avatar

burns46824

Gold Finger

Posts: 262 Member Since: 07/02/2011

Lead

Jan 19 17 7:10 PM

Tags : :

Question for you mastering engineers...

If you're mastering from analog tape, do you think it sounds better to capture mixes at high sample rates (24/96 or 24/192) and downsample to 16/44.1 compared to capturing at 24/44.1 or even 16/44.1 (to avoid SRC) if CD is the intended release format?
Quote    Reply   
avatar

chrisj

Gold Finger

Posts: 938 Member Since:22/02/2011

#1 [url]

Jan 19 17 7:14 PM

I'm interested to hear about this, too. I could make arguments either way, but that's not the same as checking with trusted ears (and people who have real mastering-grade monitoring, mine is only impressive compared to prosumer crud).

Thoughts? I might learn something :)

Chris Johnson, airwindows.com

Quote    Reply   
avatar

burns46824

Gold Finger

Posts: 262 Member Since:07/02/2011

#2 [url]

Jan 19 17 8:46 PM

I feel like the best remasters came out in the late 90s/early 2000s and I feel like mastering trends with sample rates have probably changed since then. Limiting wasn't egregious until the early 2000s I feel, but I suppose it had started getting bad earlier. If only there were CD masters with 80s dynamics and modern converters!

Quote    Reply   
avatar

gold

Platinum Blonde

Posts: 1,468 Member Since:27/01/2011

#3 [url]

Jan 19 17 9:43 PM

I used to capture everything at the target sample rate because I didn't like any SRC I had. The Prism AD2 has excellent realtime SRC but that's impractical to use. After I got Saracon I had no qualms about switching to capturing everyting at 24/96. That's what I do now.

Quote    Reply   
avatar

burns46824

Gold Finger

Posts: 262 Member Since:07/02/2011

#4 [url]

Jan 19 17 10:24 PM

gold wrote:
I used to capture everything at the target sample rate because I didn't like any SRC I had. The Prism AD2 has excellent realtime SRC but that's impractical to use. After I got Saracon I had no qualms about switching to capturing everyting at 24/96. That's what I do now.

Yeah, I've got the AD-2 at my studio, as well.  You think capturing at 24/96 and downsampling to 16/44.1 in Saracon sounds better than capturing at 16/44.1 on the Prism?

Quote    Reply   
avatar

gold

Platinum Blonde

Posts: 1,468 Member Since:27/01/2011

#5 [url]

Jan 19 17 11:24 PM

burns46824 wrote:

 

When I first got Saracon I did a few experiments and couldn't tell the difference between the two methods on all but a couple of things. The difference was very small and didn't really bother me. So I never looked back. I  often have to deliver at multiple sample and bit rates. Recapturing everything is not practical.

The AD2 will output two sample rates. If you have two capture devices you could record both at once.
 

Last Edited By: gold Jan 19 17 11:28 PM. Edited 1 time.

Quote    Reply   
avatar

burns46824

Gold Finger

Posts: 262 Member Since:07/02/2011

#6 [url]

Jan 20 17 12:45 AM

gold wrote:

burns46824 wrote:


 

When I first got Saracon I did a few experiments and couldn't tell the difference between the two methods on all but a couple of things. The difference was very small and didn't really bother me. So I never looked back. I  often have to deliver at multiple sample and bit rates. Recapturing everything is not practical.

The AD2 will output two sample rates. If you have two capture devices you could record both at once.
 

Yeah I do have two SD recorders at my disposal (Tascam DA-3000).  I only see one "wordclock" output on the AD-2, which is what I've been using to clock the one DA-3000.  But there are two other BNC outputs...you know what the difference is?

Quote    Reply   
avatar

reynaud

Silverado

Posts: 194 Member Since:13/07/2011

#7 [url]

Jan 20 17 1:23 AM

If the source is analogue and the processing chain is purely analogue, capture will be at the target rate (i.e. 44.1k for CD release) and a higher rate (i.e. 96k) for asset repurposing (e.g. Hi-res download or Pure Audio Blu-ray, or even for vinyl).  

This is the same process employed with straight transfers and DSD. 

If the source is analogue and the processing chain is mostly analogue but with a digital limiter at the capture end, and further editing is required, capture is always at 96k.  

If the source is analogue but the processing chain is purely digital, capture will always be at 96k. All targets are generated from the 96k master (i.e. 48k for video and 44.1k for CD). 

The quality of the SRC available today is much improved over what was available only a few years ago and I have no issue at all converting between "disparate" rates (e.g. 96k and 44.1k). I wouldn't hesitate using Saracon or iZotope or the newer Sonic SRCs.

Last Edited By: reynaud Jan 20 17 1:51 AM. Edited 2 times.

Quote    Reply   
avatar

tbethel

Gold Finger

Posts: 296 Member Since:21/02/2011

#8 [url]

Jan 20 17 5:04 AM

I try and get the best possible transfer at 44.1 and 24 bits. The initial transfer is what is important IMHO. I always spool the tape at spooling speed and check the splices as I go. Then the tape deck cleaned, demag'd, and checked for proper operation. Tones off tape calibrated, Then a careful playback with me listening to the whole transfer in real time. The same process for cassette or vinyl. It is the initial transfer that is the most important. Yes we have RX-4 but I would rather get a clean transfer so I don't have to rely on post processing as much. FWIW

-TOM-
Thomas W. Bethel
Managing Director, Acoustik Musik, Ltd.
www.acoustikmusik.com/
Celebrating 22 years in business in 2017

Quote    Reply   
avatar

jerry tubb

Gold Finger

Posts: 621 Member Since:06/02/2011

#9 [url]

Jan 20 17 7:27 AM

I've also done all of these approaches.
Capturing at 24/44.1 for CD is a great option, at 24-bit you can still make transparent level and fade adjustments.
Bypasses the need for SRC from a 96k capture.
Capturing at 24/96 for HiRes distribution, or just general HD Archiving.
Then down sampling for CD or other internet distribution.
Which one you choose might depend on your hardware and software setup, and/or if you want to run the analog tapes twice.
JT

Quote    Reply   
avatar

gr

Gold Finger

Posts: 350 Member Since:21/02/2011

#11 [url]

Jan 20 17 9:54 AM

To the original question: it depends on the converters you have. Some sound better at 96k and others better at 44k1 but there are other variables that make it difficult to compare directly.

Assuming good converters, capturing at 96k gives you more options for other release formats.

Quote    Reply   
avatar

Justin P

Silverado

Posts: 69 Member Since:03/07/2014

#12 [url]

Jan 20 17 10:58 AM

I'm with those that process/play/capture at 96k. It keeps your options open for other release formats and both my HEDD and HILO perform very well at 96k. If you have a great SRC like Saracon (or RX5 or Myriad w/Goodhertz SRC) you will have no problems downsampling to 44.1k when needed. I personally think that doing digital processing/play/capture at 96k yields the best results for the total summation of the project. With a good SRC and best practices, the SRC down to 44.1k is no problem if you are careful about dither and headroom.

Quote    Reply   
avatar

burns46824

Gold Finger

Posts: 262 Member Since:07/02/2011

#14 [url]

Jan 20 17 12:18 PM

Interesting to read your opinions, not being a mastering engineer myself!

So, if you were to capture at 44.1 kHz, the advantage of capturing at 24 bit is so that you can "sequence" a CD in a mastering DAW without losing quality? What about the change from 24 bit to 16 bit at the end of the day? Isn't that also a "conversion"?

Quote    Reply   
avatar

weedywet

Ruby Baby

Posts: 5,774 Member Since:20/01/2011

#17 [url]

Jan 20 17 4:39 PM

people (especially on some websites) love to quote the part of that Lavry article that says " 60 KHz would be closer to the ideal" sampling rate.

but they choose to ignore the rest:

At 60 KHz sampling rate, the contribution of AD and DA to any attenuation in the audible range is negligible. Although 60 KHz would be closer to the ideal; given the existing standards, 88.2 KHz and 96 KHz are closest to the optimal sample rate. At 96 KHz sampling rate the theoretical bandwidth is 48 KHz. In designing a real world converter operating at 96 KHz, one ends up with a bandwidth of approximately 40 KHz.

Quote    Reply   
avatar

burns46824

Gold Finger

Posts: 262 Member Since:07/02/2011

#18 [url]

Jan 20 17 5:30 PM

Will be curious to hear, after doing some SRC tests on my own, if capturing at 24/96 actually sounds better than capturing at 24/44.1 if CD is the intended release format.  Any of you done this?

Last Edited By: burns46824 Jan 20 17 5:49 PM. Edited 1 time.

Quote    Reply   
avatar

biigniick

Silverado

Posts: 157 Member Since:16/02/2011

#19 [url]

Jan 20 17 8:13 PM

burns46824 wrote:
Will be curious to hear, after doing some SRC tests on my own, if capturing at 24/96 actually sounds better than capturing at 24/44.1 if CD is the intended release format.  Any of you done this?

once I was able to run my chain at 96kHz and I found SRCs that I like, I capture at 24/96 and SRC for CD release. I have several SRCs that I like and it gives me options for Hi-Res file export for other formats if needed.

- nick

Quote    Reply   
avatar

adam

Silverado

Posts: 92 Member Since:21/02/2011

#20 [url]

Jan 21 17 9:13 PM

gr wrote:
To the original question: it depends on the converters you have. Some sound better at 96k and others better at 44k1 but there are other variables that make it difficult to compare directly.

Assuming good converters, capturing at 96k gives you more options for other release formats.

Agreed.

And I can't think of any reason to capture at 16 bit, unless you were somehow printing direct to LBR for glass mastering...

FWIW, most of what I receive has been coming in at (and mixed at) 48k – and it's natively suitable for every format except CD.

Quote    Reply   
Add Reply

Quick Reply

bbcode help