people can see the exact same thing differently, two people can hear the exact same thing differently. So "accurate" and "correct" are highly subjective, and no one is going to win that fight. Interestingly, I've found that there is a whole sea of audiophiles out there who feel that the tonal accuracy of audio is of primary importance and won't even begin to consider any other aspects of a piece of equipment if this aspect is not to their standards. Still, ever since the 70's, the bias and misconceptions about specifications that arose from that era largely still prevail today.
But if you pause to consider it, are good specs. really the final word on sound quality? How many of you have heard an amplifier or receiver with a jaw-dropping low distortion spec. that sounded absolutely terrible? For me, too many to count. And how many 1000 watt per channel amps or receivers can you find on Amazon or Alibaba - none of which will set you back for more than two or three Benjamins? How utterly dreadful do you think they sound? I've never bought one to find out, but I doubt anyone here would trade what they currently own for one. Or how many of you have a friend, like I do, who has come back from one of the big box stores to show you the "incredible" stereo he got, with awesome specs? If you're a good friend, you're too nice to tell him how terrible it is. But if you did try to explain how poor the equipment sounds, the first thing he would say in his own defense is that it has "awesome" specs! And technically, it probably does. So what do you say then?
So specifications can serve as a guide, but they should not be interpreted as the absolute ends when it comes to either sound or performance. I've got a twenty five year old 45wpc Rotel receiver that I know will blow away any of those $200, 1000wpc receivers on Amazon. And there are many, many tube amps with a higher distortion spec that will sound far better than many SS amps with incredibly low distortion.
Regarding the higher output impedance of tube amps, which is a considerable difference from solid state designs, this is something to be aware of, but not wary of. In short, just as with a solid state amp, you should take into consideration whether an amp is an appropriate match with your associated equipment, especially your speakers. With regards to PrimaLuna's newer Dialogue Premium HP integrated and power amps, according to John Atkinson's measurements:
"...these impedances are significantly lower than with other PrimaLuna amplifiers I have measured, and the modulation of the amplifier's frequency response by the interaction of its output impedance with that of the loudspeaker was relatively mild. It ranged from ±0.8dB from the 4 ohm tap in triode mode (fig.1, gray trace) to ±1.3dB from the 8 ohm tap in ultralinear mode. Note the peak between 30 and 50kHz in fig.1. This peak was at its highest from both taps into 8 ohms and higher impedances, but disappeared when the load impedance was well below the nominal transformer-tap value. But with the tap matched to the load, this peak was associated with a significant amount of overshoot on a 1kHz square wave (fig.2), though a 10kHz square wave revealed that the consequent ringing was critically damped (fig.3), the amplifier maintaining its stability. The flat tops and bottoms of these square waves correlate with the amplifier's extended low-frequency response".Certainly it is easier and therefore more common (pretty much ubiquitous) for solid state amps to provide full power across the entire audible spectrum (20Hz - 20kHz), but as seen above, with careful investigation it is possible to find tube products that do the same. To do so at higher power ratings will definitely lead you into higher price tiers, though the PrimaLuna (Disclaimer: I have no association or interest with PrimaLuna) Dialogue HP line is relatively quite a bargain, especially when you consider that even the integrated version has more than enough power reserves for at least 90% of the speakers in the industry, including many of the ribbon and electrostatic models. What's more, since nearly all of PrimaLuna's amps and integrateds can be strapped to mono, if a customer decides to invest in a top model of electrostatics, he or she can simply add a second unit instead of having to start from scratch due to the new amplifier demands necessitated by a new, demanding speaker load. As a result, owners have more than enough power - with the Dialogue Premium HP, up to 192 tube watts per channel with KT150 tubes - for less than 4K per amp. Of course KT150s are expensive, but with the supplied EL34s you're still at 148wpc with two mono amps. This means that your initial investment (if you start with a single amp) is not only flexible, it's also rather future-proof.
