This is what I was thinking.
Set the recorder for mic-in, but don't actually plug
a mic in. Then get the recording started and run the
recorder through the various levels of gain, from low
to high. Take the resultant recording and load it into
your audio editing program and check the dB
levels at the different gain settings that were used.

Any reason why this wouldn't work and be accurate?

Set the recorder for mic-in, but don't actually plug a mic in.
Then get the recording started and run the recorder through
the various levels of gain, from low to high. Take the resultant
recording and load it into your audio editing program and
check the noise levels at the different gain settings.

Any reason why this wouldn't work and be accurate?

Yes!

The preamp's noise level is affected by the source impedance driving it. So
you need to attach a resistor of the same value to the mic input.

This assumes a 600 ohm mic "really" looks like a 600-ohm resistor. It might
not. But the principle is correct.

wrote:

This is what I was thinking.
Set the recorder for mic-in, but don't actually plug a mic in.

Mistake number one ! You may get some consolation by being told it's a popular
mistake though.

If you want to do this substitute a 20 ohm resistor for the mic. And don't turn
it up full unless that's what you intend to do with a mic connected.

Graham

William Sommerwerck wrote:

Set the recorder for mic-in, but don't actually plug a mic in.
Then get the recording started and run the recorder through
the various levels of gain, from low to high. Take the resultant
recording and load it into your audio editing program and
check the noise levels at the different gain settings.

Any reason why this wouldn't work and be accurate?

Yes!

The preamp's noise level is affected by the source impedance driving it. So
you need to attach a resistor of the same value to the mic input.

This assumes a 600 ohm mic "really" looks like a 600-ohm resistor. It might
not. But the principle is correct.

Few mics are 600 ohms. Most decent dynamic low impedance mics vary from 150 -
300 ohms IME.

Graham

wrote:

That's what I was kind of figuring ... that I should take into
account the mic in some way.

So, I should use a resistor of the same resistance as the mic I intend to use?

Yes.

And where would I find this resistor? The input is 1/8".

Actually it's 3.5mm !

Somehow I think it's more involved than walking into Radio Shack and asking
for a "resistor with a 1/8" stereo audio plug connected to it".

Can you solder ? It's as easy a soldering a 150 ohm resistor to a 3.5mm jack.
You can get these parts in RS.

Graham

On Sat, 03 Jun 2006 14:26:16 +0100, Pooh Bear
wrote:


If you want to do this substitute a 20 ohm resistor for the mic. And don't turn
it up full unless that's what you intend to do with a mic connected.

Graham

Why 20 ohm instead of 150?

Julian

Julian wrote:

On Sat, 03 Jun 2006 14:26:16 +0100, Pooh Bear
wrote:

If you want to do this substitute a 20 ohm resistor for the mic. And don't turn
it up full unless that's what you intend to do with a mic connected.

Graham

Why 20 ohm instead of 150?

Typo ! Should have been 200. Or indeed use 150.

Graham

wrote:
This is what I was thinking.
Set the recorder for mic-in, but don't actually plug
a mic in. Then get the recording started and run the
recorder through the various levels of gain, from low
to high. Take the resultant recording and load it into
your audio editing program and check the dB
levels at the different gain settings that were used.

Any reason why this wouldn't work and be accurate?

Because the noise level will drop considerably when a mike is plugged in.
Instead, put a 600 ohm resistor across the input and measure the noise then.
Noise with a mike will be similar (probably lower if the mike output Z is
lower).
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

Scott Dorsey wrote:

Instead, put a 600 ohm resistor across the input and measure the noise then.
Noise with a mike will be similar (probably lower if the mike output Z is
lower).

A short circuit at the input gives more consistent results when
comparing preamps. While the noise inherent in a 600 or 150 ohm
resistor isn't likely to corrupt your results, different preamps can
have different internal noise with different source resistances. A
short is a short is a short regardless of the input stage
configuration.

"Mike Rivers"

A short circuit at the input gives more consistent results when
comparing preamps. While the noise inherent in a 600 or 150 ohm
resistor isn't likely to corrupt your results, different preamps can
have different internal noise with different source resistances. A
short is a short is a short regardless of the input stage
configuration.



** How typically asinine of the Brain Dead Parrott.

No mic is ever like a short - so such a test is fake.




