Thorsten
How do you mean? That the FFTs of the input and output signals are identical, but there are audible differences? How can this be? (This seems to be the fundamental issue that this thread should be discussing.)
Are you saying that the FFT is an inappropriate tool for measuring the effects of power supply noise? Are there alternative measurements which could reveal problems of power line noise with these amps?
Hi,
No, but that the FFT of the output noise appears to shown no appreciable differences between one configuration and another. However an FFT with a multitone signal can show differences.
PSU noise will only manifest itself directly in a few cases, in most cases the problem is intermodulation with signal present.
An FFT with suitable resolution can reveal problems if the user of the system is capable of using it suitably and of provoking the problem to be measured.
Specific Measurements in themselves are not always conclusive. Any measurement simply offers a specific answer to a specifc question.
If the question is misunderstood or generalisations are made from specific results that do not support such generalisations, then the measurements may even become counter-productive.
Ciao T
No, but that the FFT of the output noise appears to shown no appreciable differences between one configuration and another. However an FFT with a multitone signal can show differences.
PSU noise will only manifest itself directly in a few cases, in most cases the problem is intermodulation with signal present.
An FFT with suitable resolution can reveal problems if the user of the system is capable of using it suitably and of provoking the problem to be measured.
Specific Measurements in themselves are not always conclusive. Any measurement simply offers a specific answer to a specifc question.
If the question is misunderstood or generalisations are made from specific results that do not support such generalisations, then the measurements may even become counter-productive.
Ciao T
If there's intermodulation or power supply noise that appears at the output, the frequency domain spectra will be different. I can't imagine not using that tool to detect power supply noise or intermodulation, tracing the source, and verifying the fix.
edit: xpost with Thorsten. I see that he's qualified his claim. Excellent!
edit: xpost with Thorsten. I see that he's qualified his claim. Excellent!
Hi,
Funny.
At the cost of another false generalisation, most of what I find the most artifical and unnatural (most hi-fi sounding) in speakers and amplifiers is american designed and made and what one might call "typically american".
Then again, some of my all time favourites (like the JBL Everest and Urei 813 in speakers and the Bel Cantor Designs 845 Amp) are also US Made and I guess "US designed" (the Bel Canto Amp was actually based on a Portugese design published in french mag...).
Just shows the wisdom of "DE GVSTIBVS NON DISPVTANDVM EST".
Ciao T
Funny.
At the cost of another false generalisation, most of what I find the most artifical and unnatural (most hi-fi sounding) in speakers and amplifiers is american designed and made and what one might call "typically american".
Then again, some of my all time favourites (like the JBL Everest and Urei 813 in speakers and the Bel Cantor Designs 845 Amp) are also US Made and I guess "US designed" (the Bel Canto Amp was actually based on a Portugese design published in french mag...).
Just shows the wisdom of "DE GVSTIBVS NON DISPVTANDVM EST".
Ciao T
It must surely be the case that the designers of audio equipment will tend to audition it using the type of music that they personally 'understand'. I, personally, would have no concept of what I 'was supposed to be listening for' in oriental music, for example. If compromises were necessary, I would no doubt favour those compromises that enhanced my own personal favourite type of music, which would no doubt tend to correlate with my own country's typical tastes.
That's where Peter Walker of Quad's philosophy would come in: start with the design criterion of aiming for a 'straight wire with gain', and then don't listen to the equipment, only check that the design is working as intended, by measuring it.
That's where Peter Walker of Quad's philosophy would come in: start with the design criterion of aiming for a 'straight wire with gain', and then don't listen to the equipment, only check that the design is working as intended, by measuring it.
Hi,
Then perhaps it is telling that I "understand" a wide rage of music.
Understanding in this case that implies being able to follow the score where such exists, understanding some of the musical theory behind the score as well as the historical and musical context, understanding the instruments, their acoustic output and radiation pattern and how they are played, not that you would want to listen to me playing drums, piano, horn or guitar - but I at least tried to learn these and attained a certain minimal technical proficiency.
I also understand how to record music, how to place microphones and how to achieve certain effects and so on. I have recorded minimalistic miked classical and early music, I have set at the mixing desk (both live and recording) for Rock Act's, but also folk music and others all the way to "Schlager" and for a while with band playing what you may call NuJazz.
Mind you, I have not made many recordings in the last 20 Years or so since I got of the music industry in disgust over the production values (or rather the lack thereof), something I should really change...
I went to classical concerts as often as others (concert tickets are now rather expensive and most amplified concerts have bad sound, so now I only occasionally go classical concerts.
When designing and voicing stuff I always refer to un-amplified music though, if a system handles that right the rest is usually not a problem.
Here an illustration of what I am trying to achieve:
The irony in this is that even "a straight wire" does not really reliably behaves as the proverbial "straight wire"...
An no equipment ever made by Acoustical Manufacturing came even remotely close to a real piece of wire and excluding their ESL Speakers, came ever remotely close to allowing a recording to sound like music.
So one may argue that it is on surface a reasonable proposition and much use as a propaganda slogan, but falls down badly in reality.
Ciao T
Then perhaps it is telling that I "understand" a wide rage of music.
Understanding in this case that implies being able to follow the score where such exists, understanding some of the musical theory behind the score as well as the historical and musical context, understanding the instruments, their acoustic output and radiation pattern and how they are played, not that you would want to listen to me playing drums, piano, horn or guitar - but I at least tried to learn these and attained a certain minimal technical proficiency.
I also understand how to record music, how to place microphones and how to achieve certain effects and so on. I have recorded minimalistic miked classical and early music, I have set at the mixing desk (both live and recording) for Rock Act's, but also folk music and others all the way to "Schlager" and for a while with band playing what you may call NuJazz.
Mind you, I have not made many recordings in the last 20 Years or so since I got of the music industry in disgust over the production values (or rather the lack thereof), something I should really change...
I went to classical concerts as often as others (concert tickets are now rather expensive and most amplified concerts have bad sound, so now I only occasionally go classical concerts.
When designing and voicing stuff I always refer to un-amplified music though, if a system handles that right the rest is usually not a problem.
Here an illustration of what I am trying to achieve:
An externally hosted image should be here but it was not working when we last tested it.
The irony in this is that even "a straight wire" does not really reliably behaves as the proverbial "straight wire"...
An no equipment ever made by Acoustical Manufacturing came even remotely close to a real piece of wire and excluding their ESL Speakers, came ever remotely close to allowing a recording to sound like music.
So one may argue that it is on surface a reasonable proposition and much use as a propaganda slogan, but falls down badly in reality.
Ciao T
Very true, and I agree wholeheartedly.
Which is why I was so insistent on the fact that this way just MY choice, no universal truth carved in stone.
You and I are living proof that even nominally opposing camps (your above comment, the fact that you have strong leanings towards, and I strong leaning away from tubes, etc) can get along just fine for, what has it been now, 10 or 11 years, since the days of tnt-audio.com?
True to my own words, other US amps I liked and remember fondly are the old Krell KSA-100, and recently, John's own Parasound HCA 1205 A. I narrowly missed out on actually buying it - bad luck.
THE one single do-it-all, have-it-all loudspeaker of my life is JBL's 4312 Monitor. I readily admit it is colored, but where it excells is in lfe - it has life pouring out of its ears, er, connectors. And there's the miraculous ease with which it goes about its job; my own speaker's moniker "1043" is in fact my own hommage to JBL, 10" for bass driver, 4" for the mid driver and "1" for the tweeter. It was jointly developed after some of my ideas on speaker upgradeability (a different story) by myself and a guy who used to make speakers for a living (why he doesn't any more is yet another story, one with a sad end all around), my pair serial number is "0001". I believe we managed to actually improve on the JBL theme, except on that one item life bubbling out of it; sure, it's got plenty of life, fire and brimstone, but it just doesn't have that JBL feeling, it's more sort of well mannered for lack of a better phrase. Whatever, the JBL is still better in that.
On the other hand, with the notable exceptions of Norway's Elektrocompaniet and Denmark's Ortofon, I have to date yet to hear a Scandinavian audio product which I would want to keep. Nada.
Even in headphones, my leanings are more towards USA. Ever since 1976, I have never been without a Koss pair of 'phones, currently Pro 4 AA Ti, and keeping it company so it doesn't get lonely is a pair of Grado SR-80 cans (used for TV mostly, but used every day). True to my regard for tradition, I also keep at hand, as I have from 1966, a pair of Sennheiser cans, currently HD 598.
Japanese audio? Well, my Yamaha CDX 993 CD player is Japanese, even if it's used as a transport only, with my Australian made real time DAC. And my old Sony 3950 tuner is Japanese, as is the Sansui AU-X701 integrated amp my son is using, which I keep for sport, really. But every day, I listen to my reVox B 760 tuner. Very hard to beat, that one.
Hi,
To me the 4312 and LE100 is where interesting speakers start. It is still too low efficiency to be really "live-like", but it does get a lot more of this than many modern HiFi Speakers.
On the other hand try a pair of well restored Urei 813, preferably the ones with the original Altec 604 Coax and not the later JBL Coax equipped version.
Correctly set up (which means soffit mounted or LF equalised, toed in at the listening position) they do "live" with ease and generally offer very low colourations, not as low as a Dynaudio Standmounter, but also not that far off and definitly lower than the 4312.
Once you have such a speaker you may also revise your view on tube gear, a pair of Urei 813 with something like old Bel Canto 845 SE Amplifier can do "virtual reality" on a level unattainable for conventional speakers with big SS Amp's. On the other hand with even with JBL 4312 it would probably be a tossup for me between tubes and transistor Amp's, if tubes probably quite big, not 300B SE...
Funny. My own current speakers follow a similar recipe. But instead of the JBL cramping I use a large floor standing enclosure with Avalon/Stealth chevron's.
Bass is 10" generally quite conventional, but massive magnet and high BL for the moving mass.
Mid is an overgrown 5" fullrange driver (6" frame, but long throw surround so the cone would be more at home in a 5" driver, ND Magnet).
Treble is a circular magnetostat (kind of like the old Infinity EMIT-R), first order series crossover at 300Hz and 3KHz...
Around 89dB/2.83V/1m, 6 Ohm impedance, mostly flat (except LF Box tuning peaks). Quite lively. Closer to a conventional HiFi speaker than most of what I have used, but reasonably tube friendly and more to my taste than most HiFi Speakers...
One of these days I'll build something bigger again, maybe 15" Bass, 8" Mid and a large Hornloaded Magnetostat...
Ciao T
To me the 4312 and LE100 is where interesting speakers start. It is still too low efficiency to be really "live-like", but it does get a lot more of this than many modern HiFi Speakers.
On the other hand try a pair of well restored Urei 813, preferably the ones with the original Altec 604 Coax and not the later JBL Coax equipped version.
Correctly set up (which means soffit mounted or LF equalised, toed in at the listening position) they do "live" with ease and generally offer very low colourations, not as low as a Dynaudio Standmounter, but also not that far off and definitly lower than the 4312.
Once you have such a speaker you may also revise your view on tube gear, a pair of Urei 813 with something like old Bel Canto 845 SE Amplifier can do "virtual reality" on a level unattainable for conventional speakers with big SS Amp's. On the other hand with even with JBL 4312 it would probably be a tossup for me between tubes and transistor Amp's, if tubes probably quite big, not 300B SE...
Funny. My own current speakers follow a similar recipe. But instead of the JBL cramping I use a large floor standing enclosure with Avalon/Stealth chevron's.
Bass is 10" generally quite conventional, but massive magnet and high BL for the moving mass.
Mid is an overgrown 5" fullrange driver (6" frame, but long throw surround so the cone would be more at home in a 5" driver, ND Magnet).
Treble is a circular magnetostat (kind of like the old Infinity EMIT-R), first order series crossover at 300Hz and 3KHz...
Around 89dB/2.83V/1m, 6 Ohm impedance, mostly flat (except LF Box tuning peaks). Quite lively. Closer to a conventional HiFi speaker than most of what I have used, but reasonably tube friendly and more to my taste than most HiFi Speakers...
One of these days I'll build something bigger again, maybe 15" Bass, 8" Mid and a large Hornloaded Magnetostat...
Ciao T
All good stuff. But I'm guessing that much of the equipment used for professional recording is designed for robustness and functionality, without any provision for 'boutique' audiophile refinements. How do you cope with all those long cables of dubious origin, amplifiers with limiters, leatherette-covered chipboard speaker cabinets, twenty op amps and untold nicotine-coated potentiometers in the signal path, everything plugged into cheap mains extension blocks, spurious interference from air conditioners and traffic outside, and endless gaffa tape? (I know it's not all like that, but the average musical performance or location recording, or even studio recording, does tend to have some of those elements in it.)