Just to put things into perspective because there is so much said about "tube amps" as a generic thing, and of course there are many outstanding tube amps from many manufacturers, but the OP was asking specifically about PrimaLuna, so here is a customer's review of his newly acquired PrimaLuna Dialogue Premium HP amp:
"After years of Audio musical chairs my search is over, my PL Dialogue Premium HP Amplifier replaced my McIntosh MC 302. In terms of performance: Bass Tone, Midrange Liquidity and High Delicacy there is no comparison, the 3D imaging and soundstage bursts from the blackest of backgrounds, Bravo Primaluna! My system consists of Revel F208 Speakers, M2Tech Young DSD DAC w/ M2Tech power Supply, MIT cabling and a dedicated Solid State mini computer. Out of the box COLD it crushed my McIntosh MC 302 in all performance parameters. What really surprised me was the Bass, it had beautiful tone, pitch and depth, I heard bass in the lower registers that I thought my Revels where incapable of WOW! I was so impressed that I contacted Kevin and Purchased the PL Dialogue Preamp"As a point of comparison, the above customer's McIntosh MC302 amplifier is a $5,500, 100 pound beast rated at 300wpc - continuous, not peak! It also has a frequency response of 20Hz - 20kHz (10Hz - 1000kHz, -3db), a signal-to-noise ratio of 120db, and a distortion rating of 0.0005% at full power. It also is a high current design, providing 60 amperes of current, which is impressive. Current (amps) is what controls the motion of the drivers in your speakers. Obviously, more current is better, allowing tighter control over the drivers' movements which results in faster transients, better resolution of complex passages and tighter bass control with faster recovery. Tube amps are high current by design. Solid state amps, not so much. It depends on the manufacturer. That's why it's possible for these ridiculously cheap 1000 watt amps and receivers to exist on Amazon, Walmart, etc. They have a high wattage rating, but next to no current. They sound awful.
Now we've seen the specs of the customer's massive and well-built MC302, but what about the specs of the PrimaLuna that "crushed" it? The piece he bought, the Dialogue Premium HP amp (not the integrated) costs less than the Mac boat anchor, at $3,900. And compared to the Mac, it has a much punier 70wpc in stock configuration, though you can bring that up to as much as 96wpc with the aforementioned KT150 tubes. The (usable) frequency response is also 20Hz - 20kHz. The signal-to-noise ratio is just a bit above 75db. Distortion (THD), as per the topic of this thread, is far different from the solid state MC302. Instead of 0.0005%, the PrimaLuna has, at normal listening levels, a THD spec. of between 0.1 and 0.2%, with a full 3% THD at clipping (full output power of 70 watts). Although this distortion is inaudible below clipping at these measurements, what distortion is present is predictably the second harmonic, which is why when the amp actually crosses over into clipping (at or above the full 70wpc), it is difficult to hear evidence of any clipping, because the higher amounts of distortion present when the amp is clipping is euphonic and musical, unlike with a solid state amp that is clipping.
So, as several before me have said, what is actually audible when we are considering the specifications of an amplifier under consideration? Did the owner suffer from the 230wpc drop in power? Did he hear the less impressive signal-to-noise ratio spec. in the PrimaLuna? Was the 70wpc not sufficient for his 88 - 89db sensitivity speakers? Was the amplifier incapable of handling the speaker's natural impedance swings from 3.6 - 8 Ohms? Was it incapable of getting good bass from the Revel's two 8" bass drivers? Did he hear a muddiness or lack of cohesion due to the less impressive THD specs.?
No. On all fronts. This buyer went from a more expensive, extremely high-powered, high current solid state amp from one of the most renown manufacturers on the planet. This amp has all of the best specs: high current, crazy low distortion, ultra-quiet signal-to-noise ratio. You name it, the MC302 dances circles around the Dialogue Premium HP when it comes to specifications. But when it comes to ACTUAL performance, the buyer (as well as every reviewer who has auditioned this model) noted improved imaging, with a holographic, reach-out-and-touchable soundstage. He experienced better tonality - instruments that sound like they are supposed to sound. Mid-range liquidity, no harsh, etched and fatiguing mid-range, which is, hands-down the most important audio band. He also heard improved high-frequency delicacy, which can be critical for the realistic reproduction of a recording's acoustic space. (As a test, play a recording of someone shaking a set of keys, a baby laughing or an audience clapping. If your system isn't reproducing the mids and highs realistically, with a natural tone and delicate, relaxed and spacious highs, then these test sounds will sound artificial, canned and compressed.) But beyond all of those great improvements, compared to the 300wpc monster, the customer heard
IMPROVED BASS! His speakers suddenly went deeper than he thought they could go, with a richer, fuller, and better tone and depth. How often can you slash your budget by sixteen-hundred bucks ($5,500 vs $3,900) and actually expect EVERY ASPECT of the thing you buy to
IMPROVE?! Part of that, obviously, is the overall move from solid state to tubes. Especially in the areas of 3D imaging, tone, smooth midrange and spacious highs. The improved bass is a little unexpected, for sure.
