Slashdot Mirror
Tubes vs Transistors: An Audible Difference?
cgenman writes "Are those vaccuum tubes worth the extra price? This paper, a transcript of a speech to the Audio Engineering Society of New York, indicates so, though the reason is surprising: Overloaded tubes behave better.
While the speech itself is from the early 70's, the paper takes on new importance with the recent trend in louder is better music."
NO!
Dollar for dollar, transistor amplifiers output far more power before they're overloaded, making this discussion moot.
If you like the distortion tube amps give (remember, you're not getting the audiophile shound, you're getting "nicely" distorted sound) I'm sure a DSP can do it for you. Even an EQ would probably help.
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
The recent trend in "louder is better." Did I just read that? The recent trend? Since the first real Rock and Roll music appeared approaching, 60 years ago now, louder has been better. That's a "recent" trend?
How does a speech from the 70s, discussing how better "behaved" tubes are, have relevance today? Transistor technology has had 3 decades to grow into a more stable, mature platform for audio, and we understand a great deal more about the nature of sound and the equipment producing that sound.
Digging up an ancient speech which probably SPARKED the religious war in the first place is idiotic, in my opinion.
What's next? Will we dig up some argument from the 1880s about the superiority of DC-delivered electricity?
Better for whom? The average listener won't be able to tell the difference, this is like how theres a few nutbags such as myself that still enjoy listening to vinyl. It can just sound better sometimes.
Also how relevant is this? 30 years ago, we've got all kinds of DSP going on now and very efficient transistor amps putting out a boatload of power before they become strained.
The problem with the louder-is-better issue is the albums themselves. They're mixed horribly. You can play them on a cheap boombox or a system costing thousands of dollars. You'll just hear the garbled shit more clearly on the multi-thousand dollar system.
Presently here, but not there.
The only real place where this has any impact is in recording and performance; amps are frequently overdriven to provide a "fuzzy" effect - guitarists will know exactly what I'm talking about here. There, tubes and transistors sound quite different, and tubes do sound quite a bit nicer.
I'm sick of all the "audiophiles" who claim that a non-overdriven tube amp provides a better reproduction of any given sound than a similar, transistor-based amp. The fact of the matter is, transistors provide a better sound reproduction, as there's less interference from things like the tube's heater or outside magnetic fields. Whether it sounds better or not is up to you, but don't try to tell me that it's a better reproduction.
That's it. I'm no longer part of Team Sanity.
The record labels want to ruin the CD format
The CD has outlived its usefulness to the labels. They want to move people onto a copy-protected medium so that the MP3 problem is squashed. And think how much better the properly leveled SACD will sound next to the clipped CD.
No, I don't trust in god. He'll have to pay up front, like everybody else.
While the speech itself is from the early 70's, the paper takes on new importance with the recent trend in louder is better music.
I think when loudness becomes music's most important quality, the word "music" should be placed in quotes.
Really, why care about perfect reproduction when your ears are bleeding?
"You mortals are so obtuse." -Q
I really wish AOpen had more success with their Tube Sound motherboards... If they had released one that supported the CPU I wanted I would have bought one. :(
Any unintended (i.e. can't shut it off if you want to) effect on the audio is distortion. Period.
Some distortion sounds better than other types. But in the end, you are still getting a signal that is not reproduced faithfully.
(As an aside, modern MOFSETs produce even-order harmonics in an overload situation, just like tubes. This is opposite earlier IC-based gear that produced odd-order harmonics, which are much harder on the human ear. I think this is what the linked talk is going on about. I might also note that audio technology has grown by leaps and bounds since the 70s.)
If you like the "warmness" of a tube, then grab a tube preamp and a modern amp and you can now have the best of both worlds.
The "Audiophile" business is chock full of snake oil, even moreso than many others. $1000/ft "de-ionized oxygen-free" cables? LOL.
Natural != (nontoxic || beneficial)
Typical solid state amplifiers have increased in power an headroom to the point that you are unlikely to want to listen to them at clipping.
