John Curl's Blowtorch preamplifier part II

[jacco vermeulen]

The margin of error ?

[jan.didden]

Originally Posted by jacco vermeulen

The margin of error ?

Jacco, you just made it 50.0001%. And with this post, 50.0002%.
Sorry John, couldn't resist....

jan

[jacco vermeulen]

Not 75.0001% ?
Jan, you're too kind. (btw, very enjoyable read Volume 3 is)

Arm length is measured from top of thumb to shoulder, right ?

[scott wurcer]

Originally Posted by simon7000

John,

Here we strongly disagree.

One of the nice side effects of having test equipment is that when you use both you learn the signature sound of certain things you are measuring.

ES

Yes we do, I thought the crux of these discussions was where there was no (meaningful) measurable difference. If we are back at annecdotal comments about resistors seperated into -140dB vs -120dB distortion bins, the connection has not been proven.

If we are going to practice relativism, why aren't my experiences (no sonic difference between many brands of film resistors) on equal ground.

[Steve Eddy]

Originally Posted by scott wurcer

If we are going to practice relativism, why aren't my experiences (no sonic difference between many brands of film resistors) on equal ground.

'Cause that's how religion works.

se

[jan.didden]

Originally Posted by jneutron

The ridiculously foolish phrase " I trust my ears even though I can't prove it", is an abysmal affront to intelligent and educated people. [snip]cheers, j

The interesting observation is that people who refuse any form of controlled listening are implying that, in contrast to what they say, they DO NOT trust their ears.
I DO trust my ears - I am willing to shut off all other external influences and just listen with only my ears. I DO NOT need to know the color of the amp, the price of the speaker, the reputation of the CD player. I only want the sound to be the arbiter - and thats why we need controlled listening.

"I don't need controlled listening because I trust my ears" must be the biggest bs statement in audio.

jan

[abraxalito]

Originally Posted by janneman

The interesting observation is that people who refuse any form of controlled listening are implying that, in contrast to what they say, they DO NOT trust their ears.

Just a moment - its an 'observation' then its an implicaton? How can one observe an implication? Implicatons are inferred through a process of reasoning based on assumptions are they not? So how about you make your assumptions open to cross examination here?

[ThorstenL]

Scott,

Originally Posted by scott wurcer

If we are going to practice relativism, why aren't my experiences (no sonic difference between many brands of film resistors) on equal ground.

They are on equal ground.

There is just a minor sticking point.

Your experience shows no difference, which is fine.

But as it did not reliably cover all brands and constructions of resistors the negative outcome cannot be generalised, disregardless of any other issues, such as experimental design etc.

Incidentally neither can individual positive outcomes of tests, e.g. old Holco "sound good" (new ones no longer BTW) be construed to provide proof that all resistors sound somehow different or even that one particular brand or type of resistor is reliably superior to others.

In fact, there is no scope for generalisation in these tests as the statistics are simply to marginal to conclude anything with any degree of confidence that is beyond taking a leak in a very dark place with srong gusty and varying direction wind.

BTW, the old Holco's had crucial differences in construction to "generic film resistors" which where not duplicated in other brands of very expensive, tight tolerance Metal Film Resistors (which sounded just like the cheapest meatl film resistors). In the process of ROHS compliance etc. Holco changed their construction to match their competitors and lo and behold, current Holco indeed show no appreciable difference (to my ears and testing) over their competitors and you might as well buy the cheapest chinese made MF resistor.

I previously mentioned the blind (and "ultra everything blind") test we did with comparably minor power-supply changes.

We did a similar test with SMD Film resistors and they confirm your impression.

Most generic metal film resistors are equally BAD. It is perfectly possible to compare several dozens of Brands for SMD film resistor and not find an appreciable audible (or measurable!) difference.

What I am quite certain you did not do, however, was to test resistors that are mechanically compatible, but use alternate construction/design/materials.

We did. And found material differences in testing, with strong preferences compared to generic film types, in BLIND testing (single blind preference test).

I cannot tell you how displeased I was by the tests outcome, even though I had staked my reputation in the company on "my magic resistor" (note, tongue firmly in cheek for the magic), because they where only available in truely massive MOQ per value (100KU - seemed we had to do a new production run for each and every value) and we somehow had to find the money to pay for this MOQ on something like 50 Values, a few month after the economy of the western world had been flushed down the toilet, so cash was all but plenty or easy...

I would have been pleased as punch had the outcome produced a "no difference" or a "no difference worth having" outcome...

BTW, in a certain specific part of an MC input stage the difference in noise between a "generic film resistor" and my "magic resistor" came up to around 10dB at LF, starting below 1KHz. The resistor formed the tail of a folded cascode BTW, with a 1:1000 ratio between cascode impedance and resistor value...

This is the only situation are where I was able to MEASURE a reliable difference between "generic" and "magic" resistors, using AP2 or EMU1616m.

Ciao T

[ThorstenL]

Hi Jan,

Originally Posted by janneman

The interesting observation is that people who refuse any form of controlled listening are implying that, in contrast to what they say, they DO NOT trust their ears.

Now, who please refuses ANY form of controlled listening tests, as opposed to having specific or general objections to specific kind of tests.

For example, personally I object to tests that have more than 4-5 presentation in a row. And to ones in unfamilar settings etc. My ability to be bothered about sound simply evaporates. So if you for require me to produce 16 positives in 20 trials in a row, in an unfamiliar setting, for known audible differences, forget it.

Now four times 4 positives in 5 trials for know audible differences I find not difficult, with decent intervals between trials and in a context I am familiar with, as well as with the fundamental protocol under my control.

I also personally object to tests being called "controlled" because they control a single variable, while omitting both positive and negative controls and are open to many criticisms of method and statistics. Such tests at best should be called "poorly controlled" and a s a result any data derived from such tests has to probably be considered irrelevant or unacceptable, because of the precise lack of decent controls.

Originally Posted by janneman

"I don't need controlled listening because I trust my ears" must be the biggest bs statement in audio.

So, who actually makes these statements? Who (names please) actually claimed they do not need controlled tests because they trust their ears (as opposed to objecting to specific methods etc.)?

For now I somehow seem to have a major smell of staw in my nose...

Ciao T

[abraxalito]

Originally Posted by ThorstenL

Now, who please refuses ANY form of controlled listening tests, as opposed to having specific or general objections to specific kind of tests.

I'll raise my hand here, with a couple of qualifications.

I choose not to take part in such tests (rather than 'refuse' which sounds like there's some forceful determination not to take part). I don't take part because I see no benefit to myself in discovering whether having a more controlled listening environment than normal can make my placebo effect (if placebo there indeed be) disappear, or not. Hence taking part has no advantage from my side to offset the time required.

Put very bluntly, tests are boring.