When considering tubes, if you are put off by the scary 3% number (only once the amplifier reaches clipping at full power), you have to ask yourself, "how often am I going to listen to my music at
full output power? Even if you are the type of listener who frequently drives your amp to clipping (aside from considering a bigger amp or bridging two amps to mono), you should also be aware that tube amps have what is called "soft clipping". This means that even when the amp is over-driven, it clips in a euphonic way - you likely won't even notice it. Where you hear it is in a compression of the imaging and a reduction in the dynamics, and it just won't get any louder, but it doesn't sound
BAD. Unfortunately with solid state amps, clipping sounds
AWFUL. There is no soft clipping with solid state amps. When they clip, they clip HARD. Of course the dynamics and soundstage become compressed, but also the music sounds HARSH and unpleasant. Everything begins to sound like mush, just noise, full of dissonance. There is zero chance you won't notice when your solid state amp is driven to clipping. So specs should be considered with the knowledge of what these things mean relative to the particular component being considered.
I agree with you absolutely here... right up to the point where conclusions become assumptions.
You are absolutely correct that, if a lot of feedback is used to reduce certain types of distortion, the result
MAY be truly awful sound.
However, the reason isn't at all mysterious (even though it might have been in 1976.) The reason is that, by using heavy feedback to reduce distortion
IN DESIGNS WITH AN INSUFFICIENT SLEW RATE, you reduce the amount of lower order harmonics, but at the cost of increasing higher order harmonics. And, because higher order harmonics are more discordant, it's a bad tradeoff - because even smaller amounts of higher order harmonics sound worse - while relatively large quantities of lower order harmonics don't sound especially annoying - and may even sound "good" under some circumstances.
Therefore, you're entirely right - you can't always tell what an amplifier will sound like by looking at the THD (
TOTAL harmonic distortion) spec. In order to predict how an amplifier will sound based on distortion measurements you need to look at the actual spectrum showing the relative levels of the first several harmonics. (And, since both modern solid state circuits and tube circuits have "typical" distortion spectra, you can get some vague idea of how a unit will sound by knowing whether it's tube or solid state. However, it isn't conclusive, and not all units of either type will necessarily be "typical" - so you can still in fact tell more accurately by looking at the measurements - if you know how to interpret them.)
The other thing is that there is still in fact a context. For example, when talking about THD, a tube amplifier with 5% THD will almost certainly sound better than a solid state amplifier with 5% THD (because,
TYPICALLY, the tube amp will be producing mostly second harmonic, while the solid state amp will be producing a significant amount of third harmonic). However, if you had a tube amplifier and a solid state amplifier, each of which was producing 0.01% THD, then you wouldn't notice any difference at all - because 0.01% THD isn't audible - regardless of which harmonics it consists of. But, in this case, the context is also that many solid state amplifiers can achieve 0.01% THD, but I have never yet heard of a tube amplifier that was that low in distortion.
This is an excellent post. Yes, indeed, negative feedback can be both good or bad depending on how it is used. What's interesting is that most people don't know that most solid state amps use
extreme amounts of negative feedback to tame their transistor circuits. Most tube circuits also apply negative feedback to reduce distortion, but at the expense of frequency response. It is a delicate balance, and these kinds of give-and-take trade-offs exist throughout audio engineering. What's interesting is there are some tube amps that actually have
adjustable levels of feedback. Some amps have a two or three position switch, and some amps allow you to apply no negative feedback at all! I am not aware of any solid state amp without negative feedback, or even adjustable feedback.
More importantly, I can show KeithL, and everyone else,
EXACTLY what he hasn't seen - a tube amp with an
actual distortion spec. of 0.01% !! It is, indeed, a rare bird, but the genuine article exists. Of course, I am talking about the legendary
"Presence" amp by Henry Wolcott. This may actually be the
BEST amp ever built. Ever. Bar none. Tube, solid state, police state, you name it. If I had the resources, I would track down and buy a pair (unfortunately Henry is now too old to continue his company), then I would track down and buy another pair and keep them in my closet. Whatever people are paying for Mac 275s these days, they'll be paying ten or a hundred times as much for these amps fifty years from now. To read the details about these incredible amps, and their ground-breaking design, check out this link:
www.enjoythemusic.com/magazine/equipment/0700/wolcottaudio.htmThere are a few other reviews here and there, but the above link is the most comprehensive and shows you that yes, a tube amp can have distortion specs. like those of a solid state amp. Incredible.
But with all of this back and forth between tube and solid state, I think the most interesting part of the above debate are the comments concerning measurements:
"Bear in mind that there really is no such thing as magic... and nothing exists which inherently cannot be measured (only things that so far we have failed to measure)."Logically we could conclude that all aspects of audio should be able to be measured, and those measurements should then be capable of informing us about, not just the performance, but the qualities of the sound itself. This is a fascinating concept that I find very seductive. On paper (no pun intended) it makes perfect sense.