It is certainly true that some people like the coloration introduced by tube amps. Guitar players routinely treat tubes as musical instruments by overdriving them.
Another (non-disjoint) set of people enjoy the coloration and noise of vinyl recordings.
The bottom line is that you can make a digital recording of your favourite vinyl/tube/whatever golden-ears setup, and be unable to distinguish it from the original in controlled A/B comparisons.
If you want to color your music, use tubes. If you want high fidelity, don't.
Whether it's tubes vs. transistors or vinyl vs. CD, it's worth keeping in mind the distinction of "sounds better" vs. "reproduces accurately". You may *like* the sound of tubes or vinyl better, but within normal limits of operation, there is no way tubes or vinyl more accurately reproduce sound than CDs or well-designed solid-state equipment.
As far as the article - the THD levels (3% to 30%) aren't unusual for 60's era equipment. Since the late 70's it's no big trick to design "transistor" equipment that has essentially unmeasurable THD even approaching rated power levels - it just requires lots of feedback and a better power supply than most consumer equipment has.
There isn't much point in observing that tubes clip waveforms more softly when you can design solid state equipment that never clips at all. However, some people may prefer the distorted output of tube amps to the accurate output of solid state amps.
I still use tube amps for guitar ("sounds better"), but all solid-state for playback ("more accurate"). Fender (and probably others) now offer DSP based amps that will emulate tube amplifier sound - haven't ever tried them, so I'm not sure how good they sound.
I love reports that tell us what is musically "better". It reminds me of the debate over, of all things, guitar strings.
Some people (Angus Young of AC/DC, for example) swear by using new guitar strings, replacing them as soon as they get a bit worn. Others (e.g. Neil Young) won't use 'new' ones and actually have roadies break their strings in before they will play them.
(Angus also likes to use no effects pedals, while Neil loves effects. Just picking those 2 at random 'cuz I read up on them. Which is better-- straight guitar or with effects?)
Which is "better"? The answer is 'whatever gives you _your_ sound'. You like tubes, go for it! Solid state give you what you want, more power to you!
With amps, people get distracted by engineering gobblygook, but the truth is: to get 'killer tone', you need to choose your own mix. Guitar choice, strings, amps, heads, effects, EQ, there's a fucking reason you can buy a million and one of each-- there is no one right path!
You can't define sound. It's experiential*. There's no one right set of gear. There's no one best type of music. There's no one best musician. There's no best album of all time.
Freebird! Freebird!
*(sonically, you can usually define 'sucky' due to poor audio quality, but when you get into 'good' you start getting into taste as much as specs)
A.
This topic is just not news: good audio-amp books that deal with it well have been around for years.i nsley_Hood/searchBy_Author.html .
For example, some really good explanations and designs relating to this topic are given in a series of books by John Linsley Hood, findable at http://engineering-books-online.com/search_John_L
(Some knowledge of analog(ue!) audio electronics is needed to follow some of the points fully.)
IMO some of the information can be summarised like this: Very good amplifiers can be made both with vacuum tubes (or valves!) or with transistors, and very good examples of each tend to sound alike. Some quite subtle distortion issues can arise in transistor amplifiers, from details of the way in which high-frequency rolloff is applied to obtain feedback-amplifier stability against unwanted high-frequency oscillation.
In an earlier life (!) I built/modified some audio amps to JLH's designs, I also decided to choose commercial amps on the basis of checking their design circuitry, (where the manufacturer would agree to disclose it, which not all did), to see if their hf stability circuitry is applied in the way that JLH's design criteria indicate that they should be. Not all high-price audio amps do that.
With examples that do, I found that my ears can (or at least they used to be able to) distinguish what I would call an unforced, neutral, clean sound quality, with undistorted transients, specially audible (for example) in the way that a triangle-sound is left clean and un-fuzzed, and in the way that the sounds coming from the mass of a band or orchestra emerge as distinguishable individuals rather than as a fuzzy sound-mass. Of course, good recordings and input signals
as well as good speakers are needed for any such subjective aural tests, and naturally any amp suffers to some extent if overloaded. It needs also to be noted that the standard that is met by an overloaded tube amp but not by an average overloaded transistor amp is a standard that tolerates a very high and audible level of certain kinds of distortion.
