It would only be intellectual dishonesty if done with conscious intent.
For humans in general, each day many judgments are made automatically using System 1 (from two system model of cognition) mental heuristics and without triggering focused review by conscious awareness.
Looked more like Jakob was trying to point out that even those who consider themselves aimed at maintaining robust even handedness are not immune to some of the difficulties of being human.
Hopefully we at least have some influence over how our feelings (I was just attacked, my reputation impugned!) may dispose us to respond when such is pointed out. (It is my accuser whom is at fault!)
EDIT: the dramatic exclamations are to show how the auto-protect system tends to respond, not for ridicule.
Also, I do not claim to be perfect, just trying to point out how humans may sometimes tend to interact publicly.
Last edited:
That's because it isn't a thing. It was a throwaway comment from Earl which Mark has sadly hung onto like a terrier. For sure normal distribution suggests that some will have slightly better hearing than others but that doesn't support some of the fantastical claims made on here. And all these claims are anecdotes and lots of anecdotes does not make hard data.
OK I have to pull you up here. We know who you are talking about BUT the fact he is/was a moderator has nothing to do with the discussion. A moderator who hasn't started a post with the hat emoticon is just another member. Either call him a member or use his name. Silly word games don't help...
This is unfortunately the case in many instances: you can sell musicians on a studio's sound implying the equipment within may have made famous hits. I have heard that there is a console spec called "mojo" but cannot find a meter to measure it, I believe it is rated in the number of popular hit songs (or maybe $$) made through it.
In addition to people re-installing ancient consoles, some of which work really well, there has been a resurgence in old console circuitry design applied to new consoles. I have had the pleasure of debugging installs with new, very expensive consoles with single-ended, cap coupled circuitry with extremely low PSRR. In the original implementation individual channel transformers were used to provide some galvanic isolation to limit grounding and loop issues, but not in the new implementation. Do they sound good? if you discount the buzz, hum and other noises that come and go, probably.
Regarding new digital consoles; all I have seen have IC front end with perhaps a discrete mic pre...some have switchable soft-knee compression to sound like a saturated transformer like in the good old days...there is money in nostalgia.
Howie
@DPH,
we had some discussions about the underlying problem of these imaging methods wrt Oohashi et al. in the past, after Craig Bennett's publications about the problem:
http://prefrontal.org/files/posters/Bennett-Salmon-2009.pdf
As short presentation accompanied by their more refined publication.
They used a dead salmon as an example what could happen if _no_ correction for the multiple comparison problem was used. In this case it was an active region in the brain of the dead salmon.
A moderator, who always stated that Oohashi et al. were flawed, but never came up with arguments before, jumped on the train and asserted that Oohashi et al. did not do multiple comparions (), and later that they did not use any correction for the multiple comparison problem.
When it was pointed out that they actually did use a correction method for the multiple comparison problem, the assertion was changed to, that it was the wrong correction method because they should have used a method introduced by Benjamin/Hochberg.
While the Benjamini/Hochberg is a better balanced method (means more balanced wrt avoiding false positive but retain statistical power) the method Oohashi et al. used was the more conservative one, means lowered the risk of false positives by accepting the higher risk of false negatives.
To be fair, newer publications since ~2011 (like you mentioned) aruged based on large scale tests on simulated data, that better methods exist compared to the one that Oohashis et al. used.
Where is the time machine when you need one ....
Beside the not so important audio related issue, I was shocked reading that so many experiments were done without usage of any corrections method at all.
Actually I'd never thought that these machines even allow such use (thinking about the consequences of wrong diagnosises)
Thanks for the link! fMRI is a relatively new field and I'd argue the majority of early research falls into the vein of "oooo look at this new shiny machine our university purchased, let's use it" coupled to and excitement to fund said research by various private and public agencies. Then reality struck when the body of work was put under an appropriate magnifying lens.
Beyond the methodology problems, many of which are hopefully in our past, my issue lies more with the insinuated efficacy of fMRI to understand underlying neural networks. The fact that large regions (voxel size is still sadly huge) light up metabolically is very very helpful when looking at certain processes (diseases, gross pathology, genetic/developmental neurology) and on a pure technological level, this stuff is amazing. fMRI + EEG provides another layer of concomitant data (voxel size on EEG is atrocious though as we're not keen on putting a bunch of electrodes IN people's brains), but these methods still seem more in their scientific infancy. Fortunately, things are calming down.
I find the weak correlation between what fMRI says and the ostensible claims of activated neural networks, sadly popularized as "you're thinking these things" or "these groups have this pattern of thoughts, so if you want to be like them, think this way; here's the fMRI data to prove it" to be so far beyond the scope of the data.
The biggest problem I have with papers coming out of Oohashi's group is their extremely low N coupled with a weak correlation to hard-to-quantify-data. It leaves me with a strong sense of "who cares?" There's been zero follow up in almost 30 years by any other group, so all the papers are referencing themselves. Everyone does it, but that doesn't make the practice right.
It perhaps makes for an interesting series of hypotheses but I only see a few puffs of smoke, not a fire.I'm not one to usually throw the baby out with the bathwater, but frankly the whole of audio-related perception studies have instilled a *lack* of confidence in me. For one, the amount of "motivated statistics" is horrifying. Of course, the more I put a lens to a lot of the biological/medical studies done (many famously not-reproducible), the stickier the situation becomes. Coming from a "cleaner" science background, (i.e. not relying on living creatures
), I have to take a pragmatic approach: have these things been translated or rocked the scientific world? Yes? They have a lot of meat and are actionable. No? Might be interesting but could also run the gamut of noise/spurious correlation/loose statistics/experimental bias/terrible SNR/just plain hard, none of which necessarily speaks bad of the research (well, the loose statistics and poor experimental design does), just that the claims are hardly backed up.
Last edited:
Really,do I?
How should I interpret the comment that I would accept results only if they agreed with my beliefs.
JC likes to tell the story of his customers insisting he remove the 711 from an offset servo, usually as an insult. OK, possibly just a coincidence.
I don’t want to pour fuel on an existing fire, but just want to mention that a Pet-scan as used by Oohashi is not at all the same as a fMRI.
A Pet-scan, that needs radio active glucose to be injected, does not produce an image, and that’s why it is combined with an MRI to “paint” the detected spot in the MRI image.
It is a fascinating technology when used properly such as for detection of cancer cells.
Hans
A Pet-scan, that needs radio active glucose to be injected, does not produce an image, and that’s why it is combined with an MRI to “paint” the detected spot in the MRI image.
It is a fascinating technology when used properly such as for detection of cancer cells.
Hans
Op amps are definitely better today, yet we still do miking with a Shure SM57 and we still can't abandon compressors before ADC unless we go synth all the way .
Well... i do have original Signetics, Philips, early Texas instruments and late texas instruments ne5534 and 5532 , but i won't dive into discussing it cause that is really boring.
Originally Posted by hhoyt View Post
To reiterate: the vast majority of recordings people are using to evaluate their hi-end playback systems have been passed through many op amps before being delivered in a consumer playback format.
Which is exactly why I suggest you directly compare using real acoustic sounds played in your listening room. Like a familiar voice or cymbal, piano, etc. Not a recording. Listen to the quality difference (in near-field).
THx-RNMarsh
IMO: IC opamps - VFA/CFA are really good now. If we could just get them into everyday consumer level electronics and in recording consoles ($$$).
To reiterate: the vast majority of recordings people are using to evaluate their hi-end playback systems have been passed through many op amps before being delivered in a consumer playback format.
Which is exactly why I suggest you directly compare using real acoustic sounds played in your listening room. Like a familiar voice or cymbal, piano, etc. Not a recording. Listen to the quality difference (in near-field).
THx-RNMarsh
IMO: IC opamps - VFA/CFA are really good now. If we could just get them into everyday consumer level electronics and in recording consoles ($$$).
Last edited:
Scott, you got it wrong. I started using the AD711(12) for servos and I used them for many years. It was only when I tried to use the AD712 for an INPUT STAGE, ahead of a discrete power amp that I got into trouble. That was the primary difference between the Parasound HCA2200 and the HCA2200 MK2. I took out the AD712 from the input but left them (usually 711) in the servos of the same product.
Hey Bill,
I think it is a thing (I know 5% is just a guesstimate) because it certainly makes sense, I don’t base that on marks comments alone but after a couple years of digging further into audio than I ever have before I’m finding the same results all over.....there’s a small percentage that pay better attention, maybe hear differently or could be some are better connected to their subconscious? I do some of my best testing when I get lost in the music.
Whatever it is, it’s real.
Yes, it's very often combined with a CT scan also. I have worked on an infusion system for PET contrast with a very short half-life. It doesn't cross the blood-brain barrier like FDG, so it's for other organs.
Tournesol, what was the last time you've dealt with modern music production?
Just google Neve 1073 and you'll see.
Or watch this movie in full:
YouTube
Its not only about the vibe but about the console sound too.
P.S. My respect for Paul McCartney grew enormously after seeing him play "that" type of music.
P.P.S. This one's about the original studio
YouTube
It might give some of the readers here an insight in "how it's done"
Last edited:
If i remember right from your last discussion we had in 2019, you said that Parasound's engineers asked you to recommend an op-amp for one of their phono preamps and that you chose to advise them of using ad712... Nobody told you to ask Scott for a recommendation of a good op-amp...There was nothing in the AD712 that would have recommended it for a high quality phono preamp...I wonder if you even knew Scott at the time ...For me it sounds ridiculous to take Scott's side in this argument.I see no reason for why he's even talking to you or me.It's just that the internet allows for that , but honestly ...we're light years away from each other , in any possible way...
I think it is a statistical thing. A 2 standard deviation from normal covers 95.44% of population. The rest 4.56% are those beyond 2 standard deviation from normal, half of of which (2.28%) are exceptionally good while the other 2.28% are exceptionally bad.
Pouring much resources into research targeting the most difficult to satisfy 2.28% of population would be foolish. Expect to find very sparse hard data beyond that applicable for the 95.44% and learn to patiently sift through the mountain of anecdotes and snake oil.
One interesting thing I noted when looking up that record on Discogs is that they had an engineer from EMI on board. Maybe that's the reason it sounds different? EMI classical releases from '60s and '70s usually sound very good.
Igor Strawinsky* — London Symphony Orchestra* * Claudio Abbado - Der Feuervogel (Firebird) * Jeu De Cartes (1975, Vinyl) | Discogs
Michael Gray | Discography | Discogs
i don't understand where you want to go. I was referring about what happens in the music industry in France. All the big studios disappear one after the other and protools is near everywhere. The business is no more the music, but movies and TV post production. and even in this activity, business is very critical. The companies don't survive very longTournesol, what was the last time you've dealt with modern music production?
Just google Neve 1073 and you'll see.
You are right on one point, I switched to the movie industry a little more than 20 years ago and after that was the technical manager of several post production facilities in Paris. Only doing few music sessions (recording and mixes) for old friends after that as a free lance.
I had worked a lot on the various versions of the 1073 and their 5 different preamps and don't need to "google" about. They are copied everywhere and slides used as effects in various studios.
It was my prefered mixing desk ever. For me, things started to go wrong with SSL. (the last i worked on was a SSL C300 HD)
On my side, I refused to work on digital mixing desks when it was possible. A very different approach. But it is just a personal preference. Nowadays, it seems the market is at Harrison MPC 5d, Avid S6 etc... That is why I was talking of nostalgia.
I don't follow what happens in the US musical industry. It it not the center of the world neither and I find the actual level of the musical production very poor on average. ;-)
Last edited:
All are subjectivists. Because objectivists mostly can not understand the relationship of measurement and the sound.
Dreamth, I think that you got that wrong. I have never recommended the AD711-12 for phono or anywhere but servos, once I had trouble with it as an input buffer for the HCA2200 power amp, and that was over 25 years ago. I did recommend the AD797, which is entirely another design, for the input phono stage for the JC-3.
- Status
- Not open for further replies.