I tend to agree with that.
I hate you 😡
Quite possibly, but we get very much opposition along the lines 'sine waves are not music' (in itself debatable) so sine wave testing doesn't tell me how it sounds.
I would like to go around that in some way.
Jan
Not to offend, but I'd expect neither of these two devices to be transparent. Preference in this situation is then a really complex situation that probably also depends upon the music that is being played through each system. I could imagine a situation where the setup with higher distortion sounds worse with complex material and better with simple material.
I guess that's why I prefaced my statement with 'simplistic', since my goal is transparency, and I think that one can measure and design for transparency fairly easily. Designing for the most excellent coloration is a situational task, dependent upon the program material and the nature of one's aesthetic.
So, comparing these spectra is tough, except to state that neither would be likely to be transparent with a wide range of program material. Again, whether either (or both) would be pleasing depends upon the program material and one's aesthetic.
I guess that's why I prefaced my statement with 'simplistic', since my goal is transparency, and I think that one can measure and design for transparency fairly easily. Designing for the most excellent coloration is a situational task, dependent upon the program material and the nature of one's aesthetic.
So, comparing these spectra is tough, except to state that neither would be likely to be transparent with a wide range of program material. Again, whether either (or both) would be pleasing depends upon the program material and one's aesthetic.
It is rare for an equipment demo that complex music is played - often variations on a nice solo sexy female voice 😉
Almost anything sounds good that way.
Jan
I would recommend you take the Richard Clark amplifier challenge using one of these amps versus an amp that you would consider transparent. It's an easy $10,000!
A technician friend poked at me for spending $$ on the ability to measure stuff that "we cannot hear", but my thoughts on this are that the ridiculous numbers, like -150dBC or whatever distortion with sine waves, aren't to be interpreted as what we hear, but instead viewed as an overly sensitive way to predict the (probably worse) distortions that will happen with dynamic and complex program material. So, the goal isn't these subterranean numbers, and one ought not expect those magnitudes of error with real programs either, but they serve as a way to identify known problems in a way that can lead to 'fixing' the circuit.
For example, 2nd harmonic could point to an asymmetric Vin/Vout error, 3rd could point to a symmetric Vin/Vout error, and frequency dependance of HD could point to a bunch of other problems, such as simple decrease in the feedback factor at 6dB/oct. A simplification, but you get my point.
So, while these simple tests have no direct link to the audibility of distortion, I think it's important that a piece of gear passes 'the basics' in order to be considered good enough for complex, 'real' program material. I can't prove it either, but as you imply, if your goal is to make transparent gear, it's at least necessary, but not sufficient, to pass these basic tests, done to a high resolution. The rest has to be done with good design principles IMHO.
For those folks chasing the 'magic soup' of a perfect coloration, that quest is endless… not something that results from logical, time invariant, and situationally independent analysis. More power to them...
Yes, and in my former life as a recording engineer, I've decided to build that coloration into the recording itself, and insist upon a clean playback system. That way, when I want to listen to well recorded, dense rock and roll, it too will be presented in all of its splendor, with little removed and no need for enhancement.
I think that's the proper way to arrange things, and those folks who want to mess with their playback system to achieve greater artistic meaning ought to get some mikes and a recorder and have a go at making a recording. IMHO, the playback system is not the place for editorials. Sure, a 'bad' recording could be altered to good effect, but why not get some gear to specifically do that alteration, print your new version, and then listen to that through a clean, basically linear playback system.
I used to find a bit of motorhead can be good for testing systems. I figured if you could listen into the mix with that you were on the right track. Plus what other use is there for a thrash metal 12" 🙂
Ref the distortion I'm currently building up a quartet of the composite LM3886 amplifiers that Tomchr designed. Distortion wise they are at the limits of what the AP 5xx series can measure and lower than the built in oscillator. And whilst they don't have the peak current delivery some would consider vital I am building them as, for the money I don't think I can get closer to reference grade. But I have got into a number of spats when I mention that I don't expect them to have a sound, and that I don't want a sound from my power amps. In fact I will be disappointed if I can hear them other than in gross overload situations.
I think thats a long way of saying in total agreement 🙂
Ref the distortion I'm currently building up a quartet of the composite LM3886 amplifiers that Tomchr designed. Distortion wise they are at the limits of what the AP 5xx series can measure and lower than the built in oscillator. And whilst they don't have the peak current delivery some would consider vital I am building them as, for the money I don't think I can get closer to reference grade. But I have got into a number of spats when I mention that I don't expect them to have a sound, and that I don't want a sound from my power amps. In fact I will be disappointed if I can hear them other than in gross overload situations.
I think thats a long way of saying in total agreement 🙂
Bill how about putting them all in series (with attenuators in between of course) and see if you can hear a difference with that series inserted in your regular system?
Jan
Jan
Tempting, but that would delay my active project even further. Plus the current power amp is really on its last legs and I need a reliable reference to play music on before I mess with too much audibility testing. Expectation bias can be a great if you embrace it for purely personal pleasure 🙂
Basically agree with all of yours points , except ...
that simple JLH amp shows much lower generation of high ordered THD components ,
where both JLH & Arcam perform much better on 15-ohm load than on the lower 8,2-ohm ,
Amp `sound transparency` sometime mean how amp perform in range from 1Khz -20Khz and beyond , since finest music micro-details is mostly concentrated on upper sonic spectrum .
-Personally I will prefer JLH A-class amp with 16-ohm load(speaker)
Best Regards !
It may still happen. Although as part of my recovery after 20 years in the wilderness I do have many long standing perceptions to test and dismiss.
But a functional, hum free, stable full range system is needed as a starting point !
I'm going with http://hifisonix.com/error-correction/ (aka AFEC) which I can switch in and out to allow me to compare very low distortion to moderately high distortion at the flick of a switch.
Last edited:
You could build a low distortion transconductance power amp based on a chip amps current source, and compare that to the king of the hill FW F1.
(couple of freebee designs on the web, besides Mr. Rasmussen' I'm NOT frantically yelling Thunder Ute on AC/DC chords)
I could, but once you have a miniDSP in the chain and can eq things out where is the difference? Were I thinking of building a voigt pipe with a fostex/lowther full range transconductance makes a lot of sense, but I'm not. So parked until I have time to do things for Lolz (oh and money). Neither of which will happen soon.
Or in other words, want to, know I don't have time to.
And before I do anything I need a reference to start from, which I don't have.
Or in other words, want to, know I don't have time to.
And before I do anything I need a reference to start from, which I don't have.
Older.
After a couple of weeks, all she wanted to talk about was my money and what she wanted to do with it. She had my whole life planned, after knowing me for only a few weeks....🙄
I dumped her on new years eve, and took another woman out instead. I'm such a bastard.
I simulated the same 200W mosfet amp without local negative feedback around enhanced VAS (EF+VAS) and whit it. The loop gain dropped by 20 dB (80 dB and 60 dB) and THD20k doubled when local VAS NFB was deployed.
Does it possible to decide from the FFTs which one could sound better or both would sound the same?
Damir
Does it possible to decide from the FFTs which one could sound better or both would sound the same?
Damir
