it's $10 at Farnell.
regarding feedback vs no-feedback, here's a wild theory. what if most audiophile music happens to have been be mixed and mastered with high-distortion, low-NFB amps (very likely)? would that not be an explanation for the reported dull sound of low distortion, high-NFB ones at the playback side?
All wrong, mr_push_pull.
First and foremost, I am NOT a SR and bandwidth looney, I simply feel that they should be pushed to a certain level in a certain way so that we improve the sound of the amp. Namely, a volateg slew rate of 40V/uS is quite enough for real world applications, and that number is about double of what is really enough, just to be on the safe side. If it comes out more, fine, I won't complain, but I'll never go for specifically that.
As you may remember, the decade 1975-1985 was spent in Slew Rate Wars among the manufacturers, notably Sansui, Kenwood and Pioneer. They would have had us believe that a high slew rate meant better sound. Time has proved them wrong, of course, as it probably always will each and every time any one single factor in an amp is blown out of proportion.
And secondly, in case of headphones, the actual, real world requirements are at least an order of magnitude below power amps, simply because our output voltage requirements are so much smaller. Remember, headphones very typically need like 1 mW to shout at 96+ dB. You work out what that 1 mW is for 30, 100 and 600 Ohms headphones. Then for safety's sake, multiply the result by 10, just to be safe.
And lastly, I have already built 3 decicated headhope amps, one in bipolar (BJT) technology, another in FET/MOSFET technology and the third in (gulp!) tube technology. BJT and FET/MOSFET are DC coupled, fully complementary designs, and so forth. Tube job has zero global NFB.
Thus, I am in no rush to experiment on that front, I think I did just fine as it is.
I think dull amps are ones where only the oscilloscope listened ? I suspect most good measuring amps can be transformed by a little use of intelligence and pushing the boundaries . I almost feel guilty about saying 70 kHz - 3 dB . Someone will listen with their oscilloscopes and fix it at that . I was saying I wouldn't be upset if it took that to make an amp nice to use .
well, here's a view from a computer engineer with a tiny bit of analog knowledge.
please don't take it the wrong way, it's just my state of mind when I read some of the discussions.
what I see from some of the amp designers (and is very evident in this thread) is an almost complete unwillingness to try and correlate sound with what is known about the human hearing and with some objective measurements (whatever they are). it's like the mere act of analyzing a signal and trying to understand why it sounds how it sounds is a sin in itself.
maybe if I were an EE, I don't know... but I'm not and I always look at amps as boxes that do something to a signal. I often read statements as (I'm just making something up but I'm sure you get the idea) "I feel that a certain amount of input pair degeneration sounds like this or that" but I rarely see stated "and maybe it's because it does this or that to the output signal". I have the gut feeling that EEs almost hate looking at things from a system perspective and love to ignore the bigger picture
maybe it's just a different mindset that comes with the job? maybe it's because there's a certain enjoyment in fiddling with that kind of stuff? I don't know. I'm not saying it's a bad thing, but from my perspective it's intriguing.
and it's even funnier because little of what the title suggests is actually done in this thread
sorry if this sounds blunt, maybe it's better to just leave things as they are and not stir the waters.
please don't take it the wrong way, it's just my state of mind when I read some of the discussions.
what I see from some of the amp designers (and is very evident in this thread) is an almost complete unwillingness to try and correlate sound with what is known about the human hearing and with some objective measurements (whatever they are). it's like the mere act of analyzing a signal and trying to understand why it sounds how it sounds is a sin in itself.
maybe if I were an EE, I don't know... but I'm not and I always look at amps as boxes that do something to a signal. I often read statements as (I'm just making something up but I'm sure you get the idea) "I feel that a certain amount of input pair degeneration sounds like this or that" but I rarely see stated "and maybe it's because it does this or that to the output signal". I have the gut feeling that EEs almost hate looking at things from a system perspective and love to ignore the bigger picture
maybe it's just a different mindset that comes with the job? maybe it's because there's a certain enjoyment in fiddling with that kind of stuff? I don't know. I'm not saying it's a bad thing, but from my perspective it's intriguing.and it's even funnier because little of what the title suggests is actually done in this thread
sorry if this sounds blunt, maybe it's better to just leave things as they are and not stir the waters.
Last edited:
Quite seriously, Nige, I pity those who "listen" by 'scope only. The poor souls have no idea of what fun they are missing out on.
I can't speak for others, but for me, the greatest fun starts once you sort it out electrically and it works with no glitches. Then come the freal fun, auditioning it, trying things out, playing with various materials, trying to catch it out and if you do, resolving the problem.
If I had the time, I'd spend literally months playing with it and I'd never be bored. To me, itr's fun to discover how changing say cap materials from polypropylene to polyethyline to polycarbonate will impact the sound, if at all. Discovering where to use standard resistors, and where to use Dale resistors because it sounds that eeny weeny bit better.
To me, that's the real challenge.
well, here's a view from a computer engineer with a tiny bit of analog knowledge.
please don't take it the wrong way, it's just my state of mind when I read some of the discussions.
what I see from some of the amp designers (and is very evident in this thread) is an almost complete unwillingness to try and correlate sound with what is known about the human hearing and with some objective measurements (whatever they are). it's like the mere act of analyzing a signal and trying to understand why it sounds how it sounds is a sin in itself.
maybe if I were an EE, I don't know... but I'm not and I always look at amps as boxes that do something to a signal. I often read statements as (I'm just making something up but I'm sure you get the idea) "I feel that a certain amount of input pair degeneration sounds like this or that" but I rarely see stated "and maybe it's because it does this or that to the output signal". I have the gut feeling that EEs almost hate looking at things from a system perspective and love to ignore the bigger picture
and it's even funnier because little of what the title suggests is actually done in this thread
sorry if this sounds blunt, maybe it's better to just leave things as they are and not stir the waters.
No offence, mr_push_pull, but in case of this part of the forum, I'd say you are wrong. Dead wrong.
Just as an example, do you seriously think that John Curl went into all audio history books by just observing a 'scope? Trust me, that is simply not so, rather it's because he watched it for a while, and after that, he used mostly his ears.
If an amp sounds dull but is not dirt cheap, then the chances are that its designer listened via a scope, or is musically deaf, the remedy being that in either case, he should be fired.
If an amp, any amp, shows any spirit, perk, liveliness and such like, you can bet your bottom dollar that designer listened. Ultimately, any decently designed amp will reflect the hearing and the tastes of its designer, or the lead designer of a team.
The bottom line is, you're being too harsh.
More simply put, take it back, son, or I'll start gathering a posse.
see, that's what I meant.
I wanted to say it for a long time but refrained from doing so because these discussion downspiral very rapidly in a very predictable way. it's one of those cases when saying and not saying it is equally as bad
and your reply confirmed it. to me it looks like the subjective camp is kind of stuck. ok, if an amp measures perfectly but it sounds bad, well, it sounds bad. and my amp measures horribly while the former measured much better. guess why I own the current one?
but, gee, I can't stop but feel that the subjective/empirical camp really needs to move on and actually try to correlate sound with something palpable. it looks to me they simply don't want to do that they can but they won't it's what 10 years on this forum taught me.
waiting for the posse
I wanted to say it for a long time but refrained from doing so because these discussion downspiral very rapidly in a very predictable way. it's one of those cases when saying and not saying it is equally as bad
and your reply confirmed it. to me it looks like the subjective camp is kind of stuck. ok, if an amp measures perfectly but it sounds bad, well, it sounds bad. and my amp measures horribly while the former measured much better. guess why I own the current one?
but, gee, I can't stop but feel that the subjective/empirical camp really needs to move on and actually try to correlate sound with something palpable. it looks to me they simply don't want to do that
they can but they won't it's what 10 years on this forum taught me.waiting for the posse
exactly. it looks to me if one day the perfect amp arrives, which simply renders a scaled copy of the input, someone will still say it must be fiddled with in order to be made "right". how can that be?
So can we look at the schematic of an exciting, spirited amp and point to the 'perkiness components' that the 'scope-watching philistine would have missed out, or the special dullness components he would have added? There's only so many places in the circuit to add or take away the magic.
You conjure up an image of the amp designer as an artist or sculptor deftly wielding his soldering iron, adding a small cap here, chipping off a resistor there, standing back and listening for a moment, "No no! The amp has lost some spirit! I need more capacitance but my rent is due. The amp comes first. If I am evicted from my garret, so be it."
What load is the amp driving during all this creative activity? Is the extra liveliness being conjured up at the expense of stability? What music is the designer listening to? Dave Brubeck's Take Five, Girl and Guitar? (Amplifier designers are obviously well known for their musical literacy and radical tastes that push systems to their limits.)
No, I suggest that the average amplifier designer doesn't know the first thing about music, nor have any taste in it. Like most audiophiles they will be hung up on some small aspect of the overall picture and in the end the acceptance of their work by others will hinge on marketing and expectation bias.
Good point. I've a few 'audiophile' recordings, which happen to be throwaways from the local library, I don't buy them new,
... and they are generically dull. If they had been monitored through those type of seups while mastering, it would make sense ...
Yes, except I would change 'amp' to 'system'. Every setup I've listened has convinced me that a brilliant link in the chain can subjectively help a great deal, but ultimately the sound fails at some point to deliver the goods if there is a particularly weak link somewhere.
I was having another fiddle with my "ridiculous" UULE yesterday, Aldi TV plus Blu-ray player, seeing if I could push it a bit higher without doing any internals and was pleasantly rewarded. Big, deep, sound, very close to being 'invisible' in the closing of one's eyes and pointing sense - the interesting thing is that one can push the volume right up so the cabinet is rattling and buzzing badly, and the deep bass notes sound very gluggy; yet this overt distortion doesn't irritate or get in the way of tuning into, enjoying the musical experience.
This is to point out, yet again, that an abysmally basic system can perform remarably well in key areas, so long as critical weaknesses are eliminated or bypassed in some way ...
By all means, you gentlemen keep your dull sounding amps if that's your thing, but be aware that "dull" is also a qualitative statement. Being dull is also a sound characteristic.
Let me try to explain. All else being equal, if one amp leaves me completeley unmoved during some personal musical test pieces, which I know should be wild, dynamic and exciting, to me that's a poor amp, but if I catch myself tapping my foot, then it has potential.
In my view, reproducing the emotion of music is the hardest trick to pull off. Far too many amps don't even know what that is. Unfortunately, just as many listeners don't either.
If the musical content has fire and brimstone, I want to hear it, I don't want it all flattened out and mish-mashed into dull. This is one of the reasons I dislike typical British audio, quite simply, it sounds dull to me.
Let me try to explain. All else being equal, if one amp leaves me completeley unmoved during some personal musical test pieces, which I know should be wild, dynamic and exciting, to me that's a poor amp, but if I catch myself tapping my foot, then it has potential.
In my view, reproducing the emotion of music is the hardest trick to pull off. Far too many amps don't even know what that is. Unfortunately, just as many listeners don't either.
If the musical content has fire and brimstone, I want to hear it, I don't want it all flattened out and mish-mashed into dull. This is one of the reasons I dislike typical British audio, quite simply, it sounds dull to me.
Oh yes, Frank is right. Amps can sound dull in the wrong system, like it or not, we are in fact searching for system synergy.
UI am fortunate in that respect because my speakers are such an easy load to drive that even humble supermarket amps are free to do the best they can. It widens my amp choice tremendously.
UI am fortunate in that respect because my speakers are such an easy load to drive that even humble supermarket amps are free to do the best they can. It widens my amp choice tremendously.
I'm trying to find the classification for that type of fallacious argument: if you don't agree with some opinion on a technical matter then that must mean you are an ignorant philistine or perverse in some way. Agree, and you're one of the boys without a stain on your character.
List of fallacies - Wikipedia, the free encyclopedia
Is it the old chestnut of ad hominem, or is there a more accurate classification?
So is this reproduction of emotion a function of simple design parameters e.g. by limiting slew rate, the emotion is killed automatically? Or is it hidden somewhere in the details of the design? As I asked previously, can we look at an existing amp with built-in emotion, and spot the 'emotion components', or point to areas of the circuit that maintain stability while allowing the emotion through? Regardless of whether meaurements after the fact can reveal the emotion, what in the design is responsible for reproducing or killing the emotion? Funny that amplifiers fail to pass on the emotion but never fail to pass on, say, the woodwind.
"Here's the classical music section (beautiful oboe network there) and these large devices are for rock music. We include a small speech circuit for completeness. Moving onto the more abstract areas, we have ambience, excitement and a very large emotion circuit with SOA protection; we're very proud of our emotion circuit. For the British market we can switch in a conventional dullness module." etc.
Last edited:
I was told by a company's technical rep that a typical amplifier might take 18 months to design ( sorry I forget who ) . Very early in the project the amplifier would meet a design specification . Over the months it would evolve sonically . Then the spokesman very proudly said . The finished one and the prototype measure essentially the same and we would insist on that . They sound totally different . His best comment was that if the technical version had been OK a year of development would have been avoided .
His final words were . Specs we do for the Germans , power for the Americans and sound for the Brits . When I asked about the Japaneses he said they would follow the Brits but have a slight preferences over bandwidth ( wider ) . He also said ultimately it is the sound that sells equipment , the specs guide people to the product . He felt his company could engineer good products whilst still chasing a specification war . One could almost say the specifications were a setback that the designer had to live with . I am not advocating horrid errors of design when I say that . Just that we never discount any idea without trying it .
Real music is very odd . Take Oxfords Sheldonian theater . It has a sound rather like a zero feedback valve amp . A glorious sound with loose bass which is a bit lumpy . I know it is a bit wrong but it works very well . Same orchestra, same night ,different music . If a variation in sound like that was heard in a blind hi fi test one would have to believe different equipment used . The only difference was interplay of timing . The difference is like 10 kHz and above is missing , then it is back with a different piece of music . Most Dull Fi makes it sound about the same all the time (stifled ) . Ah that is it being correct I hear , no it isn't . I absolutely insist that no amplifier need sound that way . The bandwidth , gain , coupling , time constants , VAS cap and size of PSU caps need to be listened to . With average test gear most of us own I doubt any measured change will happen . I don't count it was 0.008% and now it is 0.005 % as my goodness with a change that large no wonder it sounds better . Many are guilty of that I feel . If owning an Audio Precision test set doubtless the game is easier ? One of my amps was measured on one ( I didn't know ) . It was one of the lowest distortion amps the guy had ever measured . I am told it gave him sleepless nights as it was impossible what he saw . As my friend said who took it to him , his training stopped him seeing possibilities . The amp for all the world sounds like a valve amp with rock and roll ability . Only 8 devices per side and 100 watts 8R . The amplifier is absolutely accurate . It is fast , open , sweet , very wide bandwidth . It lacks a bit of weight which can be cured . It is even reliable ! Does it sound like the Sheldonian ? Not exactly . The closest commercial amp would be Goldmund ( Out of H H ) .
The strangest part of this is speaker problems do not bother me . They are good or bad . Amplifiers do as there is no easy explanation as to why one descent design is trounced by another descent design . My recent valve design has told me where I should look next . If I was to use it with an electrostatic unit directly and use a beefed up versions of my low distortion amp to drive twin subs I might have something worth getting excited about .
His final words were . Specs we do for the Germans , power for the Americans and sound for the Brits . When I asked about the Japaneses he said they would follow the Brits but have a slight preferences over bandwidth ( wider ) . He also said ultimately it is the sound that sells equipment , the specs guide people to the product . He felt his company could engineer good products whilst still chasing a specification war . One could almost say the specifications were a setback that the designer had to live with . I am not advocating horrid errors of design when I say that . Just that we never discount any idea without trying it .
Real music is very odd . Take Oxfords Sheldonian theater . It has a sound rather like a zero feedback valve amp . A glorious sound with loose bass which is a bit lumpy . I know it is a bit wrong but it works very well . Same orchestra, same night ,different music . If a variation in sound like that was heard in a blind hi fi test one would have to believe different equipment used . The only difference was interplay of timing . The difference is like 10 kHz and above is missing , then it is back with a different piece of music . Most Dull Fi makes it sound about the same all the time (stifled ) . Ah that is it being correct I hear , no it isn't . I absolutely insist that no amplifier need sound that way . The bandwidth , gain , coupling , time constants , VAS cap and size of PSU caps need to be listened to . With average test gear most of us own I doubt any measured change will happen . I don't count it was 0.008% and now it is 0.005 % as my goodness with a change that large no wonder it sounds better . Many are guilty of that I feel . If owning an Audio Precision test set doubtless the game is easier ? One of my amps was measured on one ( I didn't know ) . It was one of the lowest distortion amps the guy had ever measured . I am told it gave him sleepless nights as it was impossible what he saw . As my friend said who took it to him , his training stopped him seeing possibilities . The amp for all the world sounds like a valve amp with rock and roll ability . Only 8 devices per side and 100 watts 8R . The amplifier is absolutely accurate . It is fast , open , sweet , very wide bandwidth . It lacks a bit of weight which can be cured . It is even reliable ! Does it sound like the Sheldonian ? Not exactly . The closest commercial amp would be Goldmund ( Out of H H ) .
The strangest part of this is speaker problems do not bother me . They are good or bad . Amplifiers do as there is no easy explanation as to why one descent design is trounced by another descent design . My recent valve design has told me where I should look next . If I was to use it with an electrostatic unit directly and use a beefed up versions of my low distortion amp to drive twin subs I might have something worth getting excited about .
Emotion is killed when our hearing system has to work too hard to digest the musical message. An analogy is listening to a live talk by a real person, no PA, just pure, acoustic speech: one person is nearby, has no trouble hearing every word uttered, picks up all the nuances in the voice and is deeply moved by what the person is talking about - supporting some cause to help alleviate human suffering; another person somewhat further away, is near some people who rudely keep talking amongst themselves, and is struggling to follow the precise flow of the words of the speaker - this people ends up quite irritated by the exercise, and in fact leaves before the end of the talk, because he's had enough.
This is like the difference between a system working well, and another going through the motions - what was fundamentally replayed is not different, but in the latter case too much lack of clarity, and offputting 'noise' was also present ...
There is an easy explanation (for at least some cases), but some people get offended when 'euphonic distortion' is mentioned. When it is obvious (e.g. typical Chinese low voltage 'tube buffer', for adding 'tube warmth' to a 'nasty' 'sterile' 'dull' SS system) it is easy to spot and laugh at. When it is more subtle it is easier to deny.nigel pearson said:
I am not saying that all sound preferences can be explained in this way, but I am quite convinced that some can - possibly many can. The problem is that this widespread phenomenon gets in the way of finding what engineering changes can lead to genuinely better sound.
- Status
- Not open for further replies.