No, if I want to measure frequency response of THE AMP I load it on plain dumb most linear resistor. Period. Otherwise all I want to see would be hidden behind errors of that "real speaker or dummy circuit".
People who claim that measurements and listening tests don't correlate don't have a damn idea what and how to measure.
Indeed. But they also confirm my hypothesis about the lack of correlation between Stereophile reviews and reality. 😀
Why would I spend $350,000 on an amplifier to drive a resistor?
If it's that obvious, then why use a load at all then? Why not use a dumb open circuit to measure THE AMP?
It does not heart the amp if it drives the resistor in the designer's laboratory! 😀
Like, your favourite shirt is not bad because the designer used straight ruler calibrated in equal millimetres to measure it on a flat table.
If it is some kind of a voltage amp, I would follow your advice. If it is a current amp, I would short it's output measuring magnetic field around the shorting wire. But if it is a power amp, please forgive me if I use a resistive load to measure it's frequency response measuring unknown voltage drop on very known resistance. It is the same, as to use a flat straight ruler measuring the shirt. And I would put the shirt on the dumb flat surface before measurement.
Last edited:
Peruse the collected works of Jonathan Scull and Michael Fremer. If you are still sane at the end of that, you'll certainly be stupider than when you started. Not quite as stupid as you'd be for the equivalent time reading Six Moons, but nonetheless, stupider. 😀
Into a single resistive load, you can't separate out the amplifier's 'internal' frequency response from the effects of its notional output impedance (which will not just be simply resistive). To separate the two, you would have to test into multiple 'dumb' resistive loads. The review could then attempt to model the amplifier's series output impedance (as L, R and C) as well as its notional 'internal' frequency response - but of course there would be some interaction between the two which might be difficult to specify.
How should the casual reader of the review respond to this wealth of information?
Alternatively, the review could simply take the approach of measuring amplifier output into a standardised representative speaker load, which at least would instantly convey to the reader how much bass boost/treble loss he should expect to hear, and also tell him whether he needs speakers designed to compensate for the amplifier's 'warm tone'.
Reality Check: Essentially ALL you would have to do is to insert a 4 ohm resistor in series with the output of MOST power amps, and you would have essentially the same frequency response with ANY power amp. Try it yourself, and listen to the difference. However, that is NOT what makes that expensive amp sound good.
Is it the high level of high order harmonics that make it "sound good"? The ringing? The terrible transient response? Would you feel comfortable releasing an amp like that? You'd feel comfortable taking the money (hell, so would I!), but wouldn't you be a bit... embarrassed?
Stuart, we speak now about accuracy of frequency response measurement reflected in the article. How the rest was measured, is totally different question that has no answers. Either the author has obvious hearing problems, or he does not know what and how to measure. It is already established fact that he does not know how to measure frequency response. The article is totally useless for our discussion, it neither proves, nor denies nothing. It is irrelevant and any other quotes of the article, of the author, and of the magazine are useless. They have no credibility at all, except may be pictures that reflect real things, until proven that they are also edited.
The measurements (and excuses) were written by someone different than who "reviewed" the product for its... ummmm... sonics.
I see nothing wrong with the frequency response measurements. A dummy load which simulates a loudspeaker is quite a valid way to see how the amp reacts to real-world loads. Resistor loads are necessary (and done here as well) but not sufficient.
I see nothing wrong with the frequency response measurements. A dummy load which simulates a loudspeaker is quite a valid way to see how the amp reacts to real-world loads. Resistor loads are necessary (and done here as well) but not sufficient.
Well, if an amp changes frequency response with load, that's kinda important to know. Or looking at it another way, if a speaker sounds one way with a low impedance low distortion amp, it will sound different with a high impedance high distortion amp. Wouldn't you agree that this is something a purchaser should consider? Yeah, you or I could look at the source impedance and the impedance curve of our speakers and draw the appropriate conclusions, but the dentists and hedge fund managers who read Stereophile don't know an ohm from their elbow.
nah, but they have nice pictures of new stuff
and its not all stupidly expencive 😉
a claim here still rings in my head...why expencive gear sounds good
I would have preferred ...why good designs sounds better
but at least its good to know we can build stuff that competes with anything
Expensive gear sounds good to those with strong religious beliefs - like 'one needs to spend a lot to get good sound'. Those kinds of beliefs. If a person's not as religious then they'll likely hear the benefit of a good design no matter what the price.
Just today I learned that the designer of an uber-expensive DAC is disputing the review there of the Metrum Octave DAC - which by their standards is rather cheap 😀
Just today I learned that the designer of an uber-expensive DAC is disputing the review there of the Metrum Octave DAC - which by their standards is rather cheap 😀
- Status
- Not open for further replies.