Some people might doubt that it's possible to accommodate all of that in one's brain when when listening to it at home, and yet pick out the -100dB of imperfection in the final amplifier when listening to the recording. Is it really possible to distinguish differences in the final 0.5m of boutique interconnect, against the 100m of 'screened cable' that preceded it in the signal path?
Sell my revered Marantz collection, after waiting for 27+ years to get my hands on it?
You're outta your mind, Tex!
Bad T-bone for lunch? 
Got your point, though. Don't mind us, friend, we like to take a break from the main subject here and there and get carried away a wee bit. Thorsten and I go back like 10 or 11 years, and some of the other guys here make for a hell of an interesting read, but you can't be dead serious ALL of the time. It gets boring.
BTW, I don't sell in North America since 2003.
Last edited:
Has anyone heard or read the JA "Heyser" lecture recently delivered at AES? I was, I readily admit, from what I had heard about it, somewhat disposed to be critical of it at the outset. However, having gotten, finally, a pdf last night from a friend, it's actually rather fascinating, and has a bearing on the thread topic. I recommend getting hold of it. John ends up wrestling with some ideas about mental representations that are eerily close to some of my own speculations.
Brad
Hi,
I have it and have read it, I can make it available to anyone looking for it but wishes not to create extra work for John.
And i agree, it is very interesting and illuminising.
Ciao T
I have it and have read it, I can make it available to anyone looking for it but wishes not to create extra work for John.
And i agree, it is very interesting and illuminising.
Ciao T
Not all recordings are made with utilitarian "pro" gear. Some are made with equipment whose specs and SQ would please the most ardent "audiophile". Many smaller labels (and some larger ones) use minimalist miking techniques and a purist audio chain. Then there is the the BBC. Regards from a retired "minimalist" recording engineer and producer.
I understand it differently: the author knows that some designers try to minimize wrong errors while don't hear that they overlook something else more significant. We discussed already hundred pages ago how improvement of DF more that 10 does not matter, and THD less than 0.01% on full power on 1 kilo Hz frequency does not matter, while dynamic distortions added as the result of such heroic minimization start dominating.
Last edited:
We already discussed fifty pages ago that "adding nice sounding distortions" is a myth. I especially designed and built such device with a knob, but nobody wanted to add "nice sounding distortions". I knew already they don't exist, and wanted to demonstrate to other people on the experiment, how wrong is this myth.
Well, nobody wants to do it knowingly. It's like wine, where people say they want dry, but if you add some residual sugar without telling them, they like it better and the wine sells like crazy. Do you remember a fellow who popped up here a few months ago extolling the virtues of an aftermarket effects box but got quite angry when it was suggested that the coloration (EQ and distortion, in the case of his favored gadget) was what he liked?
It's well known in the food service industry- add sugar and MSG, people will love it as long as you don't make it so overtly sweet that they figure out the trick. If you tell them that's what you're doing, they'll hate it.
As John Dozier pointed out, there have been (and continue to be) recordings made with some wonderful gear, and not beaten to death in the post-processing --- even if these amount to a tiny fraction of recordings.
But also: even with ghastly source material, one can become very familiar with it. So it may not be such a stretch that the alterations attendant on different playback systems will be audible.
One of the things I've been amused by over the years are the accounts of listeners and reviewers who say they heard things in a given recording that they'd simply never heard, before playback system X. Well, this is often taken to be an endorsement of X, rather than considered potentially as a linear or nonlinear and/or time-variant distortion by X, which just happened to make detail Y more audible.
Does anyone remember the story of the Aphex Aural Exciter? My understanding is the company founder built a Heathkit, and he screwed up, with one channel horribly distorting. But he found that he liked a little of it mixed with the working channel, hired an engineer to analyze what was going on, and then decided to make a product around it.
As time went on the company de-emphasized that initial product, and when at the behest of a recruiter who needed money I interviewed with the then-chief engineer, he seemed almost embarrassed about it.
- Status
- Not open for further replies.