-wb-
So bloody what. This is not news, it's been known by every audiophile on the planet since the inception of transistors. Transistors clip more harshly than tubes. Tubes clip softly, transistors clip sharply. If you want to go loud without clipping, buy a better amplifier.
You make an important point. Good musicians don't want perfect reproduction. They want music. What they want is imperfect reproduction because they want a particular sound that their instrument doesn't naturally give. It's kind of like visual fidelity in movies. It you look at raw movie footage it looks very harsh, kind of like home movies. They have to artivicially color grade it to make it look good. Again they don't want perfect fidelity, because perfect fidelity looks bad. It's the imperfect fidelity that looks good. It's kind of like the old Monte Python sketch in which the american movie director explains that he is shooting snow scenes on the beach because " It looks more like snow than snow."
...is that we have achieved amplifiers based on transistors that are more accurate than human hearing. Once you achieve that, there is no point in having anything else.
:)
Any effect, such as that of a tube amp, a vinyl player, or whatever else makes music better for you, can be emulated. Any distortion, clipping, overloading, whatever.
Audiophiles live in a reality distortion field which makes Steve Jobs (Apple) look like a kindergarten magician.
Call me when TV has the same luxury problem. "This here looks completely real, but some people claim they can see the difference between this and reality. Those videophiles are crazy!". It'll take a lot more than HDTV to do that... and in 3D of course
Kjella
Live today, because you never know what tomorrow brings
Or, you could just not purchase music that sounds like crap...
Patient: Doc, it hurts when I do this.
Doctor: Don't do that.
Or, maybe you could get a really tiny tube amp, and listen at normal volumes, and hope that all the "pleasant" distortion of the tube amp in overdrive cancels out the overmodulated recording.
Solid state and tube amps have almost no comparison. Id take a tube anyday
My, how insightful. Except that it contains exactly zero facts.
Translation: I'm a dumbass and don't care about reality as long as I can live in my audiophile dreamworld.
And sure enough, if you want to hear the ultimate in reproduction from a classical orchestra it is preferable to possess your own concert hall and hire a real orchestra!
The problem with the valve (tube) Vs. silicon debate is that it doesn't relate to the 'average joe' who listens to snatches of music 'on the go' on their radio, CD or MP3 player, probably while doing other things such as sitting on a train, driving their car or working on their PC. Under these circumstances the listener isn't focusing solely on the purity of the sound reproduction but on the 'background noise' that the sound provides with a familiar or favourite tune.
Naturally, a true audiophile will have their own acoustically perfect listening room, will slip on their favourite headphones or sit in front of their favourite speaker system and will wait for their tubes or FETs to warm up - heck no, they'll never turn them off in the first place! Under these circumstances the audiophile will buy whatever they believe will do their 'listening pleasure justice' - tubes, FETs or hybrids. Fair enough - those with the money can do what they want, but the vast majority will be happy with their Sony, Panasonic, PC system etc. and won't give a stuff what actually makes the sound come out the speakers.
In a similar way, the recording industry's attempts to thwart the 'for personal use' pirates with copy protection mechanisms makes be laugh-if I REALLY want to make a copy of something 'protected' and I can't be bothered to find out where to download the latest crack or workaround off the 'net then I'll simply hook up a stereo mike in front of my speakers and make a copy that way - naturally, this won't give me a 100% perfect audio copy but that's NOT going to bother me if all I want is a 'rough and ready' copy.
AT&ROFLMAO
Sigh. This whole discussion is full of lots of opinions from people who don't seem to understand what the paper is talking about, begining with the submitter.
First, the paper is refering to microphone preamps, which are used to boost the very, very low level signals. These signals are affected by impendence, one way that vacume tubes are different that transistors. Both are good, both can be used to make very good gear, both can be used to make very bad gear.
The difference in harmonic orders generated by distortion is important because equipment is often used to intentionally generate distortion because sometimes it's pleasing to the ear. Tubes also begin to compress the waveform when driven into distortion, which often is pleasing to the ear. And sorry, there's no advances in technology that's changed those basic laws of physics/electricity. That's not to say solid state stuff is bad, just different.
Virtually every rock/country/pop CD out there has passed through a selection of vacume and solid state technology. We use the best tools to generate the tone we want, regardless of the technology. If you go to a high quality studio, you'll find that most of the audio monitors are powered by solid state amps. You'll find racks of solid state and vacume tube mic preamps, EQs, and compressors. You'll find lots of tube based guitar amps and very few solid state ones.
An LA2 compressor has tubes and sounds like god on some things. An 1176 doesn't have tubes, and sounds like god on some things. I reach for the one that best serves my needs, not what technology it's built on.
BTW, most real studios don't use the monster cables that audio stores will try to sell you. We use plain old, high quality wire with quality connectors that cost much less than any of the audiophile stuff.
As far as the loud is better stuff spouted in the submission, that has nothing to do with it. You can design a 1 watt tube amp that's very overdriven to get certain sounds at low volume. It's all a matter of knowing what your desired effect is and the purpose, and designing the equipment to deliver it. A 60 watt 4 ohm amp for home listening has entirely different design considerations than an amp designed to deliver 4500 watts 2 ohm for sound reinforcement.
I think you missed the point of the post you replied to. Screwed-up dynamic range means that the "master" copy of your art is already fsck'ed. Thats like having Mona Lisa "painted" with an inkjet printer. These aren't just bootleg copies - these are the master versions everone (and their ears) has to suffer through.
The pointy-haired twit who started this trend in the music probably thought post-modern degeneracy "Art" as well.
On a day when the real-world news is rife with examples of how faulty information processing has lead to multiple thousands of deaths, Slashdot dredges up issues with studios' technology from the 70's and claims they apply to consumer choices of today. Of course, in the fine print, NONE of the boundary conditions that are pushed, accidentally or intentionally, are similar.
...
Clueless, disingenuous or manipulative? I couldn't tell. But it's not exactly helpful in forming a well-considered mindset about audio design.
Here's my 3-bullet take on the weird juxtaposition:
* The older paper (as well as others quickly linked to) talks about how studios risked distortion by pushing amplifiers past design limits in order to escape tiresome, easily-heard tape hiss. In the 30 years since, the dynamic range of amplifiers has improved (less likelihood for over-the-edge conditions); metering and sound checks have gotten easier and faster, leading to fewer mistakes; and (analog) tape hiss, when it's an issue at all, has also dropped further down the list of concerns. Why is this archive paper relevant without those differences mentioned?
* The second-linked article vents frustrations that even live music is intentionally garbaged up by the creators. The sound is intentionally manipulated to sound "louder" which also makes it SOUND AS IF it was produced by over-driven equipment. That's the artists' prerogative, and the critic's job to carp about. Nothing to see here, folks, except that it interestingly links to
* a previous in-depth analysis of the Dark Side of the Moon SACD that details differences between formats that must have been driven by perceived preferences of listeners, not the formats themselves. Implicitly, some engineers seem to believe that CD listeners prefer LOUD while SACD listeners like "clean," because that's how they manipulated the two formats differently. For CD listeners, they clipped the sound INTENTIONALLY, and differently from any faults of the electronics, in a way that's unnecessary for the CD format. Clipping produces ugly noise on loud spots, but makes the recording sound "louder."
One might guess that engineers aim for the "cleaner" effect on vinyl, too. (Not too many vinyl fanatics risk installing their systems in cars, so they can groove while cruising along I-5, and probably not very many SACD systems, either.) And it's also not too much of a guess to assume that vinyl listeners are about 10X to 100X more likely to use tube equipment, which the owners have selected because it sounds (to them) more the way THEY prefer.
So this attempt at stoking flames under the War of the Formats (Audio Division) can be seen as having nothing to do with "Tubes vs Transistors," as titled. Rather, it oughta be, "my format Rools and yours Sux" or something more appropriate to the information that it provides to the topic. Absent the 2+2=17 faulty logic, the articles actually seem to show that engineering allows whatever "sound" the seller wants to feed the consumer, without any objective "quality" standard at all.