But what about those things we can't measure? And, what about those things that, although the measurements predict one thing, upon listening, our ears tell us quite the opposite? This is a debate genuinely worth having. Solid state guys have been bashing tubes for decades, and they're never going to win as they're arguing opinion, not fact (it's okay to like one more than the other, but just because you like it more, don't try to claim it's
better). What is amusing is that slowly, one by one, most solid state enthusiasts eventually cross over if it's within their budget. What's unfortunate is the time lost before the switch happens. It took me thirty years, and I regret every lost year. But this other question is far more fascinating...
Why, for example, can two resistors of equal value sound completely different in the signal path? One is a silver foil unit of the highest pedigree, the other a three-cent part from Radio Shack, but they MEASURE the same? What's that about? Even if we understand
why they sound differently, can you
measure that difference? Or from the other side of the coin, we know that the smooth, natural and life-like mid-range on a tube amp is a result of the way that tubes transfer electrons. But how do you MEASURE a more liquid sound compared to, say, the drier, more sterile sound of transistors? A graph of the harmonics "predicts" what kind of sound profile an amp will have, but that isn't a measurement of the sound itself. And a measurement of the sound itself will only show you frequency response and amplitude, but it doesn't measure "liquid" vs "dry".
It's safe to say that within our current state of technology that there are indeed things that cannot be measured, but can be heard, just as there are things that have identical measurements, yet sound differently. Perhaps one day we will find ways to measure these differences, but in the meantime, maybe this is where some of the "magic" lives in audio? I'd like to think so.
As a final thought, there is this little nugget: for centuries no one really knew why Stradivarius violins perform and sound so much better than any other violins of equal, or even better craftsmanship. Virtuosos claim that the instrument has no limits; whatever you are capable of offering, the violin will oblige, effortlessly, and make everything you feed it sing with endless musicality. Finally, some scientist(s)? decided to find out why. A small sample of wood was taken from an insubstantial part of the body of one of these great violins. After considerable analysis, it was discovered that the logs used as the raw materials at the time these instruments were created were uniquely affected by the river water while these logs were being transported, and the molecules of the wood had specific, unique properties as a result of the wood's exposure to this water. They also, armed with this new information, knew they could reproduce this affect on modern wood.
After creating wood with the same molecular peculiarities that were observed in an original Stradivarius, and having a master violin maker craft a replica of an original, the instrument was offered to a virtuoso violinist (I've since forgotten who, but I'll try to look it up), the musician stepped away from a session with this lab-inspired Franken-varius with goose bumps, stating that the replica offered everything that his original performance Stradivarius did, and possibly then some.
Okay, I did look it up. It took me a while. This story took me back to a news story I first heard in 1984. Now, if I could only remember what I had for breakfast... Anyway, here is a quote from a reprint of the story as found on DiscoverMagazine.com,
"Stradivari's Secret":
"[Joseph] Nagyvary's historical research, however, yielded a crucial clue. According to an account written by the aristocratic patron of Joannes Baptista Guadagnini, one of the last of the famed Cremonese violin makers, it was important to use wood that had been dry-aged, with no extra treatment. Nagyvary didn't buy it. 'I think this may have been a deliberate deception,' he says, 'in order to keep anyone from copying the great masters and lowering the value of existing instruments.' He learned that wood supplies were tightly controlled at the time by government authorities in Venice: 'If you just went out and cut wood from the forests, you could be thrown in jail.' Instead, authorized woodcutters felled trees in the highlands and dumped logs into rivers, where they were carried downstream to the capital. 'The Venetian navy got the best wood for building its ships,' Nagyvary says. 'Only after bureaucrats had taken inventory and assessed taxes could wood merchants buy their supplies—and at this point, the wood had been sitting in water for weeks or even months at a time.'"
So maybe one day we will be able to measure all of those indefinable aspects of sound we can only attempt to describe with words. I do know this: when it comes to the tone of recorded instruments, whether it's a sax, or a human voice, or an oboe or a cymbal, nothing sounds more "accurate" or life-like to me than a good tube amp. Knowing that I'll probably never have a pair of the Wolcott mono amps, it might be more cost-effective to make use of tubes
and solid state. Maybe bi-amping my speakers, using solid state for good bass grip on the woofers, with tube amplification on the mids and tweeters will give me the best of both worlds. Who knew that tubes and solid state could live together?
MasterHiFi
