The No.1 Website for Pro Audio
Why does everyone want 96khz ?
Old 15th December 2019
  #991
Deleted 2ef94c5
Guest
Quote:
Originally Posted by onewire View Post
No thanks. Don't know what a penis envy shroom is to be honest.
Just like you, I was joking as well.

Or was I?
Old 15th December 2019
  #992
Gear Addict
 

Quote:
Originally Posted by Deleted 2ef94c5 View Post
I know. I am just busting your balls. But I do think you should take an honest listen every now and then. Have you tried testing any of your theories out in the real world?
These are not my theories.
I wish I was smart enough to come up with this stuff.
I'm just relating these things to you. They have been tested, measured and confirmed for decades, in the real world.
The fact that you might have heard about some or all of them for the first time now is disheartening, if I have to be honest.

Except the binaural stuff. Nobody really cares about binaural and it is totally acceptable that you might not have wondered about it.
Old 15th December 2019
  #993
Deleted 2ef94c5
Guest
Quote:
Originally Posted by sax512 View Post
The fact that you might have heard about some or all of them for the first time now is disheartening, if I have to be honest.
You're disheartened by me never hearing that a speaker performs worse off (is less accurate) when sent ultrasonic frequencies?

As if analog consoles and tape machines were doing it all wrong for over half a century and no one seemed to notice or care...

What am I missing?
1
Share
Old 15th December 2019
  #994
Lives for gear
Quote:
Originally Posted by mpr View Post
This is where we differ. To me it is all about something sounding good, therefore accuracy is of secondary importance to sounding good. If I cant feel it in my chest, then chances are you wont either.

And I dont need perfect accuracy to make music translate. That's another illusion I had to let go of in order to get better at this game. Accurate to a point, yes, but emotionally even further.
But the emotions are in the recorded, mixed & mastered and medium bound result. Would more objective accuracy make that go away? I think not.
Old 15th December 2019
  #995
Gear Addict
 

Quote:
Originally Posted by mpr View Post

What am I missing?
An understanding of how speaker drivers work and perform.
Old 15th December 2019
  #996
Gear Guru
 
Brent Hahn's Avatar
 

Quote:
Originally Posted by Mikael B View Post
But the emotions are in the recorded, mixed & mastered and medium bound result. Would more objective accuracy make that go away? I think not.
Recording something "objectively accurately" is often a good way to make it suck.
1
Share
Old 15th December 2019
  #997
Gear Addict
 

Quote:
Originally Posted by Brent Hahn View Post
Recording something "objectively accurately" is often a good way to make it suck.
You might be surprised to hear that I completely agree with this.

Accuracy in recording is something to pursue with extreme caution, and it only works in very specific scenarios. Even an accuracy buff like me knows that.

In the reproduction chain though I personally find an almost direct correlation between how accurate a system measures and how much I happen to like it.
The only exception being the speaker voiced flat. I do not like a speaker that is equalized flat (at the listening position, that is).
Old 15th December 2019
  #998
Lives for gear
 
IanBSC's Avatar
Keep in mind, the "accuracy" Sax is describing is not the input and output of the converter sounding the same. It's the notion that lower sample rates are improving the accuracy of signal path by ridding your speakers of intermodulation Gremlins. And if it sounds less accurate to you, or if the input signal (or the high sample rate) sounds more accurate, it's only because you love your speakers distorting.
1
Share
Old 15th December 2019
  #999
Deleted 2ef94c5
Guest
Quote:
Originally Posted by sax512 View Post
An understanding of how speaker drivers work and perform.
In basic laymen terms, can you please explain (once more) why we would want to limit frequencies as they exist in nature, emanating from instruments and such, so that a speaker can be more accurate recreating content found strictly within the human hearing band?

To achieve perfect playback, shouldn't the speaker be allowed emanate all the frequencies of the cello, including the overtones existing above the hearing band?

Quote:
Originally Posted by IanBSC View Post
Keep in mind, the "accuracy" Sax is describing is not the input and output of the converter sounding the same. It's the notion that lower sample rates are improving the accuracy of signal path by ridding your speakers of intermodulation Gremlins. And if it sounds less accurate to you, or if the input signal (or the high sample rate) sounds more accurate than the converted, it's only because you love your speakers distorting.
Interesting.
1
Share
Old 15th December 2019
  #1000
Gear Guru
Quote:
Originally Posted by Deleted 2ef94c5 View Post
Wow you are in luck. I am in San Diego right now and have some very rare penis envy shrooms that will definitely do the trick. Hit me up.
Ha! Those sand worms that look like weiners?! Man that’s nuts mini cyclone thing.... gulls must be loving it!....
2
Share
Old 15th December 2019
  #1001
Lives for gear
 

