Tube vs Solid State.... is it mostly illusion?

[giorikas81]

Ok but I wouldn't mind having the power of a 2:90 with its simulated behaviour/feel in class-D 7kgr power amp in 1U. Light powerfull and no biasing or tube swapping oh no! 100$ and it does not sound as it used to. If that day comes screw tubes. Either way 80% of the annual tube manufacturing is directed to guitarists and bassists. Of course some people benefit and will benefit by producing tubes and tube amps and are not going to be put down for the sake of digital or solid state technology. It is like saying ok, LCD or LED tv's are cool but are never going to exceed the quality of old CRTube TV's. But they already did. But sound is subjective. It is like digital and analog delay, I hate analog delay I think it muddies up the sound and is not audible in live mix with another heavy guitarist. Digital is a way to overcome this analog shortcoming. Of course, tubes is something different alltogether. Then again, vintage sound is not quantisized. So it cannot be exceeded as it is not measured by mathematical forms but is subjective. A common misguidance is that solid state/hybrid or digital designs try in any case to replicate tube amp sound.

As a result if they succeed what do they get? Not a better sound, or different, just something that it can be equally good to tubes. And if some years ago this was a neccessity because tube thing cost an arm a leg and half of a right ball, now bugera and even soldano MIC gives you excellent tube tone from 300$. Some even for less, (5watt etc). So ss and co started to lose positioning in the market.

If you come to think of it, the most appreciated solid state amps in guitar world are amps that did not imitate tube behaviour. These are mainly randall, roland's jc-120 and similar amps (fender steel jazz king etc), ampeg VH4 and sunn's sludgy stoner line...Of course not all music genres require tube dynamics. Fast percussive riffs might benefit from ss. Engl Powerball the best solid state tube amplifier!!!

But in music instrument history (because good amps are music instruments in their own right) classic sounds were mainly recorded with tube amps. As a result what can be said about tube vs solid state? Certainly not an illusion but I think aphorisms are way off the truth in any case especially as guitar players are the most pambered and subjective buyers in the market!

 

[silentrage]

I certainly don't want to convince anyone if they don't believe in digital, there is something that's unquantifiably cool about having a real vacuum tube amplifier, I have a few so I know.

But objectively speaking, we're approaching the computational capacity of a human brain, we're modeling complex planet-wide weather systems, we're calculating plasma physics, super gravity and the beginning of the universe, I can't fathom how vacuum tube physics could be out of our reach for long, unless we just don't care enough to do it. I mean you could argue that these things are only done with super computers, but a common desktop today is vastly more powerful than a super computer of 10 years ago, like-wise a desktop computer 10 years from now will be vastly more powerful to a super computer of today.

So from that perspective, I firmly believe the audible aspects of tubes will be more or less replicated by digital technology and the utility aspect will be overtaken by far, if we want to do it.

The good thing is, when that time comes, having an amp that still runs on real tubes will be that much more cool/oddball, and I plan on hanging on to mine, haha.

 

[giorikas81]

I think it comes down to the real physical truth, which is that tubes make lousy amplifiers. A great amplifier has very little error, meaning you get out what you put in, but maybe louder. Transistors are very good at that - put in small signals, large signals, fast signals, slow signals, they don't care. They respond similarly in all cases - they reproduce the input faithfully.

On the other hand, with tubes, you don't know WHAT you'll get. Dial in a response, change the signal level or frequency, and you get a DIFFERENT response.

I think that is why we like tubes. Play soft. Play loud. Play fast. Play slow. Play high. Play low. Each has its own tone. So we can be MUCH more expressive. Once you learn to manipulate the changing reponse of the tube amp, you can get a wider range of expression out of it. With solid-state, you get the same stuff out regardless of how you play, with little exception.

Yes but think outside of the obvious. You describe a response to a non-linear user. As a result even tranzistors can be non-linear. The player is non linear. Your whole guitar is non-linear. Hey, if I play hard enough AND the strings have enough room to vibrate(medium high action) I sound tubey acoustically!!! And many clever manufacturers think outside the oscillioscope and put a few dynamic links here and there. Mike B says, he mods many mark III to IIIC+'s they measure and sound the same but they don't feel the same in many players to an exact IIC+. In that regard, how can amp modelling win if mike B cannot...

Then you got tech21. I am a longtime fun. Some sounds I adore some I dislike. The dynamic feel is incredible. Much better than many mediocre tube amps. Different regarding to the way it responds or it is delivered. Not a tube amp. Equally dynamic but not the same. I am betting there are some tricks inside the circuit to model or replicate or create dynamic involment. I feel it as an extension of my playing as a musical instrument even though different than a regular tube amp. And I have a great time playing through it!

Then you got Axe-fx. You can say that many companies analyzed the recorded sounds of many amps and paid tribute in their own way. Even if 1000 tube amps with the same specs don't sound the same they have a common sonic signature. Modellers do that. Catch a certain sonic signature maybe from a famous recording. But do they model all the components of the amp or just the sound? Axe fx and revalver do. And they sound and now, feel great to me. As of personal use I refer to the Revalver which I adore for the way it let's me percieve sag, overtones and little other things that change when I play differently. Also the fact that I hear it in lower volumes let me analyze it more easily. It gives me sounds that if they exist so rich from real tube amps but in high volume, my ears due to the fletcher munson curve will/may not be able to pick up. And they will hurt.

Then again when you play all out attack in a heavy sludgy or thrashy super high gain sound with low action, what kind of dynamics will you expect...? Even from the best tube amp? Play fast. Play loud. Play slow. Play soft. All out gain and compression?


So many years from now let us suppose that cellphones will have absolute power to run an accurate emulation of hetfields crunch berries mark IIC+, its components, the mark IV transformer he planted in there and the way they react in all volume with the help of a super fast CPU and fuzzy logic. So the emulation will be complete on the digital plain. All that remains will be a "perfect amp to recreate the preprogrammed digital emulation in any level without coloration". What will you chose? A solid state? A class D-digital? Some new type? A tube? what power tubes?

Still I insist. If tube tone is something that could not be touched yesteryear it is extremely affordable nowadays, especially if you know what you want before you buy. Modelling solid state etc is the king in versatility and low cost gigable amps though at least for me. You can take 230 euro peavey bandit new and gig with it with multiple sounds an effects loop, a solo boost and headroom and power that will surprise many. This you cannot do with an orange tiny terror that costs double the money. But if you want only the orange sounds the bandit won't do.

 

[silentrage]

I know I may be going pretty far here, but one day we will have the ability to scan the exact atomic configuration of any object and replicate it exactly with nanomachines, this configuration can then be stored, recalled and altered at will. That means there is almost no distinction at all between a real tube amp and a nanomachine copy, except this tube amp can shift and morph into any other tube amp, or a tv, or a toaster oven.

 

[Heritage Softail]

There is a difference the way solid state and tube components clip when being driven to distortion. You can see it on an oscilliscope in the rounded waveform of a tube clipping, and then the sound of a tube distorting is more smooth where a transistor is more angry sounding. There are also linear harmonic differences that create a more buzzy harshness in solid state, but that sense of liking or not liking the harmonic differences is a matter of taste. I think for angry metal, solid state may have an advantage. It all gets down to taste. If you like it, buy it and play the hell out of it.

 

[ando]

I know I may be going pretty far here, but one day we will have the ability to scan the exact atomic configuration of any object and replicate it exactly with nanomachines, this configuration can then be stored, recalled and altered at will. That means there is almost no distinction at all between a real tube amp and a nanomachine copy, except this tube amp can shift and morph into any other tube amp, or a tv, or a toaster oven.

Yeah, you are going pretty far there. lol! I think what you are describing is a LOT further off than you think. I would also dispute that computers are anywhere near the sophistication of the human brain. The whole "computational speed" argument is irrelevant. The brain is an incredibly complex system that is well beyond what computers are doing or are likely to do in out lifetime. Neural networking is currently only trying to replicate certain aspects of how the brain functions. Covering all the functions of the brain and hooking it up properly will be an achievement of the highest order. It begs the philosophical question of whether humans can actually conceive of any machine that rivals its own complexity. Personally, I don't think so. Humans will have to evolve further to produce computers that rival our current brain. Our understanding of the brain is still very primitive - as are our instruments for measuring brain activity. What that means is that neural network scientists are making an awful lot of guesses and premises about how the brain works and then trying to model their computers on these premises. We're not even close - trust me. And I'm good friends with a neural network professor. Even he admits we're just getting started on this stuff.

On the other hand, replicating a tube amp doesn't seem so hard in comparison!

 

[elvis]

About 15 years ago, I saw a marketing guy pitch the idea that "in 5 years" people would all have gone to full computer automation of all household functions, including television, appliance control, heat, etc. He also claimed the computer interface would have evolved to 3D with goggles and control gloves.

In the real world, things move slowly. The first widespread PCs came out in the 1980s. They can do a lot now, but they're not all that different from the originals.

 

[silentrage]

The first PCs did something like a few million instructions per second, and had like 512kb of ram, they cost $20,000.
My pc does a few billion instructions per second, and has 8gbs of ram, that's a few thousand times more powerful, and it cost $2000/
Before the first PC we had calculators, which was already hundreds of times faster than the average human,
Before that we had abacuses, which was tens of times faster than the average human.

It's only the same to you if you don't pay attention to it, but being the nerd that I am, I believe it's moving by a factor 2 every year or so, with software development anywhere from 1-3 years behind taking full advantage of all that power.

edited cuz I realized I sounded a little confrontational, I don't mean to push this down anyone's throat, just my personal observation.

I know I may be going pretty far here, but one day we will have the ability to scan the exact atomic configuration of any object and replicate it exactly with nanomachines, this configuration can then be stored, recalled and altered at will. That means there is almost no distinction at all between a real tube amp and a nanomachine copy, except this tube amp can shift and morph into any other tube amp, or a tv, or a toaster oven.

Yeah, you are going pretty far there. lol! I think what you are describing is a LOT further off than you think. I would also dispute that computers are anywhere near the sophistication of the human brain. The whole "computational speed" argument is irrelevant. The brain is an incredibly complex system that is well beyond what computers are doing or are likely to do in out lifetime. Neural networking is currently only trying to replicate certain aspects of how the brain functions. Covering all the functions of the brain and hooking it up properly will be an achievement of the highest order. It begs the philosophical question of whether humans can actually conceive of any machine that rivals its own complexity. Personally, I don't think so. Humans will have to evolve further to produce computers that rival our current brain. Our understanding of the brain is still very primitive - as are our instruments for measuring brain activity. What that means is that neural network scientists are making an awful lot of guesses and premises about how the brain works and then trying to model their computers on these premises. We're not even close - trust me. And I'm good friends with a neural network professor. Even he admits we're just getting started on this stuff.

On the other hand, replicating a tube amp doesn't seem so hard in comparison!

Fully agreed there! All I'm arguing is that we're making progress, and progress is being made faster the more we try with modern technology, so the eventuality is that it will be done.

 

[b0nkersx]

If you don't think SS amps will replace tube amps, you're kidding yourself. It may take upwards of 20 years, but it'll happen for two big reasons that I can think of:

1) Increase in technological capacity: this is inevitable and will certainly play a large part in guitar amplification in decades to come.

2) It's just a matter of time before tube production plants shut down. The demand for them doesn't go anywhere but down.

 

[ando]

If you don't think SS amps will replace tube amps, you're kidding yourself. It may take upwards of 20 years, but it'll happen for two big reasons that I can think of:

1) Increase in technological capacity: this is inevitable and will certainly play a large part in guitar amplification in decades to come.

2) It's just a matter of time before tube production plants shut down. The demand for them doesn't go anywhere but down.

1. Yes, as long as highly musical engineers are working on it and have a strong feel for what they are trying to model. It's not just a matter of input and output data.

2. Not necessarily true. Tubes underwent something of a renaissance when the audiophiles of the late 80's and early 90's decided that tube power amps were a necessary offset to the digital age of music. The high end of both HiFi and guitar amps is heavily dominated by tube technology and I have only seen the number of tube amps increasing, not decreasing. It's also happening with LP records. Everybody assumed they would die a natural death but they have also made a comeback due to their increased fidelity and dynamic range. In the long term, I think it will be the power consumption of tubes in an energy hungry/rationed world that determines how long tubes remain viable. For now, they are not going anywhere fast.

 

[silentrage]

I personally think tube amp might be like the landline telephone, sure it's still got some advantages like consistently good signal, but as soon as cell phones can do what it can do, it'll pretty much go extinct. Maybe not extinct, certainly relicated.

 

[b0nkersx]

2. Not necessarily true. Tubes underwent something of a renaissance when the audiophiles of the late 80's and early 90's decided that tube power amps were a necessary offset to the digital age of music. The high end of both HiFi and guitar amps is heavily dominated by tube technology and I have only seen the number of tube amps increasing, not decreasing. It's also happening with LP records. Everybody assumed they would die a natural death but they have also made a comeback due to their increased fidelity and dynamic range. In the long term, I think it will be the power consumption of tubes in an energy hungry/rationed world that determines how long tubes remain viable. For now, they are not going anywhere fast.

I disagree. Imagine it's 1945 and somebody tells you that in a few years, people are going to recreate the bombs dropped at Hiroshima and Nagasaki on a smaller scale in order to power your houses. People today are still extremely skeptical and would rather rely on older methods, but truth is, nuclear power is better for everybody. Pardon the analogy, I'm just trying to draw parallels.

I'm not saying that this technology (with regards to guitar amplifiers) will evolve rapidly at all, but it'll happen, more than likely in my lifetime.

That being said, the market for tubes for guitar amplifiers might be increasing, but the market for vacuum tubes is, in general, decreasing.

 

[giorikas81]

There is a difference the way solid state and tube components clip when being driven to distortion. You can see it on an oscilliscope in the rounded waveform of a tube clipping, and then the sound of a tube distorting is more smooth where a transistor is more angry sounding. There are also linear harmonic differences that create a more buzzy harshness in solid state, but that sense of liking or not liking the harmonic differences is a matter of taste. I think for angry metal, solid state may have an advantage. It all gets down to taste. If you like it, buy it and play the hell out of it.

exactly!!!
You are absolutely right. So don't make em clip. Don't make distort. Pre design the whole thing and THEN make em play it. That is what is all about in modelling and other amps. Tranzistors may be the electronic world's best uncoloured classical performer. Don't make a violinist play blues on the bass. Don't make a tranzistor imitate the electronic behaviour of a tube. Mosfets come close though.

 

[Anonymous]

I think my big problem with digital is that it largely removes the "art" element of designing amps from the ground up. Instead of Randall Smith toying with different components, you get a computer engineer. I'd rather spend $1000 on a tube amp than $100 on something digital that does the same thing. Don't get me wrong, I love computers and know enough about them to build my own systems, but I haven't got a very good idea about how they actually work. Tube amps? I've had several problems and have resolved them all myself without any prior knowledge.

For me, until digital can do something better than tube amps tone-wise - and I mean raw signal, not effects processing - I'm not interested.

 

[silentrage]

Actually I think peavey is on the right path with the revalver.
If you can use a program to accurately replicate an analog signal path, then you can play around with arranging caps, resistors, tubes, pots, etc like you would a real tube amp, this way it's much more cost effective, and you could get unique sounds that no one else has. We're not quite there yet but it'd be fun for a noob tinkerer like me, it's quicker and less likely to kill me. :lol:

 

[giorikas81]

I think my big problem with digital is that it largely removes the "art" element of designing amps from the ground up. Instead of Randall Smith toying with different components, you get a computer engineer. I'd rather spend $1000 on a tube amp than $100 on something digital that does the same thing. Don't get me wrong, I love computers and know enough about them to build my own systems, but I haven't got a very good idea about how they actually work. Tube amps? I've had several problems and have resolved them all myself without any prior knowledge.

For me, until digital can do something better than tube amps tone-wise - and I mean raw signal, not effects processing - I'm not interested.

Wait a minute here. When these legendary builders started there were no solid state or digital. So, they worked with what they had. Randall Smith said that he is absolutley fine with using whatever sounds good, but personally he doesn't know HOW to design a solid state amp. He is an open minded honest man and of course a legend on his own game.

As for the art and experimenting, what is the really trully new sound or amp that happened in tube land nowadays? None. You know why? Because tube amps are LEGOS. They actually work pretty similar with the same rules so you only get variations of the same sound in essence. Why ? Beacuse tubes are of certain type. The new mesa series comes in and what do you expect? A 18-25 watter EL84 a 50 watter 6L6 etc...Preamp wise you can play with 12AX7's all day plus some clever way to manipulate the position of the tonestack and some capacitors for push pull filters and voicings, but they choose an eq and the overall gain and that is that. They cannot deviate more than what is already done. Cabinet comes in mind but things are pretty detrimental there too for some years now.


If you expect a trully boutique amp builder to be there for you personally to custom tailor an amp to your liking plus its responses through some components, the same thing Cliff from Fractal audio/Axe-fx does. But he does it on the digital domain? Why? Because he can. Others can't. He can. He designed the Axe-fx soleheartedly, the results are spectacular, each model consists of 50000 lines of code, is as custom as any tube amp can get with bright virtual capacitors, Sag switch etc. He is every bit a guitar audiophile as Randall Smith and he succeeded in much more difficult situations and not on a blooming market but on a rather dead one. He is not just a computer engineer. Maybe not a true pioneer but every bit a butique builder for what I'd want from one. His product isn't cheap in fact you pay for it as much as a mark V or mark IIC+ used and still you need a power amp and speakers. But its sound is top notch and as a preamp it matches any marshall or mesa preamp we managed to put side by side for raw signal...

 

[elvis]

I am a bit offended by the idea that a computer engineer cannot be an "artist", only Randall Smith (and other tube tinkerers) can. Would he be less of an artist had he studied C++? After 20 years of designing ICs, I can tell you without hesitation that engineers ARE artists.

I can also tell you that the idea that tube distortion is fundamentally different from solid state distortion is misinformation at best. It's not OK to compare two totally different circuits on an oscilloscope and claim the result is entirely due to tubes vs. transistors. In addition, a simple sine-wave comparison covers 1% of the claimed differences between tubes and transistors, i.e. dynamics.

 

[elvis]

OK, "offended" is the wrong word. More like "disappointed". The people that invented things like ICs, spacecraft, cellphones, and nuclear power are artists. Very smart artists.

 

[ando]

Elvis, I think the misunderstandings can creep in when the engineer is not a serious musician. I have no particular allegiance to tubes, but tube amps in general, especially very good ones, just happen to embody many of the qualities that musicians want when playing their guitar. The dynamic feel and just the right amount of break up. They are very responsive and allow the full range of desirable musical elements that an artist wants.

I have no doubt that solid state electronics are capable of the same thing - it's just that at this stage, engineers are still side tracked by the notion of re-creating the tube response and feel. Actually, I think they already have the sound just about nailed - it's more the actual player response that is still lacking. Quite possibly that is because a lot of code is needed to model the sound and that creates a latency or buffer between the player and his dynamics. That may be just a matter of better hardware or more intuitive programming. It seems that there aren't too many engineers on the case who have an innate grasp of what the player wants and how the electronics can achieve that. Tube amp designers have a huge advantage because the main component is the tube, they don't have to design that part of the circuit. It's a reliable and known quantity.

Being a SS amp designer actually requires a lot more of all your abilities. The real progress will happen when there is no thought of re-creating the tube, but an just an innate feel for what parameters will produce the tone that is desired. I don't think many people are in the position of having that innate feel plus being a true musician at this time. When that does happen then, as you say, that will be real artistry. Most tube amp makers are musicians who learned a little electronics. Tube amps are pretty simple when you look at it. The vital connection of technical nous and musicianship is easier with tubes.

I haven't tried an AXE-FX yet, but that does sound quite promising. I think I'll wait for a while though until they get it right.

 

[elvis]

Thanks for the explanation, Ando.

I wasn't really trying to get into the technical issues (which are valid, frankly), as much as I was responding to the sentiment in MagicEight's post.

I can't understand why people who tweek on ancient technology are automatically viewed as artists, while those pushing the envelope (programmers, for example) are considered to be posers or incompetent.