KVR Audio

farlukar · Post by **farlukar** » Wed Apr 05, 2006 6:52 pm

phase difference does not equal wave length

nuffink · Post by **nuffink** » Wed Apr 05, 2006 6:55 pm

friteuse wrote:
farlukar wrote:Being able to hear up to 24 kHz - not very likely but also not impossible, certainly at younger age.
But I can hardly imagine it making a lot of difference in a piece of music OTOH.
We do often forget, that we don't only hear frequencies as notes, but as phase (or location information) too. Recent experminets showed, that we are able to perceive run time differences (uh, is this the right word in engl.?) of 5-10 µs between left and right ear, corresponding to a freqeuncy of 50-100 kHz (see german "Studio Magazin" 12/05). So, we cannot hear a note at this frequency, but still an information, namly the location of a sound source. This information gets lost when working with sample rates < 200 kHz, so 192 kHz should be a good choice to leave most of them in the source signal...

Some confusion here.

Aleksey Vaneev · Post by **Aleksey Vaneev** » Wed Apr 05, 2006 6:57 pm

friteuse, do you have a link to that paper in English? I'm just curious - to which perception channel that 'ultrasonic phase difference detection' maps.

bduffy · Post by **bduffy** » Wed Apr 05, 2006 7:04 pm

dimme3 wrote:Is there such thing as a dual band EQ, who can equalize Left & Right side seperate? if yes, which ones?

Many do, these two come to mind first: http://www.elementalaudio.com/cgi-bin/s ... nch-fireqm

deaf dunderkwac · Post by **deaf dunderkwac** » Wed Apr 05, 2006 7:38 pm

nuffink wrote:
friteuse wrote:
farlukar wrote:Being able to hear up to 24 kHz - not very likely but also not impossible, certainly at younger age.
But I can hardly imagine it making a lot of difference in a piece of music OTOH.
We do often forget, that we don't only hear frequencies as notes, but as phase (or location information) too. Recent experminets showed, that we are able to perceive run time differences (uh, is this the right word in engl.?) of 5-10 µs between left and right ear, corresponding to a freqeuncy of 50-100 kHz (see german "Studio Magazin" 12/05). So, we cannot hear a note at this frequency, but still an information, namly the location of a sound source. This information gets lost when working with sample rates < 200 kHz, so 192 kHz should be a good choice to leave most of them in the source signal...
Some confusion here.

It means that at 44.1 kHz sampling rate (cd quality) there is a perceptable time interval difference between the left/right channels. At 192kHz the difference is in-percepable (except by camsr, of course

)

This is a known and much-discussed anomoly (not necessarily a kvr discussion) about this. The 44.1 standard and how the bits are aligned in the serial stream was developed before we knew just how accurate our hearing is when it comes to time difference.

farlukar · Post by **farlukar** » Wed Apr 05, 2006 7:49 pm

dunder wrote:It means that at 44.1 kHz sampling rate (cd quality) there is a perceptable time-lag between the left/right channels. At 192kHz the difference is in-percepable

Direction perception is about the time interval between the arrival of the soundwave at your left and right ear, has fsck all to do with frequency.

Next thing you know we'll be using a delay for EQing...

jupiter8 · Post by **jupiter8** » Wed Apr 05, 2006 7:55 pm

farlukar wrote:
dunder wrote:It means that at 44.1 kHz sampling rate (cd quality) there is a perceptable time-lag between the left/right channels. At 192kHz the difference is in-percepable
Next thing you know we'll be using a delay for EQing...

Hate to break it to you but we are.

deaf dunderkwac · Post by **deaf dunderkwac** » Wed Apr 05, 2006 8:08 pm

jupiter8 wrote:
farlukar wrote:
dunder wrote:It means that at 44.1 kHz sampling rate (cd quality) there is a perceptable time-lag between the left/right channels. At 192kHz the difference is in-percepable
Next thing you know we'll be using a delay for EQing...
Hate to break it to you but we are.

what jupiter8 said,

sorry, didn't take time to read throught the whole thread properly, but the time-lag thingy is sometimes perceved as a anomoly in the frequency content, when in reality it's just the time-difference. (I know I heard this effect as such, with an additional not-quite centered phantom image in the stereo, and when I reverse my headphones, the anomoly follows.)
Only really heard by the golden-eared group (and those who wonder why things sound different from one stage to another in the production process)it's meaningless to the rest of the world.

Anyway, I'm just repeating PBS studies and my own experience when designing audio-file server systems (those derned radio announcers could hear every little thing, even the stuff that wasn't really there

)

jupiter8 · Post by **jupiter8** » Wed Apr 05, 2006 8:17 pm

How hard is the percieved timing difference between channels to solve technically ? Not at all!!!
You put a 1 sample delay on one channel.If this were a big problem it would have been solved years ago,if it isn't already. Raising the sampling rate to 192 to solve this would be like cracking a nut with nuclear bomb.

deaf dunderkwac · Post by **deaf dunderkwac** » Wed Apr 05, 2006 8:20 pm

I believe it was solved by most of the newer d>a convertors. Over-sampling and resampling and all that crap. But I got a sound card or two...

deaf dunderkwac · Post by **deaf dunderkwac** » Wed Apr 05, 2006 8:26 pm

And really, to my ears when I was testing diffferent sample rates and bit rates , the HUGE difference is quality happened when you went from a 16bit@44.1kHz to 20bit@44.1kHz. Any additional improvement was very tiny indeed.
That said, listening to 192kHz @ 24 bit was like listening to a piece of wire.
The point to be made is... who gives a rats ass, when the bulk of your audience was using cassettes and now mp3 players. (and cd's)

Kingston · Post by **Kingston** » Wed Apr 05, 2006 8:29 pm

jupiter8 wrote:How hard is the percieved timing difference between channels to solve technically ? Not at all!!!
You put a 1 sample delay on one channel.If this were a big problem it would have been solved years ago,if it isn't already.

It's not quite as simple as that, but yes, it was solved years a go. It's called an allpass filter, sometimes referred to as fractional delay. Those can be somewhat difficult to do in high quality without high sample rates, but oversampling can overcome it easily. And yes, the perceived phase differences are in the end all time based. 44.1khz can do it all, but not so well with prosumer level AD/DA and badly programmed plugins.

As a related sidenote, this is one of the reasons there isn't a hi-end phaser plugin available yet. None of them have oversampling unfortunately.

jens · Post by **jens** » Wed Apr 05, 2006 8:37 pm

Kebmaster wrote:Remember, another heavily overlooked factor is cables and leads. More expensive cable tends to add less colour than cheap cable. Cheaper cable tends to sound more muddy and bass heavy.

yeah, right....

jens · Post by **jens** » Wed Apr 05, 2006 8:40 pm

jupiter8 wrote:At least there is no Placebo left for the rest of us.I believe this guy drank the whole bottle.

friteuse · Post by **friteuse** » Wed Apr 05, 2006 8:46 pm

Aleksey Vaneev wrote:friteuse, do you have a link to that paper in English? I'm just curious - to which perception channel that 'ultrasonic phase difference detection' maps.

No, I'm afraid, I only have a printed version in german...

But if you are interested, I could scan it for you. I also could search for you the desired information in the article. Just explain me, what you exactly mean with "perception channel". Are you thinking of frequency perception / direction perception? It's again my weak english, damn!

At least, I will give you some information about the related papers and persons - but not at the moment, because the magazin is in the studio and I'm at home...

Analog summing emulation idea