The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
When 192kHz isn't good enough... Digital Converters
Old 11th October 2017
  #1
Lives for gear
 

Thread Starter
When 192kHz isn't good enough...

Old 11th October 2017
  #2
Lives for gear
 
Cellotron's Avatar
 

Verified Member
The reason to use sample rates for PCM audio over 96kHz is so that you can have a much larger file size, that takes up more cpu to process, while making it sound worse. The reason people go for doing this is because of the obviously understandable belief that "bigger is better" - something that counter intuitively is simply not necessarily true with PCM audio. Unfortunately some manufacturers attempting to push their hardware add to the disinformation around these things.

http://www.realhd-audio.com/?p=1739

http://lavryengineering.com/pdfs/lav...ing-theory.pdf

Best regards,
Steve Berson
Old 11th October 2017
  #3
Lives for gear
 
Cellotron's Avatar
 

Verified Member
Let me temper the above by stating that there are valid reasons to choose DSD for recording. But once that has been converted to PCM for processing the potential benefits of that kind of go away.

Best regards,
Steve Berson
Old 11th October 2017
  #4
Lives for gear
 
Cellotron's Avatar
 

Verified Member
The DAW software might make for a nice app and an inexpensive way to acquire some Sonnox plugins regardless though.

Best regards,
Steve Berson
Old 11th October 2017
  #5
Lives for gear
 
kuasalogam's Avatar
 

$149 and includes six Sonnox plugins!
Old 11th October 2017
  #6
Gear Nut
Even when I leave DSD, and go back to pcm, the sound is better for my projects..
Old 11th October 2017
  #7
Lives for gear
 
Cellotron's Avatar
 

Verified Member
Quote:
Originally Posted by MR WOO View Post
Even when I leave DSD, and go back to pcm, the sound is better for my projects..
Was that determination made by doing a direct comparison with the same source recorded to PCM and given the same exact processing? Do you have any files of these comparisons to share?

Best regards,
Steve Berson
Old 11th October 2017
  #8
Lives for gear
 
Cellotron's Avatar
 

Verified Member
Looking further at the feature set of this app - DDP creation, batch processing, DSD recording and playback, bundled set of plugins - it looks like a really good deal for those in the need of a mastering app.

Best regards,
Steve Berson
Old 12th October 2017
  #9
Gear Nut
Quote:
Originally Posted by Cellotron View Post
Was that determination made by doing a direct comparison with the same source recorded to PCM and given the same exact processing? Do you have any files of these comparisons to share?

Best regards,
Steve Berson
Yes.... I can't share them at this time...I'll ask for permission when the project is done.
Old 12th October 2017
  #10
Lives for gear
 
inlinenl's Avatar
 

Verified Member
while I'm absolute a BIG fan of 192kHZ when capturing mastering from tape , I see reasons for DSD and higer sample rates. specially when your final master/mix format is not in the digital domain, it's nice to have some options in capturing/distributing it in different digital formats. If it makes money, it makes money !
and some people just use music to listen to their stereo's and for me there is nothing wrong with a listeners point of view, which sometimes get's ridiculed by "pro's"


anyway )) lot's of great plugs for the $
Old 12th October 2017
  #11
Lives for gear
 
Adebar's Avatar
Quote:
Originally Posted by MR WOO View Post
Yes.... I can't share them at this time...I'll ask for permission when the project is done.
I think this often depends also on the converters. The Meitner ADC8 MK IV was great at DSD with 128 FS. But the PCM output of the same unit at 24/96 wasn´t at the same level.

Now IMHO there are some PCM converters which are better at 24/192 than the Meitner ADC at DSD 128FS.

Do we need more than 192k?
I think the step from 48k to 96k is obvoiusly an improvment. The difference from 96k to 192k is smaller but with great releases on adequate systems I hear an advantage to 96k.
More than 192k? Probably no benefit.
Old 12th October 2017
  #12
Its for measurement purposes.........................
Old 12th October 2017
  #13
Divide 768 by 2 and write down what you get.

Then go do some science
Old 12th October 2017
  #14
Lives for gear
 
Cellotron's Avatar
 

Verified Member
Quote:
Originally Posted by MR WOO View Post
Yes.... I can't share them at this time...I'll ask for permission when the project is done.
That would be really nice if the client allowed a small excerpt for comparison to be released - I would be very curious to hear these.

Best regards,
Steve Berson
Old 12th October 2017
  #15
Gear Maniac
 

Quote:
Originally Posted by Nonlinear View Post
Can someone tell me why we need 768kHz sample rates?
Is it because of higher rate DSD "needing" higher rate dxd, double 352/384 ? Just asking.

Last edited by walter88; 13th October 2017 at 08:00 AM.. Reason: missing end quote
Old 12th October 2017
  #16
Lives for gear
 

Thread Starter
Quote:
Originally Posted by walter88 View Post
Is it because of higher rate DSD "needing" higher rate dxd, double 384 ? Just asking.
As I understand it, DSD records the 1-bit delta-sigma signal directly rather than decimating that data down to a 96kHz (or other) PCM sample rate. However, the 1-bit sample rate is up around 2MHz - so I don't know what we actually have here (maybe I need to research DSD more!).

It appears from the Sound It! website that 768kHz is the actual PCM sample rate it supports.

Not all VST plugins support even 192kHz - so the plugins that come with this software must either be modified versions or they are actually downsampling at the plugins (defeating the purpose).

Makes no sense to me.
Old 13th October 2017
  #17
Gear Nut
Quote:
Originally Posted by inlinenl View Post
while I'm absolute a BIG fan of 192kHZ when capturing mastering from tape , I see reasons for DSD and higer sample rates. specially when your final master/mix format is not in the digital domain, it's nice to have some options in capturing/distributing it in different digital formats. If it makes money, it makes money !
and some people just use music to listen to their stereo's and for me there is nothing wrong with a listeners point of view, which sometimes get's ridiculed by "pro's"


anyway )) lot's of great plugs for the $
The capture for me at 11.2, sounds different compared to the pcm capture.. I am using the merging hapi converter....When you utilize different A/Ds the outcome of the resulting file can sound different.I'm sure everyone has their favorite pitch and capture A/D and D/A. And over time,and with some testing, you may find something else that may tickle your fancy...
Old 13th October 2017
  #18
Gear Head
 
Wire Grind's Avatar
 

Most people won't need a sample rate as high as 768 kHz, but it is not totally useless. Higher sample rates simplify the task of accurately reconstructing both the phase and magnitude of very high frequencies. Sometimes multiple samples can be combined in such a way that there is an improvement in the SNR. It is also possible that one might acquire audio that happens to have very high sample rate. For example, say you use a specialized machine to digitize a fragile 140 year old recording. It is also conceivable that one might need the extra precision for some scientific reason.

I think the real reason, though, is that the sample rate needs to be high enough for 99.9% of all potential customers. Also, adding support for 768 kHz is often a trivial task for the software developers.
Old 13th October 2017
  #19
Lives for gear
 

Quote:
Originally Posted by Nonlinear View Post
Can someone tell me why we need 768kHz sample rates? Audio editing software Sound it! 8 Pro | INTERNET Co., Ltd.

But I have to admit it's a LOT of stuff for $149!
Because some people do not love Shannon.

A sufficient sample-rate is therefore 2B samples/second, or anything larger. Equivalently, for a given sample rate fs, perfect reconstruction is guaranteed possible for a bandlimit B < fs/2.

So 768/2 = 384 kHz. Where is the mutant who hear that ?

Money, money....mastering the basics

Last edited by dinococcus; 13th October 2017 at 06:50 AM..
Old 13th October 2017
  #20
DSD256 sounds like a mirror image of a human playing music.
Old 15th October 2017
  #21
Lives for gear
 
Adebar's Avatar
Quote:
Originally Posted by dinococcus View Post
Because some people do not love Shannon.

..... perfect reconstruction is guaranteed possible for a bandlimit B < fs/2.
If you look closer at Shannon this sounds a little simplified as you can read here.

http://www.wescottdesign.com/article...g/sampling.pdf
Old 16th October 2017
  #22
Lives for gear
 

Quote:
Originally Posted by Adebar View Post
If you look closer at Shannon this sounds a little simplified as you can read here.

http://www.wescottdesign.com/article...g/sampling.pdf
Shannon is the basis for a numeric signal (a sample signal).

Have you academic document instead of commercial literature?
Old 22nd October 2017
  #23
Lives for gear
 
Adebar's Avatar
Quote:
Originally Posted by dinococcus View Post
Shannon is the basis for a numeric signal (a sample signal).

Have you academic document instead of commercial literature?
Did you read the paper of Wescott? There is no academic conflict with what Shannon says.

The first sentence in this paper states what everybody knows.

"The assertion made by the Nyquist-Shannon sampling theorem is simple: if you have a signal that is perfectly band limited to a bandwidth of f0 then you can collect all the information there is in that signal by sampling it at discrete times, as long as your sample rate is greater than 2f0."


The next statement often is neglected.

