John Curl's Blowtorch preamplifier part II

[jan.didden]

Originally Posted by fredex

Hi all, good reading and it has made me think...

When a golden eared audiophile has a real inner experience of hearing minute differences (say between electronics) that later disappear when he is not allowed to see the equipment, I conclude that his inner experience is not based on an external reality. However to reach this conclusion I have to make the assumption that his actual hearing sensitivity is exactly the same in both instances. How can I know it is?

This is a good point. I think you must make a difference between his actual hearing mechanism, like external ears, eardrum, cilia, tympani etc. and the perceptive engine between those ears.

It stands to reason that the mechanical part, the ear etc, is not different in the two situations. But the great difference is in the perception engine; it now has a very different set of sensory inputs (if the test is done well, ONLY the sound and not things like vision and knowledge of which component is playing). So that can very much change the outcome.

In reality it is even more involved because the perception engine also has a feedback link to the ears, like to the cilia. So depending on what the perception engine 'thinks' or 'expects', (sorry for the pun) it can modify the ear parameters like supressing certain tones/levels and enhancing others. Remember, the brain is a prediction machine and will always try to 'fill in' the rest of the sheet from what it deduces from the first few lines.

jan didden

[Roxanne]

Im struggling to find affordable PCBs for students and I was wondering if any of you guys could let me know where you are purchasing yours

Roxy

[Telstar]

Originally Posted by soundmeister

What are the typical conditions of formal ABX testing ????.
Through long audio experience I am well able to discern and describe reasonably subtle differences quickly and reliably on music and systems and environment that I am well familiar with...for REALLY fine differences I may require multiple AAA-BBB-AA-BB-AB plays of particular passages on music that I know intimately to isolate and define sonic subjective differences.
Then after that comes longer term listening over multiple tracks/albums/genres that can extend to hours or days to properly establish subjective opinions.
Can it be that typical ABX testing of FINE differences is fundamentally flawed because it does not meet the above conditions ????.

Dave.

I agree with you.
I like to do quick A/B tests, without the X (see below), to help me find the differences between two components. Adding the random part just plays tricks to my mind in the hard task of finding SUBTLE differences.
Blind ABX works only when the differences are so big that even a deaf person (which usually match with one looking at the scope first) can hear them.
Once it is found that there are audible differences, it's often more difficult to decide what is BETTER, and that usually require a long-term listening of both DUTs.

[jan.didden]

Originally Posted by Roxanne

Im struggling to find affordable PCBs for students and I was wondering if any of you guys could let me know where you are purchasing yours

Roxy

This is OT but here goes: I get mine from OLIMEX Ltd. - Electronic Design and PCB sub-contract assembly OEM/ODM service , see 'proto boards'.

jan didden

[Jakob2]

Originally Posted by janneman

<snip>

The point is, unless you do a controlled test, you cannot be sure whether there indeed IS an audible difference or not.
<snip>

The other point is, that after doing a controlled test, you cannot be sure either. (SCNR)

Due to the nature of statistical reasoning you will be only able to set some criteria for the error risks you are willing to accept. But of course, if your criterion is too tight, the risk to miss something existent will grow.

Beside that i really appreciated that you use the term "controlled" instead of "blind test" because the former phrase addresses a lot more points than the latter.

[johnferrier]

Originally Posted by jlsem

i have learned since my last screed that the newish Mariinsky Concert Hall in St. Petersburg, which is now regarded as one of the finest concert halls in the world, was in fact done by the Japanese firm Nagata Acoustics. Wood was used throughout its interior construction.

John

re: new Mariinsky Concert Hall

Picked up The Enchanted Wanderer and The Nose. Listened to the former and find the sound quality impressive. All recordings on the Mariinsky label are SA-CD. Wonder if vinyl enthusiasts would find such sound lacking ?


.

[jan.didden]

Originally Posted by Jakob2

The other point is, that after doing a controlled test, you cannot be sure either. (SCNR)

Due to the nature of statistical reasoning you will be only able to set some criteria for the error risks you are willing to accept. But of course, if your criterion is too tight, the risk to miss something existent will grow.
[snip].

Yes, good to remind us of it. If a difference is heard it has a confidence associated to it, like: "we can say with 97% confidence that there IS an audible difference".

Then again, if no difference was heard, it doesn't mean there isn't any.
You can't prove a negative. But you know all that.

Beside that I really appreciated that you use the term "controlled" instead of "blind test" because the former phrase addresses a lot more points than the latter

jan didden

[john curl]

Well done, Joshua. I'm glad that you went forward anyway, even though I had to wince a few times, when you (and I) were attacked, and people were using your academic background (or lack of it) against you with standard debating techniques. This is NOT a level playing field and everybody knows it.
Unfortunately, I do take an 'authoritarian' stance, often because I am a recognized authority on audio electronic design. I am certainly NOT the only authority on this, but I have earned that rating. However, my imperfect social skills sometimes paint me into a corner, where people will go after me on matters large and small, just to bring me down a peg.
The real problem here is a matter of 'belief'. This happens often in the history of science and psychology. Any study of the history of science will show where a dominant belief system AT THE TIME, virtually overwhelms any dissent, yet it will not last forever.
For example, Freudian Psychology, and its rigid practice of psychotherapy. When I was much younger I had 5 years of it. Boy, were they rigid! And opinionated too! While, I learned much about my self, in spite of the conditions and beliefs imposed on me, the use of a simple 'E-meter' would have facilitated getting to the 'root' of important problems much, much faster. Today, the idea of lying on a couch and telling an all knowing analyst about your 'mother' for example is almost laughable, YET it was everything, just a couple of generations ago.
I personally hold the opinion that this lack of trust of one's own personal opinions was also developed by a group of academics, starting with Dr. Lipshitz, and going on through ABX testing by BAS, and later by Dr. Toole and others, and books have been written about it, by them and many others as to how YOU too can be fooled by your own perceptions, so you shouldn't trust them. In audio, this has created this debate about whether experience and aptitude can be trusted, even if it appears to be successful in the world marketplace and opinion.
I think that this too shall pass, at least enough to remove the absurdities of not trusting yourself to make a serious decision between audio components and the often stated conclusion that everything essentially sounds the same.

[scott wurcer]

Originally Posted by hhoyt

http://wxyc.info/playlists/searchPla...ageToDisplay=1



p.s. let me head this one off at the pass, no; I do not eat fast food...from anywhere!

You wouldn't say that about the food carts outside Hamamatsuchō.

[scott wurcer]

Originally Posted by Roxanne

Im struggling to find affordable PCBs for students and I was wondering if any of you guys could let me know where you are purchasing yours

Roxy

Download any recent issue of Elecktor and there are numerous listings for services at any level. I use ExpressPCB $51 for three 1/2 eurosize boards. Good for student projects.