R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: [1]   Go Down

Author Topic: Question about SR/ Bit depth conversion in the Masterlink  (Read 2411 times)

zulusound

  • Newbie
  • *
  • Offline Offline
  • Posts: 2
Question about SR/ Bit depth conversion in the Masterlink
« on: April 22, 2005, 02:09:34 pm »

Greetings,
If I am (in the end) delivering a 16 bit/ 44.1 replication master on CD from an Alesis Masterlink..  

Should I
A: mix to the ML at 24/96, then burn a redbook 16/44.1
or
B: initially mix to 16/44.1 so the ML doesn't have to crunch any numbers?  

Thanks.
Logged

ted nightshade

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1272
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #1 on: April 22, 2005, 03:59:46 pm »

zulusound wrote on Fri, 22 April 2005 11:09

Greetings,
If I am (in the end) delivering a 16 bit/ 44.1 replication master on CD from an Alesis Masterlink..  

Should I
A: mix to the ML at 24/96, then burn a redbook 16/44.1
or
B: initially mix to 16/44.1 so the ML doesn't have to crunch any numbers?  

Thanks.



If you have to use the SRC and dither on the Masterlink, I would mix straight to 44.1.

If the CD you are burning is really a replication master, and will definitely receive no further mastering, do it at 16 bit.

If you want to be able to do further mastering of any kind, even a gain change, do it at 24 bit.

The dither on the Masterlink is not too bad, but not really that good either. The SRC is pretty rough though. I would especially avoid the SRC, and the rest of the DSP as well, including gain changes.
Logged
Ted Nightshade aka Cowan

There's a sex industry too.
Or maybe you prefer home cookin'?

bblackwood

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 7036
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #2 on: April 22, 2005, 07:48:02 pm »

Avoid the Masterlink's SRC at all costs. It sucks beyond belief.
Logged
Brad Blackwood
euphonic masters

bobkatz

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2926
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #3 on: April 22, 2005, 07:52:09 pm »

zulusound wrote on Fri, 22 April 2005 14:09

Greetings,
If I am (in the end) delivering a 16 bit/ 44.1 replication master on CD from an Alesis Masterlink..  

Should I
A: mix to the ML at 24/96, then burn a redbook 16/44.1
or
B: initially mix to 16/44.1 so the ML doesn't have to crunch any numbers?  

Thanks.




The Masterlink has no dither, just some noiseshaping and some pseudo-mastering algorithms that are not particularly good. So, soundwise, you're better off preparing everything externally at 4416 properly and then feeding the ML. But why are you using the Masterlink at all to cut a replication master? If there's a computer involved in the first place, then you probably have better tools available on your computer to cut a replication-ready master.

BK
Logged
There are two kinds of fools,
One says-this is old and therefore good.
The other says-this is new and therefore better."

No trees were killed in the sending of this message. However a large number of
electrons were terribly inconvenienced.

Ed Littman

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 877
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #4 on: April 22, 2005, 07:54:32 pm »

ted nightshade wrote on Fri, 22 April 2005 15:59


The dither on the Masterlink is not too bad, but not really that good either.


before mine smoked up a few years back it only used noise shaping & had no dither. Has dither been added since a few years ago?
Ed
Logged

Ronny

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2739
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #5 on: April 22, 2005, 11:10:42 pm »

bobkatz wrote on Fri, 22 April 2005 19:52

zulusound wrote on Fri, 22 April 2005 14:09

Greetings,
If I am (in the end) delivering a 16 bit/ 44.1 replication master on CD from an Alesis Masterlink..  

Should I
A: mix to the ML at 24/96, then burn a redbook 16/44.1
or
B: initially mix to 16/44.1 so the ML doesn't have to crunch any numbers?  

Thanks.




The Masterlink has no dither, just some noiseshaping and some pseudo-mastering algorithms that are not particularly good. So, soundwise, you're better off preparing everything externally at 4416 properly and then feeding the ML. But why are you using the Masterlink at all to cut a replication master? If there's a computer involved in the first place, then you probably have better tools available on your computer to cut a replication-ready master.

BK




Bob, if it doesn't have dither and it has noise shaping, than what exactly is it that you are noise shaping. Are you positive that it doesn't dither the 24 to 16 bit reduction? This just doesn't compute for some reason with me. Maybe it dithers, but only has one type of dither, no user choices on dither type, but a choice of noise shaping the stock dither.
Logged
------Ronny Morris - Digitak Mastering------
---------http://digitakmastering.com---------
----------Powered By Experience-------------
-------------Driven To Perfection---------------

bobkatz

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2926
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #6 on: April 23, 2005, 05:55:59 am »

Ronny wrote on Fri, 22 April 2005 23:10



Bob, if it doesn't have dither and it has noise shaping, than what exactly is it that you are noise shaping. Are you positive that it doesn't dither the 24 to 16 bit reduction?




The DSP in the Masterlink is extremely primitive. They don't have enough MIPs to do anything very well. So they may have cut corners. There is absolutely nothing in the Masterlink manual about dither, last I looked.

Ask George M about his attempts to design an equalizer for the Masterlink until he discovered the MIPs budget they could give him.

It is possible to noise shape a signal without dithering it. It just doesn't sound very good because you get noise modulation. I invented a simple noiseshaping algorithm that digidesign used for a while until they added dither, way back in the Sound Designer II days. Maybe it gave us 1/2 a bit to 1 bit additional resolution.

BK
Logged
There are two kinds of fools,
One says-this is old and therefore good.
The other says-this is new and therefore better."

No trees were killed in the sending of this message. However a large number of
electrons were terribly inconvenienced.

ted nightshade

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1272
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #7 on: April 24, 2005, 08:58:40 am »

All I know is, the damage is real but not too appalling when going from 24 to 16 on the Mlink. I wouldn't care to do that on a replication master.
Logged
Ted Nightshade aka Cowan

There's a sex industry too.
Or maybe you prefer home cookin'?

Ronny

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2739
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #8 on: April 24, 2005, 10:30:49 pm »

Quote:

 title=bobkatz wrote on Sat, 23 April 2005 05:55The DSP in the Masterlink is extremely primitive. They don't have enough MIPs to do anything very well. So they may have cut corners. There is absolutely nothing in the Masterlink manual about dither, last I looked.


That doesn't mean that dither isn't applied on bit reduction without giving user options. It's more likely that they dither the reduction than not. The ML may not be the most ingenious device created, but Alesis aren't new kids on the block WRT, digital technology, besides eliminating dither as a cost cutter would probably be one of the last things that they'd do. I wouldn't assume that just because there are no user dither options that they are truncating the 24 to 16 bit reduction without it.


Logged
------Ronny Morris - Digitak Mastering------
---------http://digitakmastering.com---------
----------Powered By Experience-------------
-------------Driven To Perfection---------------

Bob Olhsson

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3968
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #9 on: April 24, 2005, 11:16:23 pm »

I'm afraid the word straight from the horse's mouth is that they noiseshape and truncate due to extreme DSP limitations.

Ronny

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2739
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #10 on: April 25, 2005, 01:18:05 am »

Bob Olhsson wrote on Sun, 24 April 2005 23:16

I'm afraid the word straight from the horse's mouth is that they noiseshape and truncate due to extreme DSP limitations.



What noise are they shaping, is what I want to know?
Logged
------Ronny Morris - Digitak Mastering------
---------http://digitakmastering.com---------
----------Powered By Experience-------------
-------------Driven To Perfection---------------

Ronny

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2739
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #11 on: April 25, 2005, 02:55:40 am »

Bob Olhsson wrote on Sun, 24 April 2005 23:16

I'm afraid the word straight from the horse's mouth is that they noiseshape and truncate due to extreme DSP limitations.



1. It makes no sense to have a noise shaper unless you have noise to shape.
2. The most common application for adding noise to the digital signal is dither.
3. Alesis is too smart not to dither the CD24 file to redbook.
4. I think that they are following advice that I believe I've heard you mention and that I agree with and that is to dither, regardless of whether you give the user option of type of dither.  


Logged
------Ronny Morris - Digitak Mastering------
---------http://digitakmastering.com---------
----------Powered By Experience-------------
-------------Driven To Perfection---------------

Bob Olhsson

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3968
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #12 on: April 25, 2005, 08:53:55 am »

They are shaping the "truncation noise" which is digital newspeak for distortion. (Digital newspeak would make a good topic!) Bear in mind that simply adding noise is only one method of dithering. The main thing is de-correlating the bottom bit from the program material. The idea of noise shaping is to move the distortion away from the 3k area where our ears are most sensitive using, if I remember/understand correctly, a feedback and filter technique.

Ronny

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2739
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #13 on: April 25, 2005, 01:03:51 pm »

Bob Olhsson wrote on Mon, 25 April 2005 08:53

They are shaping the "truncation noise" which is digital newspeak for distortion. (Digital newspeak would make a good topic!) Bear in mind that simply adding noise is only one method of dithering. The main thing is de-correlating the bottom bit from the program material. The idea of noise shaping is to move the distortion away from the 3k area where our ears are most sensitive using, if I remember/understand correctly, a feedback and filter technique.


I know what noise shaping is all about Bob. If they could move quantization errors to the inaudible part of the spectrum, than there wouldn't be any need for dither, it would simply be moved out of the human hearing range. While that may be possible in the future, at this present time I don't believe that it's possible.

I talked to the cowboy instead of the horse, this is what Alesis  says in their FAQ's.

Does the ML9600 do sample rate and word length conversion?

Yes. If you burn a Red Book CD from a playlist containing tracks burned at sample rates other than 44.1 kHz or word lengths higher than 16-bit, it will automatically convert them using a proprietary high-quality dithered process that results in tracks that sound almost identical to the high-resolution originals. Back

Can I record at high resolution and still make standard CDs to send out?

Yes, you can make 24-bit/96 kHz mixes that are DVD ready, and still create standard Red Book CDs from that playlist. MasterLink will use its state-of-the-art internal sampling rate converter and dithering algorithm to make a standard audio CD that sounds almost as good as the high-resolution originals, even though it's 16-bit/44.1 kHz. Back


http://www.alesis.com/support/faqs/masterlink.html

Go down about 3/4's of the page.
Logged
------Ronny Morris - Digitak Mastering------
---------http://digitakmastering.com---------
----------Powered By Experience-------------
-------------Driven To Perfection---------------

bobkatz

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2926
Re: Question about SR/ Bit depth conversion in the Masterlink
« Reply #14 on: May 07, 2005, 08:33:11 pm »

Ronny wrote on Mon, 25 April 2005 02:55

Bob Olhsson wrote on Sun, 24 April 2005 23:16

I'm afraid the word straight from the horse's mouth is that they noiseshape and truncate due to extreme DSP limitations.



1. It makes no sense to have a noise shaper unless you have noise to shape.
2. The most common application for adding noise to the digital signal is dither.
3. Alesis is too smart not to dither the CD24 file to redbook.




Unfortunately, Alesis did not provide enough power in the unit to dither, so all they do is noise shape. It take far fewer DSP cycles to do a primitive noise shaping than it does to dither. The advantage of noise shaping is about 1/2 to 1 bit instead of far more with dither. Thus, the grungy, noise modulated sound.

But as another poster recently pointed out, it's subtle, but I wouldn't want to make a replication master with it.

Quote:



4. I think that they are following advice that I believe I've heard you mention and that I agree with and that is to dither, regardless of whether you give the user option of type of dither.  




In this case, Alesis is simply doing almost nothing. As I said, there is nothing mentioned about dither in the manual, but there is something mentioned about noise shaping, which is not the same thing.

Ronny, you can have noise shaping without dither. You can noise shape dither or you can noise shape without dither. Noise shaping without dither produces horrid noise modulation and distortion. The Waves type 2 is primarily noise shaping without dither!  One of the Waves plugs lets you (or used to let you) choose noise shaping on or off and/or dither on or off. You can see or hear the effect of each option.

Noise shaping is simply feedback from an output point back to an input point, or from multiple output points back to multiple input points with varying amounts of delay. This produces a high frequency boost in the feedback path and it is commonly known as noise shaping. If you insert dither noise in that path, then it is known as noise shaped dither.

BK
Logged
There are two kinds of fools,
One says-this is old and therefore good.
The other says-this is new and therefore better."

No trees were killed in the sending of this message. However a large number of
electrons were terribly inconvenienced.
Pages: [1]   Go Up