....... Phil

A short circuit at the input gives more consistent results
when comparing preamps. While the noise inherent in a
600 or 150 ohm resistor isn't likely to corrupt your results,
different preamps can have different internal noise with
different source resistances. A short is a short is a short
regardless of the input stage configuration.

And if you're using 600-ohm mics, then THAT is consistent.

You can only get valid results by testing under the conditions under which
the product will be used. A short circuit is not that condition.

William Sommerwerck wrote:

A short circuit at the input gives more consistent results
when comparing preamps. While the noise inherent in a
600 or 150 ohm resistor isn't likely to corrupt your results,
different preamps can have different internal noise with
different source resistances. A short is a short is a short
regardless of the input stage configuration.

And if you're using 600-ohm mics, then THAT is consistent.

You can only get valid results by testing under the conditions under which
the product will be used. A short circuit is not that condition.

100% agreed. The difference Mike talks about must be caused by the noise
component of input bias current. The only way to replicate its effect is to use
a suitable resistor or you're simply ignoring that part of the mic amp's noise.

Graham

Can you solder ? It's as easy a soldering a 150 ohm resistor to a 3.5mm jack.
You can get these parts in RS.

Graham


When thinking of tip/ring/sleeve, which of these two should
be used for the solder points?

Do you think there'd be much difference in the result if I
got, say, a 150 ohm or a 600 ohm?

Also, when I went to look at Radio Shack I see
the resistors are offered in "watts", such as 1/4 watt,
1/2 watt, etc. Any guidance could be offered
in choosing the watts?

Do you think there'd be much difference in the result if I
got, say, a 150 ohm or a 600 ohm?


** Around 6 dB more noise with 600 ohms - if the pre-amp is a good one.


Also, when I went to look at Radio Shack I see
the resistors are offered in "watts", such as 1/4 watt,
1/2 watt, etc. Any guidance could be offered
in choosing the watts?


** Get *watt*ever's available.

Makes no difference to the result.




........ Phil

wrote:

Do you think there'd be much difference in the result if I
got, say, a 150 ohm or a 600 ohm?

There will be a measurable difference if the mic amp's any good ( 600 ohms will
be noisier ) not sure if you'll hear it.

Also, when I went to look at Radio Shack I see
the resistors are offered in "watts", such as 1/4 watt,
1/2 watt, etc. Any guidance could be offered
in choosing the watts?

Common or garden 1/4 W is fine.

Graham

wrote:

Can you solder ? It's as easy a soldering a 150 ohm resistor to a 3.5mm jack.
You can get these parts in RS.

Graham

When thinking of tip/ring/sleeve, which of these two should
be used for the solder points?

Not sure what you mean.

Graham

Pooh Bear wrote:

wrote:

Can you solder ? It's as easy a soldering a 150 ohm resistor to a 3.5mm jack.
You can get these parts in RS.

Graham

When thinking of tip/ring/sleeve, which of these two should
be used for the solder points?

Not sure what you mean.

Oh you mean where to connect the resistor. Tip and ring.

Graham

wrote in message
ups.com...
Do you think there'd be much difference in the result if I
got, say, a 150 ohm or a 600 ohm?

There'll be some difference on some preamps, because their noise is
primarily the source resistance noise. There'll be little or no difference
on other preamps, because their noise is primarily internal. They're
noisier, in other words.

Since there are a lot more mics around 150 ohms than 600 ohms, I suggest you
use 150.

Also, when I went to look at Radio Shack I see
the resistors are offered in "watts", such as 1/4 watt,
1/2 watt, etc. Any guidance could be offered
in choosing the watts?

Get half-watt. Ideally you'd use something like 1% metal film, which behaves
pretty much like a pure resistance (not quite, but close). But I don't think
the Shack has 1% in 150 ohms, so if you buy their regular resistors, get 1/2
watt, which will have less excess noise than 1/4 watt. Probably.

Peace,
Paul

wrote:
Do you think there'd be much difference in the result if I
got, say, a 150 ohm or a 600 ohm?

Depends on where the main noise source in the preamp is.

Also, when I went to look at Radio Shack I see
the resistors are offered in "watts", such as 1/4 watt,
1/2 watt, etc. Any guidance could be offered
in choosing the watts?

Use whichever is cheapest. The higher wattage ones _might_ have lower
noise if they are the same technology, but if you have to worry about
source resistor noise, you have a VERY quiet preamp.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

The higher wattage ones _might_ have lower
noise if they are the same technology...

All resistors of a particular value produce the same thermal noise, and have
the same effect on the preamp's noise figure. The only thing left is excess
noise (which is related to the resistor's materials and fabrication), but it
applies only if a voltage is applied to the resistor.

William Sommerwerck wrote:

You can only get valid results by testing under the conditions under which
the product will be used. A short circuit is not that condition.

How useful is a test which is restricted to a single mic and preamp? At
least if you make measurements with a shorted input, you don't have to
worry about capacitance and inductance of the source and you can
compare results between preamps. Is a input-short-circuit test any less
meaningful than a resistor if you're connecting a mic with an output
transformer?

But then, how much difference in residual noise would you find in ANY
preamp between using a dummy source of 600 ohms, 150 ohms. or zero
ohms?

This is all bogus anyway, unless you're the preamp designer and you
want to know how successful your design is.

Pooh Bear wrote:

100% agreed. The difference Mike talks about must be caused by the noise
component of input bias current. The only way to replicate its effect is to use
a suitable resistor or you're simply ignoring that part of the mic amp's noise.

Numbers. Show me numbers, actual measurement, not calculations. Then,
based on the numbers, I'll tell you whether I care about the
difference. I don't trust my test equipment so I'll have to rely on
someone else's test.

What sort of difference might we be talking about here? 2 dB? 0.5 dB? 6
dB?

Mike Rivers wrote:

But then, how much difference in residual noise would you find in ANY
preamp between using a dummy source of 600 ohms, 150 ohms. or zero
ohms?

A big difference. In fact it's only poor preamps where the difference is small.

Graham

Mike Rivers wrote:

Pooh Bear wrote:

100% agreed. The difference Mike talks about must be caused by the noise
component of input bias current. The only way to replicate its effect is to use
a suitable resistor or you're simply ignoring that part of the mic amp's noise.

Numbers. Show me numbers, actual measurement, not calculations. Then,
based on the numbers, I'll tell you whether I care about the
difference. I don't trust my test equipment so I'll have to rely on
someone else's test.

What sort of difference might we be talking about here? 2 dB? 0.5 dB? 6
dB?

The difference between 200 ohms and 600 ohms is the square root of their ratios i.e.
sqrt 3 = 1.73 or 4.8dB. Between 150 ohms and 600 sqrt 4 = 2 or 6dB.

That's just their thermal noise contribution btw.

Graham

"Scott Dorsey"

Use whichever is cheapest. The higher wattage ones _might_ have lower
noise if they are the same technology,



** Yaaaaawwwnnnn.

More asinine drivel from this INCORRIGIBLE know nothing FAKE..






........ Phil

"Mike Rivers"


** The Brain Dead Parrot is having a feather pulling fit !!

LOL !!!!


You can only get valid results by testing under the conditions under
which
the product will be used. A short circuit is not that condition.

How useful is a test which is restricted to a single mic and preamp?


** False scenario.


At
least if you make measurements with a shorted input, you don't have to
worry about capacitance and inductance of the source and you can
compare results between preamps.


** False assertion.


Is a input-short-circuit test any less
meaningful than a resistor if you're connecting a mic with an output
transformer?


** A mic's transformer makes NO difference.

The mic's impedance in the audio band is the relevant parameter.

A 270 ohm resistor will exactly simulate the self noise of an SM58.



But then, how much difference in residual noise would you find in ANY
preamp between using a dummy source of 600 ohms, 150 ohms. or zero
ohms?


** For a good SS pre-amp, 150 ohms will give 5 dB less noise and a short
about 10dB less - relative to 600 ohms.



This is all bogus anyway,


** The Brain Dead Parrot's copious runny dropping certainly are.


ROTFL.....



........ Phil

Pooh Bear wrote:

The difference between 200 ohms and 600 ohms is the square root of their ratios i.e.
sqrt 3 = 1.73 or 4.8dB. Between 150 ohms and 600 sqrt 4 = 2 or 6dB.

I asked for mreasurements, not theory. If I measure the noise of a
preamp with a 150 ohm resistor and a 600 ohm resistor at the input, can
I expect the noise to change by 6 dB? I can certainly measure that, and
I will report back.

Phil Allison wrote:

A 270 ohm resistor will exactly simulate the self noise of an SM58.

The changeling is at it again. We're talking about mic preamp noise and
he smoothly changes the subject to microphone self-noise. Irrelevant,
but interesting.

Mike Rivers wrote:

Pooh Bear wrote:

The difference between 200 ohms and 600 ohms is the square root of their ratios i.e.
sqrt 3 = 1.73 or 4.8dB. Between 150 ohms and 600 sqrt 4 = 2 or 6dB.

I asked for mreasurements, not theory.

In what way do you expect measurements not to confirm the equation for thermal noise ?

If I measure the noise of a
preamp with a 150 ohm resistor and a 600 ohm resistor at the input, can
I expect the noise to change by 6 dB? I can certainly measure that, and
I will report back.

You won't see quite as much as 6dB because of the preamp's self noise. The better the
preamp, the bigger the difference you'll see. Noise sums as the sqrt of the sum of the
squares.

To measure the preamp's noise*voltage* you can indeed short the input but this won't
establish the input noise *current* which will add to the figure when the preamp sees a
real mic.

Graham

Pooh Bear wrote:
Mike Rivers wrote:

But then, how much difference in residual noise would you find in ANY
preamp between using a dummy source of 600 ohms, 150 ohms. or zero
ohms?

A big difference. In fact it's only poor preamps where the difference is small.


OK, I must have just tested a very good preamp, and I'll admit that I
was surprised at the results. Theory and practice agree fairly
closelyl.

Measurements made with a Mackie Onyx mixer from the mic input to the
"Recording" output (most direct output from the preamp). Using a NTI
Minirator as the source and NTI Minilizer as a meter on the output, I
set the preamp gain to 50 dB. Output level was +4 dBu. The resistors
were installed in XLR male connectors, between pins 2 and 3, with pin 1
connected to the metal shell of the connector for shielding. The
resistors came out of my parts cabinet, and are garden variety 1/4 watt
5% metal film of unknown quality.

Output with the input shorted: -76 dBu
Output with 150 ohms across the input: -70 dBu
Output with 600 ohms across the input: -64 dBu

Incidentally, the noise with the input open was -58 dBu. I don't
imagine this is terribly relevant to the discussion, however.

So it's clear that you can get more noise out of a preamp by not
shorting its input. I expected this, though not such a large difference
in the real world.

But (in deference to Mr. Allison who chooses to talk about something he
knows rather than the actual topic of discussion (which I quote from
the meassage header) " Is this a good way to test pre-amp hiss of a
digital recorder?" - what do we actually want to test? Do we want a
real EIN number?

We can make that whatever we want by changing the input source
resistance and it's rare that the input source is specified if EIN is
published in manufacturer's literature. Does anyone know if there is a
standard test procedure? EIN also changes with gain, which is why it's
almost always measured (for publication in marketing literature anyway)
at maximum gain, which in the case of this Mackie is about 8 dB higher
than were I measured it.

If we want to compare mic preamps, (rather than come up with an
absolute number) they have to be measured under the same set of
conditions. I'm suggesting that a shorted input is the best way to do
this because:

1. It's easier to assure the accuracy of the resistance of a short
piece of wire than it is to assure the accuracy of a 150, 270, or 600
ohm resistor.

2. A short circuit will be more immune to induced EMI than a higher
input circuit impedance. This makes testing outside of a screened room
closer to accurate. It may be significant if the unit under test is
indeed a "digital recorder" with real EMI floating around inside, and
probably getting out to find its way back into the input through a 600
ohm loop antenna.

3. The numbers look better.

By testing with a shorted input, the measurement is independent of what
microphone is used, and it allows preamps on their own to be compared
on an equal basis.

But we're getting pretty deep here when the original poster is probably
asking because he turned up the monitor volume all the way, with no
input, and heard hiss (from his un-named "digital recorder").

Pooh Bear wrote:


You won't see quite as much as 6dB because of the preamp's self noise. The better the
preamp, the bigger the difference you'll see. Noise sums as the sqrt of the sum of the
squares.

Ignoring smaller than unit values of dB, I did in fact see about 6 dB
difference.

To measure the preamp's noise*voltage* you can indeed short the input but this won't
establish the input noise *current* which will add to the figure when the preamp sees a real mic.

I follow that, this is getting too theoretical for a practical test. A
real mic has varying amounts of noise output depending on its design.
In additon, it's very difficult to acoustically isolate a mic, so you
can't get rid of any acoustic noise that gets converted to output
voltage (seen as noise at the preamp output) because of what a
microphone is suposed to do. So testing with a real mic is out the
window, and emulating a "standard" mic with a resistor probably doesn't
really emulate any real microphone. It's a way to standardize the test
procedure, however - if (and only if) everyone used the same procedure,
and the same resistor.

... (from his un-named "digital recorder").


iRiver.

"Mike Rivers **** "


Measurements made with a Mackie Onyx mixer from the mic input to the
"Recording" output (most direct output from the preamp). Using a NTI
Minirator as the source and NTI Minilizer as a meter on the output, I
set the preamp gain to 50 dB. Output level was +4 dBu. The resistors
were installed in XLR male connectors, between pins 2 and 3, with pin 1
connected to the metal shell of the connector for shielding. The
resistors came out of my parts cabinet, and are garden variety 1/4 watt
5% metal film of unknown quality.

Output with the input shorted: -76 dBu
Output with 150 ohms across the input: -70 dBu
Output with 600 ohms across the input: -64 dBu


** So, relative to 600ohms,

150 ohms = 6 dB less noise.

a short = 12dB less noise.


Almost EXACTLY as I intimated in my post to YOU yesterday:

QUOTE:

" ** For a good SS pre-amp, 150 ohms will give 5 dB less noise and a short
about 10dB less - relative to 600 ohms. "


YOU ****ING LYING ASSHOLE !!!!!!



We can make that whatever we want by changing the input source
resistance and it's rare that the input source is specified if EIN is
published in manufacturer's literature.


** WRONG !!!!!

It is almost INVARIABLY specified.

YOU ****ING LYING ASSHOLE !!!!!!



If we want to compare mic preamps, (rather than come up with an
absolute number) they have to be measured under the same set of
conditions. I'm suggesting that a shorted input is the best way to do
this because:

1. It's easier to assure the accuracy of the resistance of a short
piece of wire than it is to assure the accuracy of a 150, 270, or 600
ohm resistor.



** WRONG !!!


YOU LYING, ASININE, ****ING ASSHOLE !!!!!!



2. A short circuit will be more immune to induced EMI than a higher
input circuit impedance.


** WRONG !!!!!!


YOU VILE, ****WIT, LYING , ASININE ****ING ASSHOLE !!!!!!



3. The numbers look better.


** BUT FRAUDULENT !!!!


YOU SLIMY, VILE , ASININE ****ING ASSHOLE !!!!!!



By testing with a shorted input, the measurement is independent of what
microphone is used, and it allows preamps on their own to be compared
on an equal basis.


** WRONG !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!


YOU ****WITTED, ASININE, ****ING, ASSHOLE !!!!!!


Tell that scum of the earth CHARLATAN Dorsey to go drop dead.

Same goes for that Stamler criminal too.





........ Phil

"Mike Rivers" wrote in message
oups.com
Pooh Bear wrote:

The difference between 200 ohms and 600 ohms is the
square root of their ratios i.e. sqrt 3 = 1.73 or 4.8dB.
Between 150 ohms and 600 sqrt 4 = 2 or 6dB.

I asked for mreasurements, not theory. If I measure the
noise of a preamp with a 150 ohm resistor and a 600 ohm
resistor at the input, can I expect the noise to change
by 6 dB? I can certainly measure that, and I will report
back.


Depends on how much of the noise you measure is due to the interal circuitry
of the preamp, and how much is actually due to thermal noise in the
resistor.

The relevant (but not often given) parameter is called "noise figure". The
noise figure of a preamp is taken from the difference between the actual
measurement of the preamp, and the theoretical noise generated by the
resistor.

Therefore the noise figure of a preamp is partially dependent on the value
of the resistor used to test. A good preamp might have 1-2 dB noise figure
with a 150 ohm resistor. Often, a preamp' snoise figure *improves* with
higher valued test resistors.

If your preamp has a very good noise figure, then the increase in noise with
larger resistors will follow the square root of the resistance of the test
resistor.

"Mike Rivers" wrote in message
oups.com
Phil Allison wrote:

A 270 ohm resistor will exactly simulate the self noise
of an SM58.

The changeling is at it again. We're talking about mic
preamp noise and he smoothly changes the subject to
microphone self-noise. Irrelevant, but interesting.

No matter what, I find it to be an interesting factoid.

wrote:

iRiver.