"The difficulty with the Nyquist-Shannon sampling theorem is that it is based on the notion that the signal to be sampled must be perfectly band limited. This property of the theorem is unfortunate because no real world signal is truly and perfectly band limited. In fact, if a signal were to be perfectly band limited—if it were to have absolutely no energy outside of some finite frequency band—then it must extend infinitely in time."

Another 2 statements in this paper are:

"Nyquist didn’t say that if you are sampling at rate N that you could use an anti-aliasing filter with a cutoff at frequency f = N/2."

"If you have the freedom to set the sampling rate, this filtering effect means that you’ll need to make a trade-off between a low sampling rate and the complexity of your anti-alias filtering. It is not uncommon for systems to have sampling rates that are seemingly quite high, just to simplify the task of designing the anti-alias filters."

I cannot see an academic contradiction in this paper to that what Nyquist and Shannon says.
Old 23rd October 2017
  #24
Lives for gear
 

Quote:
Originally Posted by Adebar View Post
Did you read the paper of Wescott? There is no academic conflict with what Shannon says.

The first sentence in this paper states what everybody knows.

"The assertion made by the Nyquist-Shannon sampling theorem is simple: if you have a signal that is perfectly band limited to a bandwidth of f0 then you can collect all the information there is in that signal by sampling it at discrete times, as long as your sample rate is greater than 2f0."


The next statement often is neglected.

"The difficulty with the Nyquist-Shannon sampling theorem is that it is based on the notion that the signal to be sampled must be perfectly band limited. This property of the theorem is unfortunate because no real world signal is truly and perfectly band limited. In fact, if a signal were to be perfectly band limited—if it were to have absolutely no energy outside of some finite frequency band—then it must extend infinitely in time."

Another 2 statements in this paper are:

"Nyquist didn’t say that if you are sampling at rate N that you could use an anti-aliasing filter with a cutoff at frequency f = N/2."

"If you have the freedom to set the sampling rate, this filtering effect means that you’ll need to make a trade-off between a low sampling rate and the complexity of your anti-alias filtering. It is not uncommon for systems to have sampling rates that are seemingly quite high, just to simplify the task of designing the anti-alias filters."

I cannot see an academic contradiction in this paper to that what Nyquist and Shannon says.
Hello,

I know the problem of the numeric is the passage of the mathematic object (they are perfect) to the electronic reality.

The aliasing problem in the 70's/mid 80's yes. in 2017, i don't see.
Old 23rd October 2017
  #25
Lives for gear
 
Adebar's Avatar
Quote:
Originally Posted by dinococcus View Post

The aliasing problem in the 70's/mid 80's yes. in 2017, i don't see.
Hello,

That´s the question. You can do a lot with digital filters. But also these are not perfect when you have to make them very steep.
Old 23rd October 2017
  #26
Gear Nut
 

All the processing is done by downgrading the samplerate, so what´s the point?
Old 5th November 2017
  #27
Lives for gear
 
Plush's Avatar
I like the Japanese set and and will acquire it.
Old 6th November 2017
  #28
Lives for gear
 
Jerry Tubb's Avatar
 

Verified Member
I remember Mr. Neve saying he prefered 384kHz PCM about 20 years ago.

Seems like overkill to me, but who am I to debate Mr. Neve (?)

one time i did master a song at 192k and liked the sound of it.

but 24-bit 96k does the job beautifully to my ears.

best, jt

Last edited by Jerry Tubb; 6th November 2017 at 11:15 AM..
Old 6th November 2017
  #29
Gear Head
 

Many argue 44.1KHz is enough with proper oversampling on both ends. Doesn't quite match what I've heard, with my ears agreeing with the above statement that there's the most benefit at 96K with a slight improvement at 192. But there are so many factors involved, quality of the DAC, ADC, clocking, DSP, oversampling, mastering, etc. But the math says one can get accurate reconstruction of a 20KHz bandwidth limited signal at 44.1KHz. But whether frequencies above 20K add to the audio experience or not is often argued.
Old 6th November 2017
  #30
Motown legend
 
Bob Olhsson's Avatar
 

Verified Member
This is really all about what the filters, which vary in construction from both device to device and sample rate to sample rate, introduce below 20kHz. Adding to the excitement is that different people are more and less sensitive to the artifacts. This subject makes for endless arguments whenever it is oversimplified. I think I've literally heard every sample rate beat every other sample rate when using some specific gear. In the end, I need to use whatever sounds best using my own gear.
Mentioned Products
New Reply Submit Thread to Facebook Facebook  Submit Thread to Twitter Twitter  Submit Thread to LinkedIn LinkedIn  Submit Thread to Google+ Google+  Submit Thread to Reddit Reddit 
 
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump