Skip to main content

Topic: Francinstien audio proccessing, is this acoustically valid? (Read 19025 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
  • 2tec
  • [*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
I ran across this and immediately thought of HA. 

"this circuit could be fitted onto the output of any CD player or DAC and be used to enhance the image quality from digital." Francinstien

As proof, he offers two audio files, pre and post processed. The two mp3s are here:
http://www.richardbrice.net/demo.htm

As well, he makes this observation, "Unhesitatingly, I should say, the CD was 'nearer' the master, but vinyl produced a 'better' stereo image - in fact, better than the master tape! A conundrum indeed."

Well folks, what do you think?
Quis custodiet ipsos custodes?  ;~)

  • Jens Rex
  • [*][*][*][*][*]
  • Global Moderator
Francinstien audio proccessing, is this acoustically valid?
Reply #1
I don't even have to check the link to realize this guy is full of shit.

  • MichaelW
  • [*][*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #2
I, too, would be grateful if people who know about the technicalities would  look at this.

The page in question is plain text, with only one image. I can't evaluate what he's saying, but the language is at worst very similar to the language of people who know what they're talking about. He's also a published author (Music Engineering: The Electronics of Playing and Recording / By Richard Brice / Published by Newnes, 2001 / ISBN 0750650400, 9780750650403). Newnes aren't a vanity press.

If he *is* full of shit, it's *quality* shit. 

EDIT
AFAICT, it's a proposal for some kind of cross-feed, but applied to speakers, not just phones.
  • Last Edit: 11 October, 2008, 04:56:20 PM by MichaelW

Francinstien audio proccessing, is this acoustically valid?
Reply #3
C'm on Jens, tell us what you really think

I only skimmed it, but I think the stuff about interaural delay and bass notes is bullsh!t.  The distance between my ears is 0.2m (yes, I measured it, how sad is that? Some say it is a lot less).  If your speakers are, say, 3m apart and you sit, say, 2m from them, pythagoras says that the difference in path lengths is about .15m.  Speed of sound is ~ 340 m/s so the delay he is talking about is a little under 0.5 mS.  This is a phase angle of about 20 degrees at 125 Hz.  I really don't believe that is audible, given the way the sound is bouncing round the room anyway.

The 'explanatory diagrams' greatly exaggerate the difference in path length.  I partularly like the way the sound is going round corners.  It makes me wonder whether having a long nose would affect the outcome of an ABX test.  And did you know that, statistically speaking, 99.9% of the human population have more than the average number of legs?

Anyway, I don't believe it does anything useful.  As for me, I think I need to get out more
  • Last Edit: 15 October, 2008, 06:14:26 AM by Paul Sanders
I am an independent software developer (VinylStudio) based in UK

  • thundat00th
  • [*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #4
off topic: sorry my profile pic is not of a cat

on topic: Tried to read the article, he seems to have this "phenomenon" understood, why i need a box to fix the deficiencies of having ears on either side of my head is beyond me

then he said that vinyl has better imaging then CD and master tape, and i stopped reading
  • Last Edit: 12 October, 2008, 03:32:57 PM by thundat00th
My $.02, may not be in the right currency

  • MichaelW
  • [*][*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #5
then he said that vinyl has better imaging then CD and master tape, and i stopped reading


Yes, but he's not an ISO standard audiofool, and that statement was meant to be a provocative paradox. I had a quick look at bits of his book on Google Books. He knows quite well that properly done digital and solid state is more linear than analogue and firebottles. But because he's involved (probably more involved) in music creation as well as reproduction, he also knows that musos sometimes choose distortion (try telling a guitarist that they'd do much better with a nice solid state amp than their carefull preserved, over-driven, vintage Marshall).

The specific claim is that some distortion introduced in the cartridge of a vinyl player happens to compensate for imperfections in stereo imaging. I don't myself give a rat's whisker about sound staging, and I am quite content to be told that his technical explanations don't stack up. Sound staging seems to be a pretty hard thing to pin down, and up there with jitter as a reason for extracting large sums of money from the gullible. But he doesn't seem to be selling anything, and he didn't trigger my bullshit meter. And, after 40 years in the University-level study of English Literature, I have a well-exercised, finely tuned bullshit meter.

As I say, I'm quite happy to see that the technical stuff doesn't work out. But his stuff looked worth checking out, rather than using in an OMG audiofools!1l! thread.

EDIT
I see he *is* selling something. Sorry.
  • Last Edit: 12 October, 2008, 04:06:39 PM by MichaelW

  • thundat00th
  • [*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #6
As I say, I'm quite happy to see that the technical stuff doesn't work out. But his stuff looked worth checking out, rather than using in an OMG audiofools!1l! thread.


i agree, my post was more gut reaction then well thought out rebuttal heh
My $.02, may not be in the right currency

  • Roseval
  • [*][*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #7
|From the Richard Brice website:
Finally, the Perfect Pitch Music products are no longer in production
TheWellTemperedComputer.com

Francinstien audio proccessing, is this acoustically valid?
Reply #8
...why i need a box to fix the deficiencies of having ears on either side of my head is beyond me
That made me chuckle! Thanks.

Cheers, Slipstreem. 

  • MichaelW
  • [*][*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #9

As I say, I'm quite happy to see that the technical stuff doesn't work out. But his stuff looked worth checking out, rather than using in an OMG audiofools!1l! thread.


i agree, my post was more gut reaction then well thought out rebuttal heh


Whoops, I didn't mean to be polemical. 

Actually, I only got interested in this guy's thought from skimming a few pages of his book. His line seems to be that digital is the way to go for the most accurate reproduction of what's on the master tape, but that some people prefer certain inaccuracies (Hint: in the table of contents of the book, the heading "The Valve Sound" is immediately followed by "Harmonic Distortion").

If that point of view were accepted, it could lead to an outbreak of common sense. For starters, with clarification of expressions like "<something> sounds better." Clearly, in the world of HA, that is pretty unambiguously a statement about accuracy. In some other worlds, it could mean "sounds preferable to me," or "sounds preferable to a large number of people," which is also a reasonable criterion. It's a pity that for many people with more money than sense, the search for a sound pleasing to individual taste is masked as pursuit of some magical notion of accuracy.

At the moment, people are spending money on products that are fraudulent, because there is no way they could have an effect on anything (magic pebbles, rainbow foils), or are useless (like speaker cables that might have superior performance at microwave frequencies). Wouldn't It Be Nice If the rational engineers could persuade the audiophiles that what they really want is a linear system with a bit of discreet signal processing? Then a lot of wasted money and effort could be put to more productive purposes. Like solving the financial crisis

  • thundat00th
  • [*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #10
well said
My $.02, may not be in the right currency

  • Iain
  • [*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #11
The problem, he is suggesting, is not that we have an ear on each side of our head, but that the sound waves reaching our ears from a (intensity stereo i.e. panned) phantom source are affected differently by our head (HRTF) than if there was an real sound source in the phantom location.  He says this causes a narrower stereo image at low frequencies.

It seems his black box increases the stereo width of low frequencies. He suggests that this makes the  stereo width at low frequencies match the stereo width high frequencies, properly aligning the postion of the low and high frequency componets of the sound source. This can't be true for all cases though. Consider a signal panned hard left (that is, there is no signal in the right channel) in this situation there is no 'phantom' image as the image source is the same as the loudspeaker location, so there is no error in stereo width between the low and high frequencies. Also consider a signal panned centre (the same signal in both speakers) here agian there is no difference in stereo width between the low and high frequencies. Possibly there is an issue in between these two extremes, but I am not convinced.

He box may 'improve' the sound by increasing stereo width at low frequencies, but I am skepitcal about his explanation and reasoning.

  • MichaelW
  • [*][*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #12
I don't want to bang on, but I got interested in this guy Richard Brice because I like the way he writes in his book Music Engineering.

On his site, he lists a number of apparently impressive positions. On the internet, of course, you can claim to be Marie of Romania, but he lists his current job as President, Miranda Technologies, Hong Kong. This he is (http://investing.businessweek.com/research/stocks/private/snapshot.asp?privcapId=1502521), and his total compensation is CAD358.7K, about half that of the CEO of the parent. The company does capital goods for broadcasting, and apparently has got a contract for the introduction of HDTV to Hong Kong.

This, of course, doesn't mean that his stereo fiddling device is any good, but he is certainly not your average wanker. He's also got some bits of music on his site that are licensed under Creative Commons. They're not Steve Reich, but they're competent sound-track music, that might be useful for a low budget movie or a presentation.

Which all means that if he is wrong, it might well be interesting to work out why he's wrong, as Iain has.

  • 2Bdecided
  • [*][*][*][*][*]
  • Developer
Francinstien audio proccessing, is this acoustically valid?
Reply #13
This is, at least, interesting.

Note that he doesn't sell any of this stuff any more - these pages remain on his website simply because people still ask questions about the equipment.

http://www.richardbrice.net/
http://www.richardbrice.net/ppm.htm

It sounds like a trivial thing to do in DSP.

It's too early on a Monday morning (ish  ) to think through the psychoacoustics.

EDIT: OK, it's afternoon now!

I've never understood how that explanation of stereo (which explains how intensity stereo creates interaural delays) is supposed to work for transient signals. It's fine for sine waves, but useless for dynamic signals and signal onsets.

Also, the big theoretical "problem" with stereo is comb filtering. You have the major comb filtering due to the signal from each loudspeaker bouncing off the floor and constructively and destructively interfering with itself - playing white noise through a reasonable stereo system will let you hear this as you move slightly around the room, or just stand up / sit down in the listening position. You also have comb filtering when any signal comes from both speakers simultaneously - this is unnatural comb filtering at each ear.

The other problem are the HRTFs imprinted on the sound - they're the ones of your ears hearing a sound at the location of the real speakers - they're not the right ones for a sound located between both speakers (which is what the speakers are trying to fake). Some producers even add a little mid-range boost to sounds in the centre to try to compensate for this.

People are so used to all this that they accept it. People prefer a phantom image at the centre to a real centre speaker - a fake source to a real one - because that's what they're used to.

So I wonder how big a differences the current box of tricks makes, even if the psychoacoustics do justify it - the psychoacoustics (and simple wave/acoustic theory!) reveal much greater problems elsewhere.

Cheers,
David.
  • Last Edit: 13 October, 2008, 08:32:14 AM by 2Bdecided

Francinstien audio proccessing, is this acoustically valid?
Reply #14
Forgive my ignorance, but is any of this related in any way to the 'spatial stereo' setting you get on some audio equipment?  If so, that always sounds dreadful to me.
  • Last Edit: 13 October, 2008, 10:06:06 AM by Paul Sanders
I am an independent software developer (VinylStudio) based in UK

  • MichaelW
  • [*][*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #15
FWIW, somewhere on his site he says that his Francistein system has been used on some movie soundtracks. Scale might make a difference??

  • Woodinville
  • [*][*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #16
Whoa...

I don't know off hand the JND for 100 Hz or so, but you might just make some files in Matlab (octave, whatever) and try various DBT's to find out.

As to LP vs. CD, there is often L-R enhancement in LP due to stylus beam resonances, and often some "image enhancement" due to distortions.

Yes, these can be ABX'ed, although I'm not really free to offer the samples, they are not at all hard to make, just remember to OVERSAMPLE before you distort!

But this does not make LP "more accurate" it's addressing a preference, and one that only some people have.
-----
J. D. (jj) Johnston

Francinstien audio proccessing, is this acoustically valid?
Reply #17
Hello to all forum members.

Since you have been discussing the FRANCINSTIEN stereo enhancer, I thought I would clarify a few points.

I was really sad to see the post which simply stated, "I don't have to visit the link to see this guy is full of shit!" What an attitude? There is also a snicker that the products are no longer in production. True....and you know the reason why?... Because I got fed up with self-appointed audio "experts" and the nonsense that got talked about in audio. I found it degenerative and hurtful - as I found some of the comments here. I left the industry and went to television. When I stopped making the device, I published all the circuits... So you can all try it for yourselves... and for nothing! I don't think that's the actions of a snake-oil salesman!

Forum,

It is my belief that the ubiquity of misinformation and prejudice in audio blinds us to many possible avenues for improvement. The truth is, there still exist many areas of sound reproduction and recording which could benefit from an open mindedness and objectivity they are presently denied. The FRANCINSTIEN enhancer was based on real science: Blumlein even noted it was necessary in the original EMI stereo system. It was also tested in double-blind, audio trials at Surrey University in the UK - and the improvement was verifiable. There aren't many audio products which would pass such a test.

I recently wrote a "Francinstien Revisited" paper which I'd be happy to send to any members.

Best wishes

Richard Brice

  • bug80
  • [*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #18
I recently wrote a "Francinstien Revisited" paper which I'd be happy to send to any members.

Yes, please. 

I'm quite interested. If I understand it correctly the basic idea is utilizing crosstalk between the channels to 'improve' the stereo image. It compensates this effect:

Quote
The result, to use a visual analogy, is a slight smearing or blurring of the stereo image.


Doesn't 'crosstalk' basically have the effect of increasing the correlation between both channels? In that sense it is logical that it results in less smearing/blurring, but the question is: is it really an improvement?

Maybe there is more to this, therefore I'm interested in your paper.

  • Soap
  • [*][*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #19
I was really sad to see the post which simply stated, "I don't have to visit the link to see this guy is full of shit!" What an attitude?
... Because I got fed up with self-appointed audio "experts" and the nonsense that got talked about in audio. I found it degenerative and hurtful - as I found some of the comments here.

I think many people had a gut reaction against the very questionable line:
"Unhesitatingly, I should say, the CD was 'nearer' the master, but vinyl produced a 'better' stereo image - in fact, better than the master tape! A conundrum indeed."

I think defending this claim instead of writing off your detractors would be healing for us all.  Since CDs have a much better channel separation than vinyl, it is hard to comprehend how vinyl could have better stereo imaging.  I, and I assume many others, took the word "better" to mean "more accurate" - which is clearly impossible.

Perhaps you ment "more pleasing"?  Or perhaps the nature of vinyl channel crosstalk is similar to the crossfeed system you propose?
  • Last Edit: 23 October, 2008, 07:40:38 AM by Soap
Creature of habit.

  • bug80
  • [*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #20
Or perhaps the nature of vinyl channel crosstalk is similar to the crossfeed system you propose?

I have the feeling that this is the case. The "imperfect" channel seperation of turntables compensates some of the problems with intensity stereo, and therefore these imperfections are simulated in the crossfeed system. But again I would like to read the paper about this.

The misunderstanding here is that on this board 'better' is equal to 'closer to the original (mastertape)', whereas Richard Brice  means 'more pleasing'.
  • Last Edit: 23 October, 2008, 08:00:11 AM by bug80

Francinstien audio proccessing, is this acoustically valid?
Reply #21
I published all the circuits... So you can all try it for yourselves... and for nothing!
As a keen experimenter in the field of analogue psychoacoustic circuit design in my younger days, I'd like to at least "build" the circuit in a CAD simulator to find out what it does from a mathematical point of view. Could you please provide a link where I can download the published circuits from? A search with Google has revealed nothing so far.

Cheers, Slipstreem. 

  • botface
  • [*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #22
Hello to all forum members.

Since you have been discussing the FRANCINSTIEN stereo enhancer, I thought I would clarify a few points.

I was really sad to see the post which simply stated, "I don't have to visit the link to see this guy is full of shit!" What an attitude? There is also a snicker that the products are no longer in production. True....and you know the reason why?... Because I got fed up with self-appointed audio "experts" and the nonsense that got talked about in audio. I found it degenerative and hurtful - as I found some of the comments here. I left the industry and went to television. When I stopped making the device, I published all the circuits... So you can all try it for yourselves... and for nothing! I don't think that's the actions of a snake-oil salesman!

Forum,

It is my belief that the ubiquity of misinformation and prejudice in audio blinds us to many possible avenues for improvement. The truth is, there still exist many areas of sound reproduction and recording which could benefit from an open mindedness and objectivity they are presently denied. The FRANCINSTIEN enhancer was based on real science: Blumlein even noted it was necessary in the original EMI stereo system. It was also tested in double-blind, audio trials at Surrey University in the UK - and the improvement was verifiable. There aren't many audio products which would pass such a test.

I recently wrote a "Francinstien Revisited" paper which I'd be happy to send to any members.

Best wishes

Richard Brice

Richard,
            Welcome to HA. I for one certainly remember you from when you were involved in the audio world.

I do hope the propensity of some members to be a bit quick with disparaging remarks doesn't put you off. I think your knowledge and experience would ultimately be very useful and helpful

  • krabapple
  • [*][*][*][*][*]
Francinstien audio proccessing, is this acoustically valid?
Reply #23

Or perhaps the nature of vinyl channel crosstalk is similar to the crossfeed system you propose?

I have the feeling that this is the case. The "imperfect" channel seperation of turntables compensates some of the problems with intensity stereo, and therefore these imperfections are simulated in the crossfeed system. But again I would like to read the paper about this.

The misunderstanding here is that on this board 'better' is equal to 'closer to the original (mastertape)', whereas Richard Brice  means 'more pleasing'.



In which case he's talking about euphonic distortion.  I like some too; the kind I like is supplied by Dolby Pro Logic II, on stereo sources.
  • Last Edit: 23 October, 2008, 11:45:42 AM by krabapple

Francinstien audio proccessing, is this acoustically valid?
Reply #24
I think defending this claim instead of writing off your detractors would be healing for us all.
If I understand the article correctly, Mr. Brice is not claiming that either CD or vinyl is a superior medium. He discovered by chance that imaging of L-R panned mono sources was more correct (all frequencies at the same spot) with vinyl as a result of increased cross talk at higher frequencies. This shortcoming of vinyl turned out to be an advantage.
To localize sounds in front of us we generally rely on inter-aural level differences at high frequencies (roughly above 1600 Hz) and timing differences at lower frequencies. Intensity panning (done on most recordings) only creates L-R level differences and hence the higher frequencies will be more or less correctly positioned, but the lower frequencies will remain closer to the center. Localization will only be correct for extreme Left or Right.
Recordings that already contain interchannel timing differences (spaced microphones e.g.) probably won't benefit from the Francinstien device.
I think this stereo playback enhancement device is definitely interesting, but it's unlikely to be a cure for all recordings. IMO it's better to do this processing during the mixing. There already are "clever" pan-pot algorithms available.
The misunderstanding here is that on this board 'better' is equal to 'closer to the original (mastertape)', whereas Richard Brice  means 'more pleasing'.
Hmmm, I think he means that even the mastertape has flaws regarding stereo imaging that can be improved.
Hopefully Mr. Brice can give some answers himself.