Oh, ferchissake! Are you serious? It's not worth testing other than by
ear. And if you think it's too much, you're absolutely right.

When you get a Sound Devices recorder, or maybe one the "improved"
Marantz ones, let's talk again.

"Mike Rivers"

"Mike Rivers **** "


Measurements made with a Mackie Onyx mixer from the mic input to the
"Recording" output (most direct output from the preamp). Using a NTI
Minirator as the source and NTI Minilizer as a meter on the output, I
set the preamp gain to 50 dB. Output level was +4 dBu. The resistors
were installed in XLR male connectors, between pins 2 and 3, with pin 1
connected to the metal shell of the connector for shielding. The
resistors came out of my parts cabinet, and are garden variety 1/4 watt
5% metal film of unknown quality.

Output with the input shorted: -76 dBu
Output with 150 ohms across the input: -70 dBu
Output with 600 ohms across the input: -64 dBu


** So, relative to 600ohms,

150 ohms = 6 dB less noise.

a short = 12dB less noise.


Almost EXACTLY as I intimated in my post to YOU yesterday:

QUOTE:

" ** For a good SS pre-amp, 150 ohms will give 5 dB less noise and a short
about 10dB less - relative to 600 ohms. "


YOU ****ING LYING ASSHOLE !!!!!!



We can make that whatever we want by changing the input source
resistance and it's rare that the input source is specified if EIN is
published in manufacturer's literature.


** WRONG !!!!!

It is almost INVARIABLY specified.

YOU ****ING LYING ASSHOLE !!!!!!



If we want to compare mic preamps, (rather than come up with an
absolute number) they have to be measured under the same set of
conditions. I'm suggesting that a shorted input is the best way to do
this because:

1. It's easier to assure the accuracy of the resistance of a short
piece of wire than it is to assure the accuracy of a 150, 270, or 600
ohm resistor.



** WRONG !!!


YOU LYING, ASININE, ****ING ASSHOLE !!!!!!



2. A short circuit will be more immune to induced EMI than a higher
input circuit impedance.


** WRONG !!!!!!


YOU VILE, ****WIT, LYING , ASININE ****ING ASSHOLE !!!!!!



3. The numbers look better.


** BUT FRAUDULENT !!!!


YOU SLIMY, VILE , ASININE ****ING ASSHOLE !!!!!!



By testing with a shorted input, the measurement is independent of what
microphone is used, and it allows preamps on their own to be compared
on an equal basis.


** WRONG !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!


YOU ****WITTED, ASININE, ****ING, ASSHOLE !!!!!!


Tell that scum of the earth CHARLATAN Dorsey to go drop dead.

Same goes for that Stamler criminal too.





........ Phil

In article .com,
Mike Rivers wrote:

Pooh Bear wrote:

The difference between 200 ohms and 600 ohms is the square root of their ratios i.e.
sqrt 3 = 1.73 or 4.8dB. Between 150 ohms and 600 sqrt 4 = 2 or 6dB.

I asked for mreasurements, not theory. If I measure the noise of a
preamp with a 150 ohm resistor and a 600 ohm resistor at the input, can
I expect the noise to change by 6 dB? I can certainly measure that, and
I will report back.

If you're using a Millennia Media, you should notice nearly 6 dB difference.
If you're using a Bogen, you will not notice any.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

"Mike Rivers" wrote in message
ups.com...

I follow that, this is getting too theoretical for a practical test. A
real mic has varying amounts of noise output depending on its design.
In additon, it's very difficult to acoustically isolate a mic, so you
can't get rid of any acoustic noise that gets converted to output
voltage (seen as noise at the preamp output) because of what a
microphone is suposed to do. So testing with a real mic is out the
window, and emulating a "standard" mic with a resistor probably doesn't
really emulate any real microphone.

Yes, it does, at least close enough to make meaningful comparisons of the
noise levels of preamps. Some mics come remarkably close to behaving like
pure resistances, notably transformerless condensers. Others aren't as
close, but close enough for it to work.

It's a way to standardize the test
procedure, however - if (and only if) everyone used the same procedure,
and the same resistor.

For the most part, everybody does use the same procedure -- replace the
source with a resistor -- and pretty much the same resistor -- either 150 or
200 ohms. The AES probably has a standards document someplace on how to do
this, but meanwhile get a 150 ohm resistor and you can separate the sheep
from the goats in the preamp world quite effectively.

Peace,
Paul