Avid Pro Audio Community

Avid Pro Audio Community (https://duc.avid.com/index.php)
-   003, Mbox 2, Digi 002, original Mbox, Digi 001 (Win) (https://duc.avid.com/forumdisplay.php?f=17)
-   -   How Is 48kHz Audio That's Imported Into 96kHz Session Affected? (https://duc.avid.com/showthread.php?t=250013)

taylor4814 07-14-2009 05:23 AM

How Is 48kHz Audio That's Imported Into 96kHz Session Affected?
 
If I have rhythm tracks recorded at 48kHz, 24-bit, and later have the opportunity to import them into a 96kHz, 24-bit session, where, say, vocals and acoustic guitars will be recorded, would there be any factors against doing that -- purely from an audio quality point of view...? Thanks in advance for any insight,

daeron80 07-14-2009 06:24 AM

Re: How Is 48kHz Audio That's Imported Into 96kHz Session Affected?
 
Just make sure Apply SRC is checked in the Import dialog. Select a Quality level of "Tweakhead (Slowest)" and the interpolation will be as good as PT can make it. Which is great.

TOM@METRO 07-14-2009 03:28 PM

Re: How Is 48kHz Audio That's Imported Into 96kHz Session Affected?
 

As daeron80 said, use “tweakhead”. The math used in the conversion from 48 to 96 is a nice even computation and should not produce any undesirable artifacts. You will have a theoretical advantage completing your project at 96k, although many of us don’t really hear a difference worth the resource hit. On the other hand many swear by it. Good luck.

daeron80 07-15-2009 05:55 PM

Re: How Is 48kHz Audio That's Imported Into 96kHz Session Affected?
 
Quote:

Originally Posted by TOM@METRO (Post 1425697)
The math used in the conversion from 48 to 96 is a nice even computation

You'd think so, but actually, it has to guess the trajectory of the waveform regardless of the simplicity of the ratio. Imagine a sine wave curving between two points of 48 - the algorithm has to take into account what comes before and after to place the new bit near where it probably ought to be.

Quote:

Originally Posted by TOM@METRO (Post 1425697)
many of us don’t really hear a difference worth the resource hit. On the other hand many swear by it.

I prefer the sound of tracking at 96. Mixing, I don't notice any difference unless I apply a plug-in with Audio Suite, and even then it has to be one that really changes the sound a lot.

TOM@METRO 07-15-2009 08:18 PM

Re: How Is 48kHz Audio That's Imported Into 96kHz Session Affected?
 
Quote:

Originally Posted by daeron80 (Post 1426262)
You'd think so, but actually, it has to guess the trajectory of the waveform regardless of the simplicity of the ratio. Imagine a sine wave curving between two points of 48 - the algorithm has to take into account what comes before and after to place the new bit near where it probably ought to be.

.

I have no argument with you about the trajectory speculation involved with the up sampling we’re talking about, and that compared to tracking at 96k there will be theoretical inaccuracies. Just sayin’ that it’s likely that undesirable artifacts will be below the bar of tolerance in a pop music setting.

peppertree 07-15-2009 08:28 PM

Re: How Is 48kHz Audio That's Imported Into 96kHz Session Affected?
 
Yes any SRC algo will require filtering from any speed to any other. Quality of SRC does vary, and for anything important that I must digitally resample I use the izotope resampler in batch via audiofile sample manager. This site plots the quality:

http://src.infinitewave.ca

(Should be noted that izotope assisted the creation of that website, but I think it's fair.)

Mostly for 2 track I use the Tascam DV-RA1000HD as my mastering deck which I can run at any sample rate vs. the DAW's rate, and just work analog.

As for the effects of sample rate, the better your converters are the less it will matter at the conversion stage. What does matter however is the sample rate when doing nonlinear processing in the digital domain. Virtually no nonlinear plugins have antialiasing filters; nearly all of them will alias harmonics back into the audible band at ugly nonharmonic intervals. The higher your sample rate, the lower in level these aliases are likely to be, and eventually, they are down within the noise floor and irrelevant. 48KHz is much better than 44.1 in this regard, and 96KHz is about all you would ever need with today's mic preamp dynamic ranges.

An example of this problem affecting product design is given in the Lavry AD10 which will only do its tube/tape sims at double speeds because Mr. Lavry doesn't like inharmonic aliasing distortion and I don't blame him.

Note however that your available DSP power goes down _more than half_ if you double the samplerate. This is because DSP gets more costly as you handle frequencies further away from half nyquist. If the programmer doesn't take that into consideration, the quality of the DSP may actually suffer as you go higher in rate. The scaling of the Liquid Mix demonstrates this tradeoff. For all these reasons quad speeds are unlikely to be justifiable in practice at least in the near future.

TOM@METRO 07-15-2009 09:16 PM

Re: How Is 48kHz Audio That's Imported Into 96kHz Session Affected?
 
Some of the comparisons on that site are nothing short of shocking.

peppertree 07-15-2009 09:57 PM

Re: How Is 48kHz Audio That's Imported Into 96kHz Session Affected?
 
Quote:

Originally Posted by TOM@METRO (Post 1426325)
Some of the comparisons on that site are nothing short of shocking.

Yup. r8brain free would be a nice pickup for windows users, but the pro version is much better. Weiss Saracon may overall be the state of the art today, even if some tradeoffs (compare impulse ringing with filter sharpness) have to be made.

TOM@METRO 07-16-2009 01:28 AM

Re: How Is 48kHz Audio That's Imported Into 96kHz Session Affected?
 
I'm still doing some digesting here. Some of what I see confirms what up until now I would only question and then reject based on the apparent status of the coders involved. r8brain free looks to be a far cry from Sound Forge in areas. It would be interesting to know more about the test skew potential, though it would seem tough to fake it that dramatically.

daeron80 07-16-2009 11:53 AM

Re: How Is 48kHz Audio That's Imported Into 96kHz Session Affected?
 
Holy cow. I've tried using PEAK 6 for conversion instead of PT, but it takes SO LONGGGGGGGGG. Crazy long. It's better but it's not worth the wait most of the time.


All times are GMT -7. The time now is 04:51 PM.

Powered by: vBulletin, Copyright ©2000 - 2008, Jelsoft Enterprises Limited. Forum Hosted By: URLJet.com