This is MPR Radio! (I think therefore AM).
Chris
Old 15th December 2019
  #1002
Gear Nut
Quote:
Originally Posted by Deleted 2ef94c5 View Post
In basic laymen terms, can you please explain (once more) why we would want to limit frequencies as they exist in nature, emanating from instruments and such, so that a speaker can be more accurate recreating content found strictly within the human hearing band?
Monty has already explained it very clearly:

https://people.xiph.org/~xiphmont/de...g.html#toc_1ch
5
Share
Old 15th December 2019
  #1003
Deleted 2ef94c5
Guest
Quote:
Originally Posted by mfnickster View Post
Monty has already explained it very clearly:

https://people.xiph.org/~xiphmont/de...g.html#toc_1ch
Thank you for that. So essentially after some 70 years of sending ultrasonic frequencies to amps and speakers it was an accidental byproduct from a strict nyquist cutoff just outside the human hearing band that resulted in better speaker performance?
1
Share
Old 15th December 2019
  #1004
Lives for gear
 

Very interesting article...
Chris
Old 15th December 2019
  #1005
Gear Nut
Quote:
Originally Posted by Deleted 2ef94c5 View Post
Thank you for that. So essentially after some 70 years of sending ultrasonic frequencies to amps and speakers it was an accidental byproduct from a strict nyquist cutoff just outside the human hearing band that resulted in better speaker performance?
I don't know, but before CDs took over virtually no one cared about performance above 20 kHz. Look at any spec sheets from gear of that era, the frequency response graphs all go no higher than 20 kHz.

Some amps were rated well above that because better tolerances iDeleted 2ef94c5oved performance in the audible range, but speakers in general were not designed to reproduce ultrasonics. They were considered irrelevant.
1
Share
Old 15th December 2019
  #1006
Gear Addict
 

Quote:
Originally Posted by chessparov2.0 View Post
Very interesting article...
Chris
Indeed. But let's not make the same mistake I made 3 years ago..

Did anybody run it by Plush to see if He agrees?
Old 15th December 2019
  #1007
Gear Nut
Quote:
Originally Posted by Plush View Post
Dither is always necessary to maintain high quality in a digital audio platform. It should be enabled globally on your daw for every operation, including playback.
...

People who don't realize this are automatically getting a much lower quality of sound.
Interesting. Here's Bob Katz's take on it:

"The Cost of Cumulative Dithering"

"When feeding processors, DAWs or digital mixers to your recording unit, dither the output of the processor to a 24-bit word. Dithering always sounds better than truncation without dither. But to avoid adding a veil to the sound, avoid cumulative dithering, in other words, multiple generations of any dither. Make sure that redithering to 24- or 16-bit is the one-time, final process in your project."

https://www.digido.com/portfolio-item/dither/
1
Share
Old 15th December 2019
  #1008
Lives for gear
 

I hear you Sax. All my kidding aside, I do have high respect for Plush's level of accomplishment. And as we know, nothing beats getting to know someone in person.
The Internet is very 2D.

One of my close friends, is rather the Curmudgeon outwardly. But has a very kind heart, when you get to know him, in person.
Chris
1
Share
Old 15th December 2019
  #1009
Lives for gear
 

Quote:
Originally Posted by mfnickster View Post

Monty has already explained it very clearly:

https://people.xiph.org/~xiphmont/de...g.html#toc_1ch
So 192 is to be avoided at all costs. 96 is worthless. 16bit won't kill you. 24/48 is the safe spot. Mp3 with ill-chosen conversion parameters is your own fault. If your interface gear is crummy, it's your own fault and nothing applies no matter what bit depth or sample rate you dial in.

And now that Naugles is back from the dead with three locations, Del Taco should eventually be superseded.
Old 15th December 2019
  #1010
Lives for gear
 
IanBSC's Avatar
Quote:
Originally Posted by mfnickster View Post
Monty has already explained it very clearly:

https://people.xiph.org/~xiphmont/de...g.html#toc_1ch
In sample rate discussions the term "woo woo" is often thrown about, and here we are fretting about something that literally nobody actually knows if they have heard, that nobody can identify or describe, and that to my knowledge nobody has even measured in their own speakers. And some are even using this phantom distortion to promote the notion that if your ADC/sample rate sounds dissimilar to the input signal, it means that it is more accurate!
1
Share
Old 15th December 2019
  #1011
Gear Nut
@ mpr Dan Lavry suggests the optimal sampling rate would be around 60 kHz, for the same kind of reasons you're citing: it's better to include everything even beyond the "normal" range of hearing, simply because some instruments do produce those high harmonics and it's better to have a good margin of error. Since 60 kHz is not usually an option in digital recording, 88.2 kHz or 96 kHz would be the next best choice:

http://www.lavryengineering.com/pdfs...lity_audio.pdf

He does, however, make a point that I believe to be in error (I'm certainly no expert): that limiting every link in the chain to 20 kHz (mic, A/D, D/A, speaker) yields a combined impact of -12 dB @ 20 kHz. Oversampling A/D & D/A converters as commonly used these days don't attenuate that much because the transition band is way above 20 kHz; digital brick-wall filtering of a 4x or 8x oversampled signal can be very sharp and leave the 20 kHz frequencies unaffected. Monty's video demo even shows zero dropoff for a 20 kHz signal sampled at CD quality.
2
Share
Old 15th December 2019
  #1012
Lives for gear
 
IanBSC's Avatar
Quote:
Originally Posted by mfnickster View Post
Dan Lavry suggests the ideal sampling rate would be around 60 kHz, for the same kind of reasons you're citing: it's better to include everything even beyond the "normal" range of hearing, simply because some instruments do produce those high harmonics and it's better to have a good margin of error. Since 60 kHz is not usually an option in digital recording, 88.2 kHz or 96 kHz would be the next best choice:
I hear this all the time, but I have a really hard time believing that if I split the difference between 48khz and 88.2khz that the sound would get better than 88.2, 96 or even 192 for that matter.
1
Share
Old 15th December 2019
  #1013
Gear Nut
Quote:
Originally Posted by IanBSC View Post
I hear this all the time, but I have a really hard time believing that if I split the difference between 48khz and 88.2khz that the sound would get better than 88.2, 96 or even 192 for that matter.
I really doubt that the sound would get "better" at 60 kHz (I don't believe the ultrasonics impact enjoyment of music listening at all) but his point is that you can't reach 30 kHz bandwidth with 44.1 kHz or 48 kHz sampling rates.

Edit: apologies to sax512 for bringing it up again, I see you discussed it here a couple weeks ago!
2
Share
Old 15th December 2019
  #1014
Deleted 2ef94c5
Guest
Quote:
Originally Posted by mfnickster View Post
(I don't believe the ultrasonics impact enjoyment of music listening at all)
I agree with this, and the fact that the differences that i do hear between 44.1 and 96k are almost entirely converter dependent says way more about the converter than the rate. I can definitely see why people prefer 96k if they have done shootouts using 95% of converters out there.

Again, the only reason why I upsample to 96k is because I am strung out on plugins more than ever. Also I have noticed that plugins with internal oversampling do not always sound better than when fed a 96k source with oversampling turned off - VERY often the case for some reason.

Quote:
Originally Posted by mfnickster View Post
I don't know, but before CDs took over virtually no one cared about performance above 20 kHz. Look at any spec sheets from gear of that era, the frequency response graphs all go no higher than 20 kHz.
I dont think the old components were rolling off so i imagine they just weren't measuring that high right?

My old neve amps pass (and even create) harmonics so high you could make music for porpoises with them. No joke.
1
Share
Old 15th December 2019
  #1015
Gear Nut
Quote:
Originally Posted by Deleted 2ef94c5 View Post
I dont think the old components were rolling off so i imagine they just weren't measuring that high right?
Very likely, but my point is that in the pre-digital era, nobody was concerned about reproducing those ultrasonics to give the music more depth/life/space/warmth/clarity/detail or any number of other magic sauces that digital is supposedly missing. Why not?
Old 15th December 2019
  #1016
Lives for gear
 
IanBSC's Avatar
Quote:
Originally Posted by mfnickster View Post
Very likely, but my point is that in the pre-digital era, nobody was concerned about reproducing those ultrasonics to give the music more depth/life/space/warmth/clarity/detail or any number of other magic sauces that digital is supposedly missing. Why not?
Right. And the point is that ultrasonics don't add anything special (or explode your speakers), the only reason for high sample rates is better sounding filters. Why equipment manufacturers do it, I'm not really sure, but on the whole modern speakers sound better than pre-digital.
1
Share
Old 15th December 2019
  #1017
Gear Nut
Quote:
Originally Posted by phanlon View Post
On top of that I’ve worked on some tracks that went to one of the hugest mixers in the game who said never send me anything at 88.1 again
Well, that's probably fair...88.1 is much less common than 88.2.
1
Share
Old 15th December 2019
  #1018
Lives for gear
 
IanBSC's Avatar
Quote:
Originally Posted by thenoodle View Post
So 192 is to be avoided at all costs. 96 is worthless. 16bit won't kill you. 24/48 is the safe spot. Mp3 with ill-chosen conversion parameters is your own fault. If your interface gear is crummy, it's your own fault and nothing applies no matter what bit depth or sample rate you dial in.
What happens if you fail to avoid 192?
Old 15th December 2019
  #1019
Gear Addict
 

Quote:
Originally Posted by mfnickster View Post

Edit: apologies to sax512 for bringing it up again, I see you discussed it here a couple weeks ago!
Don't worry. I am used to people not even reading what I say around here. And when they do, but it doesn't jive with their wrong pseudo-science preconceived notions, I'm used to childish pouting like the latest comments from IanBSC.

People have a hard time coping with science, when it slaps them in the face and calls them out on their BS.

If you hang around this kind of people on GS (impossible to avoid them) you will experience this phenomenon first hand.
1
Share
Old 15th December 2019
  #1020
Quote:
Originally Posted by norfolk martin View Post
A higher tape speed is just like a higher sampling rate as more oxide particles are passing in front of the head in a certain time frame. Those millions of particles are why tape has a much higher resolution than digital audio, which only has a few bits
The very best tapes could barely hit 13 bit equivalent accuracy due to tape hiss. And this is with advanced noise reduction added in.

According to Monty Montgomery:
"That's why DDD on a compact disk used to be such a big high end deal."
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump