John Atkinson a écrit :
Howard Ferstler wrote:

Remember, I mainly deal with technical issues when I review
recordings, and not performance.


Exactly the point of my little joke, Mr. Ferstler. Perhaps Lionel
will take note of your admission :-)

I take note... ;-(

John Atkinson a écrit :
Lionel wrote:

Lionel a écrit :

Lionel a écrit :


In , Lionel wrote :



In .com,
John Atkinson wrote :

Howard Ferstler wrote:

it is a rare performance indeed that I do not find
satisfactory.

The musician community lets out a pent-up sigh of relief.
Howard Ferstler rarely finds a recorded performance that
he doesn't find "satisfactory." Such perspicacity -- such
hubris!

May we understand that Stereophile publishs mostly reviews
of "satisfactory" audio devices ?


Indeed it does, Lionel. I have explained at length on this newsgroup
what that is the case.


I was afraid that your magazine could be different from its
French cousins.

In this case is there any difference with Mr Ferstler
attitude concerning "recorded performance" ?


Indeed there is, Lionel. I guess my attempt at humor fell on fallow
ground. :-(


Not so fallow, dense perhaps but not fallow.


Perhaps you should look up the meaning of the word "hubris."

Yes.

I disagree. Even though the music may be excellent, it's better to
say, "The musical performance is excellent but the recording bites".
There is a lot of fine music which is simply not well recorded. If a
record gets an honest review like that the musicians may get another
shot under different recording practitioners.

On the other hand..the true fan transcends fandom and has a go at it
himself at some point. I'm sure there's some group you'd like to
record. Get a couple of mics and go for it.

The key to level matching is calibration. One necessary element,
lacking in most high end systems, is a precision attenuator that can be
set to a known, repeatable position. Others include calibration and/or
characterization of the entire electronic chain and characterization of
the speakers and room.

The fact is that at all these points, no one has really done the job
in a coherent fashion-anywhere-and there are literally no standards.
Some tools-test records and CDs, calibrated voltmeters of wide
bandwidth (were we saying true rms here-we'd better get a thermocouple
instrument!)-exist but the methodology is nonexistent.

Is it really worth the bother? Well, if audio is a hobby, maybe not.

John Atkinson wrote:

Howard Ferstler wrote:
Heck, John, some of the technically excellent recordings
you have engineered had musical qualities of really limited
merit. I never said that in the reviews, however.

In which, you cut a pretty sad figure as a "reviewer," Mr.
Ferstler. Ler's see: you present other people's's writings
as your own, claiming deadline pressure as an excuse; you
fabricate supposedly factual passages out of whole cloth;
you fake your published test results; and now you admit
you pull your punches as a reviewer.

I guess you are correct when you state that you are and I
are not colleagues. :-)

For sure, John.

PS: you are a better recording engineer than creditable
magazine editor.

Howard Ferstler

wrote:

The key to level matching is calibration. One necessary element,
lacking in most high end systems, is a precision attenuator that can be
set to a known, repeatable position. Others include calibration and/or
characterization of the entire electronic chain and characterization of
the speakers and room.

The fact is that at all these points, no one has really done the job
in a coherent fashion-anywhere-and there are literally no standards.
Some tools-test records and CDs, calibrated voltmeters of wide
bandwidth (were we saying true rms here-we'd better get a thermocouple
instrument!)-exist but the methodology is nonexistent.

Is it really worth the bother? Well, if audio is a hobby, maybe not.

It is tedious to level match with speakers. The way I do it
is to use two amps, with one being capable of level
adjustments. I then do a room curve run with each pair of
speakers and then adjust the levels so that the two curves
overlay as much as possible.

With amps, purists say you should use a precision digital
volt meter. (The ABX device I used a while back had one of
those built right in.) However, I have found that a quick
way to do it is use pink noise. Just adjust the levels on
each channel (left/left and then right/right) so that when
you switch from one amp to the other there is no audible
difference. This also allows one to see that in terms of
frequency response similarities (something pink noise will
highlight easily), the set up automatically proves the
identical sound. Then, you go on to comparing with music.

With wires, I have yet to find a situation where it is
necessary to work hard to level match. It happens pretty
much automatically to an accurate enough degree.

Howard Ferstler

wrote in message
oups.com

The key to level matching is calibration.

No such thing. The goal of level matching is to insure that voltage A =
voltage B. Calibration would be important if the goal was to make the
voltages equal to a certain voltage, but all that is required is that the
two voltages be the same.

One necessary element,
lacking in most high end systems, is a precision attenuator that can
be set to a known, repeatable position.

Again, no such thing. If you want to level-match two devices that have
volume and balance controls, or separate gain controls for each channel,
then you need no additonal controls. If you don't have enough control built
into the piece of equipment, then you need one or more exteranal
attenuators, but they don't need to be calibrated to any externals standard.

If you are testing all products at the same time, that would be true.
However, repeatability, and consistency-i.e. the ability to compare A-B
and A-C without having B and C there-demand adherence to objective
repeatable standards. Absolute level, not just relative, is needed, if
you want the results to be significant to anyone not there at the time.

wrote in message
oups.com

If you are testing all products at the same time, that would be true.

Look Cal, you made a global statement. If you want to admit that that you're
wrong, be my guest. If you want to redefine the problem until like one of a
million monkeys trying to type the works of Shakespere, you can waste
someone else's time.

However, repeatability, and consistency-i.e. the ability to compare
A-B and A-C without having B and C there-demand adherence to objective
repeatable standards.

If you're doing a listening test comparing alternatives, there is really
only one way to get the benefits of comparing B to C and that is to compare
B to C. So your claim is wrong on the grounds that your justification is
bogus.

Absolute level, not just relative, is needed, if
you want the results to be significant to anyone not there at the
time.

You don't need a voltmeter with NIS-tracable calibration certificate to
figure out whether your test level is more like 1 watt or 10 watts.