Head-Fi.org › Forums › Equipment Forums › Sound Science › Tube buffers, "tube sound" or just gimmick?
New Posts  All Forums:Forum Nav:

Tube buffers, "tube sound" or just gimmick? - Page 2

post #16 of 35

Shanling MC3000 measures not very well considering the price..

for example full-scale signals show a THD of over 0.8% (the specs say <0.1%, yeah right!) and over 0.2% at -10 dBFS. This means that every peak, every impulse that gets close to full-scale will be distorted a lot more than quieter sounds.

FR is down about 1 dB at 20 Hz. IMD measurement shows very weird behavior with what could almost be described as raised noise floor at over -60. Nasty IMD products are only down -48 dB. Ouch!

 

STP80 valve amp measures bad as well. Almost 0.5% THD at only 12 watts output into 8 ohms, at least in one channel (lol). Oh and the distortion rises quite a lot at low and high frequencies (over 1% at 20 Hz and 20 kHz).

 

(Sorry I cannot post the original data, it's copyrighted.)

 

 

<rant>

I mean, seriously, even the world's first CD player (CDP-101) is better in pretty much any aspect. It was released in 1982. So in a way some manufacturers went backwards from there.. and still do.

Just listen to vinyl or cassettes if you want low fidelity.

</rant>


Edited by xnor - 2/25/13 at 11:39am
post #17 of 35
Quote:
Originally Posted by xnor View Post

Shanling MC3000 measures not very well considering the price..

...but just LOOK at it!  tongue_smile.gif

 

I want one as a room accessory, just to sit there and look pretty. Might even put a  dimmed pin-spot on it.  I'll play my tunes with something better.

 

Seriously, that's got to be visual-stimulous marketing, don't you think?  Sort of like the old Transcriptors turntables...not great tables, just works of art. I hope it's tubes with the nice blue lights under them glow nice and bright.

post #18 of 35
Quote:
Originally Posted by OJNeg View Post

 



How would you have configured the resistors to impart vacuum tube distortion on the CD players output stage? I'm genuinely interested.

 

The point is that you don't. Testers have used a simple resistor network to implement EQ in blind amp testing (I think Richard Clarke in particular?) All that this does is crudely change the frequency response (or less crudely if you build a more complex network of resistors I suppose) which is responsible for the difference that people actually hear rather than the story that the marketing people tell you about even harmonics. To give you an idea how silly the usual valve tube distortion riff is, this is from an engineer who designed some best of the best stage valve amps of the classical period of rock and who still favours valves for some applications today:

 

 

Quote:

 

http://lenardaudio.com/education/14_valve_amps.html

 

 

Occult audiophile claims about valves having a magical warm sound began in the 1980s nearly 20 years after solid state technology took over.   By paying attention to this illogical use of language, the assumption is, by putting an audio signal through an infinite number of valves and an infinite length of magical cable, the sound will become infinitely warm and infinitely magical.   This type of dumb marketing hype would have been viewed as a joke, in the previous era, when all amplifiers were valve.

Valve sound.   No valve should have its own sound.   If it does, then the added distortion is imposed on the music where it does not belong.   There is a vast amount of miss-information on the web about various valves having individual characteristic sounds.   This is only true when a particular valve (valve A) biased to give optimum performance is substituted with another valve (valve B) with the same pin configuration with different internal parameters.   Therefore the biasing that was used to correctly center valve A in its linear operating position does not apply to valve B.   The majority of differences heard are simply gain related, no different than if the volume (level) is changed slightly.   Other differences are to do with valve B operating outside of its linear region and the difference heard is excessive 2nd harmonic distortion.

However, valve amplifiers do interact differently with speakers compared to conventional solid-state amps and the resultant audible difference we notice is explained in valve/solid-state page.

Myths.   To appreciate valve amplifier technology we must first be-bunk audiophile myths that in-correctly describe them.   Many Audiophiles believe in alchemy and behave similar to religious cults.   Cult driven audiophiles are easily identified by their repetitious chanting of brand names, model numbers and meaningless superlatives concocted by high priest reviewers and marketing departments.

 

 

(I wish this editor would stop messing around with the text bg colour!)

post #19 of 35

Seriously jaddie?

 

I prefer the looks of this:

over this (or this ^_^):

 

yuck, those stupid blue leds are everywhere!


Edited by xnor - 2/25/13 at 1:10pm
post #20 of 35

Something about the Sci Fi Forum makes me want to vomit......................

post #21 of 35
Quote:
Originally Posted by xnor View Post

Seriously jaddie?

No, not really.  How can I be serious about this?

post #22 of 35

Hmm the sarcasm was quite obvious. I admit I wasn't reading properly.

 

 

 

Here's a video that shows the real purpose of tubes: tongue.gif

 

 

 

Quote:
Originally Posted by Chris J View Post

Something about the Sci Fi Forum makes me want to vomit......................

Reality sucks, doesn't it..


Edited by xnor - 2/25/13 at 4:27pm
post #23 of 35

I still don't understand how you could only use resistors to implement any sort of equalization. Wouldn't you have to use some sort of RC network (at least) to change the signal in the frequency domain?

 

As I understand, vacuum tube "sound" doesn't come from any sort of magic or alchemy. It's all based on the parameters of the device itself.

 

 

Quote:
The point is that you don't. Testers have used a simple resistor network to implement EQ in blind amp testing (I think Richard Clarke in particular?) All that this does is crudely change the frequency response (or less crudely if you build a more complex network of resistors I suppose) which is responsible for the difference that people actually hear rather than the story that the marketing people tell you about even harmonics. To give you an idea how silly the usual valve tube distortion riff is, this is from an engineer who designed some best of the best stage valve amps of the classical period of rock and who still favours valves for some applications today:

 

So you're saying that the harmonics aren't what give tube amplifiers their character? What does?

post #24 of 35
Quote:
Originally Posted by OJNeg View Post

I still don't understand how you could only use resistors to implement any sort of equalization. Wouldn't you have to use some sort of RC network (at least) to change the signal in the frequency domain?

 

 

I neither know nor care. The important point is that a SS amp can pass as an arbitrary valve in blind testing via EQ. I didn't investigate the comment on using a resistor network for emergency EQing because, well, walking around with a resistor network in my pocket when I could be using my DAP's DSP chip would be silly. However, I did google for you and you can do at least some forms of EQing using resistor networks:

 

http://en.wikipedia.org/wiki/Zobel_network

 

I'm surprised that you don't think you can do any EQ this way as even the impedance of a headphone can change the fr of an amp.

 

 

So you're saying that the harmonics aren't what give tube amplifiers their character? What does?

 

I don't know how to make this any clearer - not criticism of you, I'm just struggling: I am saying that valve amps do not have any particular character. They are associated with a warm sound, because that is how they are marketed and tuned, and this can easily be duplicated with EQ - as Clarke's continued possession of his money shows. There is no intrinsic tube amp character, just marketing and the tuning of modern valve amps so that they are mostly on the warm side of neutral.

 

Or sometimes they are even worse than that:

 

 

 

Quote:

http://lenardaudio.com/education/14_valve_amps.html

 

 

Before the 1960s most small radios and radiograms were made as cheaply as possible, minimum number of pre-amp valves with one output valve (Class A) that delivered 2 to 3 Watts of power.   The higher quality, more expensive radiograms that were marketed as genuine Hi-fidelity had 2 output valves in push pull (Class AB), achieving lower distortion with a greater frequency response and were 10 to 30 Watts per channel.   The difference between a cheap (Class A) 1 output valve amplifier and a more expensive push pull (Class AB) 2 output valves per channel amplifier, was greater power, fidelity and an impressive bass response.

Leak, Quad, Mullard, McIntosh, Fisher, never made SET valve amps.   This idea to them would have been an abomination.   They only made Class AB amps.   Maybe they were ignorant not to realise that if they removed one output valve, their amps would have sounded magical to those who regard themselves as occult audiophiles.   It appears these early valve amp manufacturers were blindly obsessed with making amps that created no sound of their own.

 

post #25 of 35

Something as simple as a resistor in series with the load can raise the output impedance of a SS amp to that of a tube amp. The effects should be obvious: a flabby bass boost, with some headphones a peak in the bass response and sometimes even a little treble boost.

 

 

Attila Czirjak (developer of the ginko monitors) on tubes: "It's a fact that tubes used properly do not sound different than transistors. So from a technical point of view I see absolutely no point in using tubes."


Edited by xnor - 2/26/13 at 7:06am
post #26 of 35
Quote:

Originally Posted by xnor View Post

 

(Sorry I cannot post the original data, it's copyrighted.)

 

What exactly do you mean by "data"? If you mean something like a distortion plot, you can't copyright something like that.

 

se

post #27 of 35

Wow...this has gotten a bit deep.

 

Here's the thing on using resistors for EQ.  It doesn't work that way.  Yes, adding a resistor in series with a SS amp output may result in a change in response, but that change depends on the load impedance curve, which when combined with the resistor creates a frequency dependent voltage divider.  If the load Z were a flat curve, there'd be no change in response, no "eq" to be had.

 

The Zobel network Wiki article...um... that's not using just resistors for EQ, there's always a reactive element.  You can't EQ anything with just resistors, no matter how many there are.  Funny that article was referenced, I've used those things.  I can promise you, there's way more than resistors in there!

 

As to the sound of tubes (deep breath), there are several simple reasons for the "tube sound".  One is that the plate impedance of the typical output tube is too high to drive much of anything, so a transformer is used.  The output impedance of the combination is lower, but still not as low as a SS amp can be.  Transformers can be shockingly good, or not so much.  The quality of the transformers relates heavily to price and size.  The combination of output tubes, biasing, output transformer design, where you get your negative feedback from and how much, all of that is a study in compromise, with the goal being flat response, low distortion and adequate power.  It does take a bit of deft engineering, and high quality (expensive) parts to make a tube stage behave really well.  And so...when compromises are made, we end up with outputs with somewhat higher impedance, somewhat higher distortion, and somewhat lower power.  Drive that into a reactive load like a speaker or some headphones and you now have a non-flat response along with those other characteristics.  And yes, it does sound and measure differently.  But it's not really fair to say all tube amps sound the same any more than saying all speakers have a flat impedance curve.  None of that is true, so the end results depend on the specific amp and load.  There are some fine points to the tube sound, but that's most of it.

 

By contrast, solid state devices are by nature lower impedance, higher current and lower voltage devices.  That means they can operate directly at the desired output voltages and power while providing a low source impedance.  Designing all of that is simpler, cheaper, and easier get better results from.  And that's primarily why people recognize a difference in tube amp sound.

 

Can you make a tube amp sound like a good SS amp? Sure, but it's not done all that often.  A lot of tube designers wouldn't want to, they like being different.  And yes, some good tube amps are indistinguishable from good SS amps.

 

Can you make a SS amp sound like a tube amp?  Sure you can, and it's been done.  And a large part of making that happen was adding an output resistor to simulate higher output impedance, though, as I said before, the final "EQ" happens not just because of the resistor, it's the load too.  But, and this is just an opinion, the reason that these ideas don't succeed has to do with the fact that there aren't any tubes to look at, see glow, feel the warmth, etc.  Big tube amps fill several sensory inputs, not just ears.  And all of that input has an undeniable effect on perception. 

 

There's also a historical reason for some people's preference to tube sound.  In the early days of solid state, there were some really bad designs.  Some of them were really awful, noisy, dull sounding things, mostly because engineers hadn't figured SS out yet.  It didn't take long for those days to end, but during that time tube amps did sound much better.  I remember owning tube gear, then building a solid state amp and wondering why it didn't sound nearly as good.  But it wasn't magic, test gear showed very poor performance from the SS amp.  Anyway, the idea that tubes are better got some pretty deep roots pretty early on, and it's been hard to shake. 

 

I once had a chance to run a bunch of tests on a Marantz 8B tube amp.  It was pretty impressive driving a load resistor, but then I drove it into a complex load.  I wish I still had the response plots, but they were NOT flat, and quite dependent on the load.  Don't know why I mention that without the pictures...but oh well, it was a long time ago.

post #28 of 35
Quote:
Originally Posted by xnor View Post

Here's a video that shows the real purpose of tubes: tongue.gif

 

post #29 of 35

yeah, saw that the first time.  What's it have to do with tubes? Not a tube in the shot anywhere.

post #30 of 35
Quote:
Originally Posted by Steve Eddy View Post

What exactly do you mean by "data"? If you mean something like a distortion plot, you can't copyright something like that.

I'm not a lawyer, but: "Reproduction in whole or part is prohibited"..

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Tube buffers, "tube sound" or just gimmick?