I propose "Troll of the Week" balloting to allow us to heap opprobrium on such posts. This shouldn't even make it on a slow news day. I'm all for vigorous discussion on "stuff that matters" but articles that encourage senseless flame wars don't exactly further that goal.
"Inquiring Minds Want to Know!"
that most music these days is all recorded and processed digitally, making all these analog sound generation techinques a moot point... except that they let one audiophile prove how much more sophisticated he is to another audiophile. I will stick to my nasty old mp3s, and chuckle when your music collection takes 100000 times more space than mine and sounds no different on headphones. If I want hifi, I go to the goddam concert and hear it live!
ALL amps distort the sound, that is an absolute truth and an extension of the natural world. So, the question is what KIND of distortion do you want? Most humans prefer the natural curved distortion of a tube amplifier to the harsher square or sawtooth clipping of a transistor. Until you can make a perfectly reproducing transistor amp that is the discussion, and it's one that most pro's agree the tube wins hands down.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
The parent poster is absolutely correct in that the goal of producing an amplifier is to create a perfectly linear constant multiplication of an input signal. Any non-linearities must be viewed as flaws in the amplifier design, or at least as limits to its practical range of use. The switch to transistor technology happened so rapidly because they have much better characteristics than tube amplifiers.
All the arguments here can be summed up as such. I feel if we keep these points in mind, we can stop arguing.
1) An amplifier by definition is used only to amplify sound.
2) Distortion is a negative byproduct of a poorly designed amplifier.
3) Tubes significantly distort the sound they amplify by adding a certain color of harmonic content dependent on how much the tube is driven -- this is mainly due to NON-linearities in the circuit.
4) Live and recorded music has developed an enormous dependence on signal processing -- both linear and non-linear. Signal processors can do ANYTHING to the sound, be it EQ, delay, distortion, reverberation, etc.
5) Many sound engineers, audiophiles, guitarists, producers, and listeners have developed a distinct liking to the tone color produced by specific applications of tube amplifiers. Many others can't stand the distortion. Still many others can't stand the distortion and yet purchase $10,000 tube amplifiers. I blame fairies too...
6) The individuals that argue for tube amplifiers are actually arguing for their use as a signal processor that is often conveniently attached to an amplifier.
And here's the real kickers that I've learned through years of live sound and recording engineering:
7) Never tell someone what they like and don't like to hear. They will never change their mind.
8) Never confuse amplification with signal processing, whether it's digital (DSP) or analog (tubes, etc). If a device does both, that's great, but realize that it's doing two jobs for you.
As a side note, most DSP doesn't accurately reproduce tube sounds becuase tube distortion is very nonlinear. DSP works well for linear systems because all linear systems have a 'transfer function' that can be used to simulate that system perfectly. Nonlinear systems do not have a constant transfer function, if they have a transfer function at all, and therefore would require an exponential amount of additional processing in order to recreate nonlinear distortion. Of course, there are tricks...
~Loren
I love tubes. I've actually had dreams about tubes. (seriously!) Energy wasting clumsy things that they are, I think they are great. I enjoy watching them glow and have for may many years.
:-)
I appreciate listening to a radio program more if it's from a tube radio. To me, it "sounds better" because I appreciate it more. If I were blindfolded and was lead to believe a certain radio was tube type and another was solid state, I might perfer the tube set even if they were actually both solid state.
As far as long lasting, you'd be surprised how tough tubes can be. It's usually capacitors that wear out before tubes. Tubes survive things such as sun spots or electromagnetic interference much better than transistors too.
I for one wish more consumer electronics were tube based. Maybe it's the construction, but an old tube radio was built *way better* than any of this mass produced, cheap crap from China.
I also think that food cooked in a cast iron pan
is better than teflon. I like dumb terminals, digital clock radios with mechanical dials and analog meters with physical needles, not the digital LED crap.
Oh yea, and I perfer a command prompt to a mouse any day.