24/48 -> 16/44.1 -the d/a/a/d way

[Felix]

i would like to do this to my 24/48 mixes:
(if and when true mastering isn't an option)

-go out 24/48 d/a
-go through Fatso and Rosetta back in as 16/44.1

i can't do this with PT.
i don't think i can run 2 programs (like PT and Peak) to perform this operation successfully.
(and i'm not going to buy a separate mix-down machine at this point. way too many other things i need to buy. i just sold my obsolete 16 bit DAT for pennies on the dollar.)
so what i'm thinking is getting a nice sound card for my PC that does good 24/48 d/a and using PT Free on the PC to play(via PC sound card) 24/48 2 track files through the chain into PT on the Mac. it's basically intended to avoid dithering and esp. src crap. plus, i'd get the ssl 2 bus to boot.

: PT Free/PC -> 24/48 D/A sound card -> Fatso -> Rosetta/set for 16/44.1 -> PT24/Mac

anyone feeling the magic here?
suggestions? what sound card to get, is this a bad idea?

[Felix]

bump

[bluemt]

Why not just go:

PT24/Mac -> Fatso -> Rosetta/set for 16/44.1 -> PT Free/PC

and then copy the files back to your mac via network or removable media for further processing or to burn CD.

If you do your own mastering, I would do all that before you do the above (keeping in mind how the Fatso will change the sound). That way you're processing in 24/48 and you're just doing bit depth/sample reduction in your transfer to PC.

[univox1]

try going back to p tools 24/48 into a stereo bus return. then bounce from that source and convert the bounce to 16/44.

[Dog BBQ]

Originally posted by Felix:
i just sold my obsolete 16 bit DAT for pennies on the dollar.)

<font size="2" face="Verdana, Arial">Guess it wasn't so obsolete

[img]images/icons/shocked.gif[/img]

[bluemt]

"try going back to p tools 24/48 into a stereo bus return. then bounce from that source and convert the bounce to 16/44. "

I would definately choose ad/da over a sample conversion algorythm. They are not good for your audio and will add undesireable artifacts. Stick to the coversion method to change sample rate.

[Felix]

Originally posted by bluemt:
Why not just go:

PT24/Mac -> Fatso -> Rosetta/set for 16/44.1 -> PT Free/PC

and then copy the files back to your mac via network or removable media for further processing or to burn CD.

If you do your own mastering, I would do all that before you do the above (keeping in mind how the Fatso will change the sound). That way you're processing in 24/48 and you're just doing bit depth/sample reduction in your transfer to PC.

<font size="2" face="Verdana, Arial">yes, thanks! that makes more sense. i didn't even consider that i could run the Rosetta into the PC. so then, i would need a card for the PC that accepts digital (spdff or aes) in. however, i would be using the 882/20 for D/A which is somewhat compromising. then again, this still should at least sound better than src, right?

[love666]

you would need something like the emagic 2/6 - a USB interface with pt free. I think it has s/pdif in. but by the time you've spent that, you might consider a toolbox or an mbox instead.

[bluemt]

I would think that D/A through the 882/20 would be better than src. I would definately choose this method. The next obvious question you're probably conteplating is whether better D/A (24 bit)converters will make a difference. The short answer is yes, of course better sounding converters would be a better option. But, depending on the source material, it might not matter that much and there's something to say for doing the best you can with the equipment currently at your disposal.

If the source material has limited dynamic range to begin with or was derived from a lot of 16-bit or 20-bit samples and/or audio, than I think the 882/20 D/A would be fine. If the source has a lot of dynamic range and was recorded and processed at 24 bits than I think you're compromising a little more.

There are a couple safe guards you could do to try and optimize a conversion in the second scenario. The first would be to make sure that all of your levels are pretty hot and your master is close to full scale (with a few dbs of headroom at it's hottest point). The second is to try and use a bit depth conversion (not src) at the final stage of your master bus. I'm pretty sure the L1 Maximizer or another tool could do a 24 bit to 20 bit bit reduction using dither/noise shapping of some kind. This would optimize the files beore they hit the D/A converters of your 882/20 and your sample rate conversion via Rosetta. It would be better than allowing the converter to truncate the least signicant bits on the way out of the 882/20.

In either of the scenarios above you should get good results.

[Felix]

Originally posted by bluemt:
I would think that D/A through the 882/20 would be better than src. I would definately choose this method. The next obvious question you're probably conteplating is whether better D/A (24 bit)converters will make a difference. The short answer is yes, of course better sounding converters would be a better option. But, depending on the source material, it might not matter that much and there's something to say for doing the best you can with the equipment currently at your disposal.

If the source material has limited dynamic range to begin with or was derived from a lot of 16-bit or 20-bit samples and/or audio, than I think the 882/20 D/A would be fine. If the source has a lot of dynamic range and was recorded and processed at 24 bits than I think you're compromising a little more.

There are a couple safe guards you could do to try and optimize a conversion in the second scenario. The first would be to make sure that all of your levels are pretty hot and your master is close to full scale (with a few dbs of headroom at it's hottest point). The second is to try and use a bit depth conversion (not src) at the final stage of your master bus. I'm pretty sure the L1 Maximizer or another tool could do a 24 bit to 20 bit bit reduction using dither/noise shapping of some kind. This would optimize the files beore they hit the D/A converters of your 882/20 and your sample rate conversion via Rosetta. It would be better than allowing the converter to truncate the least signicant bits on the way out of the 882/20.

In either of the scenarios above you should get good results.

<font size="2" face="Verdana, Arial">great! you're awesome.
half of what i do is tracked 24/48. then many 16/44.1 are used also. it is also processed 24/48. it is not very dynamic overall. so going out 882 via L1 20bit dither to Rosetta/PC should work pretty good. this is in lieu(sp) of true mastering at a real mastering house ultimately, so your advice of making the best of what i have makes sense. i've thought of tracking down a Lucid D/A used for $400 or less. but maybe i'll place that a few notches down the priority list.
thanks again!