I find the distortion from the dome tweeters in my Tannoy Mercury speakers is the main frustration in my system. Coles made a supertweeter I used to like and the Celestion HF1300 was a pretty good unit too.
"Effects box" is a pejorative used by the accurate amp brigade to relegate SET an PP tube lovers to a lower echelon. Heavens to Murgatroyd, let the SET set have their enjoyment too. The Hiraga people have their preferred sound too.
You cant argue about taste, it is a waste of time.
I am glad Hugh likes the Chinese SET amp. I am v curious to build one.
Who has a recommendation for a good OT ?
Trouble is shipping iron to Oz is SO expensive.......sigh
Totally agree Bonsai, as in the workplace and business it just gets down to personalities. And Mr Nelson Pass is one such eminent and influential personalitywho IMO has made a substantial contribution to audio.
And so has Otala who is disliked by some subjectivists !!
Member
Joined 2009
Paid Member
I suggest that to Hugh many years ago - he said that this doesn't work so well for commercial sales, but I agree that for DIY it has merits. Usually it's achieved through the use of a tubed pre-amp and clean power amp. I like that approach too and you'll see many hybrid amps based on this approach. My approach to a hybrid is different though, only I haven't built it yet!
Nice straw, actualy it s the amps that are under critic not the people who loves thoses designs...
That said a high THD/IMD can be pleasing to some but for sure it s no more HiFi, and that the sound can be enjoyable wont change this fact, indeed people would be more inspired to invest 100$ in one of the item below, sound will be as "good" as SETs but with any amplifier, good or bad..
An externally hosted image should be here but it was not working when we last tested it.
I don't quite get why the big deal whether something is "hifi". Do you get a prize when your stuff gets that word applied to it? Fidelity to a very limited flawed recording system? Why lose sleep?
I guess that you would perceive more accurately what is the deal if asked to use a TV whose colours are distorted, sure that it can be pleasing for whom only know grey industrial cities environments....
The prize is the satisfaction to have an accurate reproduction, wich is not possible with badly calibrated items, and certainly not with amps whose calibration is not as accurate as a paltry 0.1%.
There are tons of records which are of very good quality, much better than what we thought when listening to their vinyl variants, despite being recorded with tapes they are still too accurate for a SET amplifier...
Have you calibrated your tv? If not, which of the multiple picture settings do you use? There are usually several in the menu, Cinema, Sports, color saturated...
You won't have accurate reproduction, not from stereo. A good illusion, yes, but it is not accurate because two soundsources cannot accurately reproduce a sound field since very little of it is contained in the two signals. The choices are good illusion or the precisely driven signals obviously playing from two loudspeakers. To do more you need to play with room acoustics and your speaker placement s, unless anechoic playback sounds good too you. Once you've done that, what precision is there to be fussy about? 0.1% distortion is a big deal, yet room enhanced sound within maybe 10dB of the anechoic is fair game? Doesn't add up.
There are good records but they have captured only so much and can only recreate at best an illusion.even if your amp and speakers are sort of flat and your amps distortion is one part per trillion. I'd prefer to have tone controls or whatever else works to make the best experience I can without worrying too much about imagined purity.
I think you're getting hung up on terminology. If a box of gain changes the sound, that's a different thing than a box of gain that is transparent. In the world I travel in, users refer to the former as "effects boxes" and that's how they're marketed and sold. If the same sound-altering function is performed in software rather than in amplification, the term "plug-in" is used.
I am in violent agreement with you that the overall sound reproduction experience is hugely flawed, and if some people find that an effects box gives them more enjoyment, why not? I for one am perfectly happy with EQ (as an example), but prefer it to be adjustable.
Good post. Getting closer.
To abbreviate more:
"A huge benefit for audio?"
I can't see it. Not even the small frequency response deviation coming from a highish output impedance.
Hardly a speaker does have a totally flat FR. In an acheoic room that is.
Put it into a normal listening environment, and 5cm more to the right or left make more difference than any 0.2 dB FR deviation.
A huge benefit would be if the suppliers would spend more time and a part of the money they gain into improving the room, hearing position and what not. Service to the customer.
Instead, it goes into grinding away the last dB of distortion.
sottomano,
There are speaker designers, though very few that delve into placement and room treatment. I still get a kick out of this so called threshold of accuracy some here tout, and think man, if they only knew. Even the hi fi recordings are coloured, if you don't think so you need to look deeper. None of these have ever claimed anything beyond full frequency and high dynamics, or flat transfer and or recording.
Colin
There are speaker designers, though very few that delve into placement and room treatment. I still get a kick out of this so called threshold of accuracy some here tout, and think man, if they only knew. Even the hi fi recordings are coloured, if you don't think so you need to look deeper. None of these have ever claimed anything beyond full frequency and high dynamics, or flat transfer and or recording.
Colin
That's just the concept. In practice there is no such bypass listening test for a power amplifier,
since you need one in the chain to play the speakers. You could put two amplifiers in series
and switch one in and out, but that introduces complications and assumes that the extra amplifier
is perfect.
Hafler devised a null test years ago that let you hear the difference between the amplifier's
input and output, but there was a second amplifier needed in the chain for that test as well.
Some try to connect multiple amplifiers in series (with each output attenuated to unity) until
they can hear a difference between just one and the chain of amplifiers. Clearly this ignores
the amplifier's behavior into a loudspeaker load compared to a resistive load.
Last edited:
except for the "in your head" "soundstage" headphones, headphone amps can be more accurate, resolving at a fraction of the price of power amps and speakers
should be able to resolve some dimensions considerably more reliably with headphones for the listening tests of loaded power amplifiers electrical output
should be able to resolve some dimensions considerably more reliably with headphones for the listening tests of loaded power amplifiers electrical output
It's actually not difficult at all.
Analog Version: You put together a dummy speaker load (R, L, and C equivalent to the impedance curve of the target speakers), then switch the amp under test in and out of the circuit, using an attenuator between the dummy load and the amp driving the speakers.
It takes special pleading to argue that the amp driving the speakers obscures the sound of the amp under test.
Digital Version: Record the input and the attenuated output with a good digital system (something like an M-Audio 192 will outperform any power amp, the Lynx L22 is ever more ridiculously good), then compare, either with software like Bill Waslo's or by ear using something like foobar ABX.
It takes special pleading to argue that a digital recording/playback system which can't be heard in a bypass test will obscure the sound of the amp under test.
I have done the analog version quite a few times. If modern digital systems had been around in those days, that's how I would have done it.
JGH also disagreed.
Who's Right? Accuracy or Musicality | Stereophile.com
"Such a perfect component should perform its desired function—transduction, amplification or what have you—and nothing more. It should neither subtract from nor add to the input signal. It should, in other words, be the equivalent of a straight wire with gain, and when inserted into the signal path, it should produce exactly the same sound as when it is bypassed. Nothing could be more self-evident. It is possible to bypass-test a preamplifier or a tape recorder. Anyone who does this, and properly interprets his findings, should be able to do a definitive report on either one of those products. Interestingly, reviewers sometimes disagree about these, too, but that's beside the point. The point is that there is no such way of bypass-testing any other component in the entire audio system!"
And as well, with gains and polarities adjusted appropriately, second harmonic distortions, for example, can cancel. This can occur with test equipment, so an Audio Precision for example can give false readings at close to the basic generator and analyzer limits.
All of this is well below most thresholds of audibility I suspect.
When Keith Howard did his investigation into the potential preferences of the Hiraga camp versus just very low distortion, he found it necessary to generate quite a lot of distortion to be audible to him to begin with. His preference was for lower distortion, but such tests would best be repeated with multiple listeners and better controls. It's reasonable to suppose that some people are more sensitive and sensitized than others.
I was given a demo of a very good soundbar yesterday, and was asked to be as frank as possible in its assessment. It's a work in progress and uses stock transducers, with custom ones envisioned, but it was quite decent already. I managed to find some material (pipe organ) that pushed the frequency-dependent limiters and loudspeakers into clear difficulties, but the SPLs were quite impressive, particularly given the dimensions of the enclosure, before problems were clearly audible. Trying to put a single number or a few numbers to what we heard when it got into trouble would be almost ludicrous. A time domain description would be more appropriate, which could be combined with time-dependent spectra.
I'm reminded of a demo of early Internet streaming audio, which was terrible, and the presenter (at an AES local event) stated that after all, it was only 8 bits. The problems were drastically worse than the implications of a low bit rate, and I knew how good dithered 8 bit audio could sound if the time base was stable. But that it worked at all was impressive---it was a wave of the future, if not the wave.
Fortunately, knowledge didn't end with Gordon's pronouncements. 😀 He was a sharp guy, most of all because he could take on correction and new knowledge.