Master the dat?

[humble]

I was reading the sawstudio manual on dithering and one suggestion it recomends is to use the higest bit rate in saw then output that to the analog inputs of your dat. My question is, do you use the dat for mastering, or do you send the dat back to your computer to create a cd copy for mastering? And if you send the dat back to your computer, do you send it back through the digital prots or analog? :confused: http://www.sawstudiouser.net/forums/newthread.php?do=newthread&f=4#
Confused

[Cary B. Cornett]

The reference to a DAT may be a little, ah, dated. The core principle of the idea was, I think, that if you had good enough quality converters you could let the converters handle the dithering (this also is sometimed a preferred method for Sample Rate Conversion). The assumption is that you have a good 24 bit converter coming out of your SAW setup, and a good 16 bit converter going into a DAT machine or any other recording device (even another computer).

However, if you can afford really good converters for this use, you can probably also afford to get the software or plugin to handle the dithering inside your computer at the quality level you need.

You might also just consider using the dithering options offered in SawStudio itself.

[Pedro Itriago]

Well...this is one of those "I have a belly button too" kinda discussion.

There're several ways that I know of on how people deal with this issue.

You can deal with this inside SAW or another program with an SRC & dither algorithms that you think/hear/know are better. Most people are just content with SAW new SRC quality.

Or, you can find a plug-in like the UV-22 or any other in the output track of your mix.

The one you mention is actually going analog from the computer to a DAT. It doesn't has to necesarily be a DAT, you can do it between two soundcards in the same computer or between computers. You go out from a hi sample rate analog output to the input of another very good converter set at 44.1/16 (or whatever the rate/depth you want).

There's also another one on which you can send you digital signal to a hardware SRC/dithering device such a TC Electronics Finalizer or an actual Apogee UV-1000 (several other hardware gear that can do this).

Which one is better? your choice.

Many people don't like the idea or simply can hear the added noise & artifacts of software/hardware SCR/dithering algo's. So it was discussed many times in here and elsewhere that some people say it's best to make a D/A--A/D than doing such processes that manipulate the audio the send it to a lower/higher resolution it.

Again, how you do this DA--AD is up to you. Many people had DAT's (and still do) to send the audio to mastering houses/cd factories to make the cd's (many people more qualified than me for this step, like Steve Berson could expand on this). That's what's being discussed in the Saw Manual.

If you don't want to use a DAT, just don't. It won't prevent you to be able to make the conversion

[Carl G.]

Bob's new algos are pretty good.
I use them all the time when bringing suff down to CD rates.

[TotalSonic]

At this point DAT is a very dated and heading to be obsolete storeage method. Transport mechanisms for DAT machines are slowly being discontinued, and they are limited to 16bit 44.1/48lHz (except for the proprietary Tascam DA-45 which can go up to 24bit/ 44.1-48kHz) so there is no reason to invest in one except if you have mixes archived on them that you need to playback, or are in the business of mastering (where they are still occasionally used by clients as delivery formats for mixes).
DAT's are subject to compatibility problems (i.e. what was recorded by one brand recorder might have dropouts when played back in another) and also reliability problems (i.e. tapes getting eaten, dropouts from age, etc.).

also - CD replicators can not accept DAT's for direct transfer to create a glass master - they would need to pre-master to a format such as DDP image, CD-R, or PCM1630 (also basically obsolete) first.

"Back in the day" many DAT players had onboard converters that were superior to the ones on the soundcards that were common at the time - but at this point this can not be said to be the case. At this point dedicated AD and DA converters by Lavry, Mytek, Prism, dcs, Cranesong, etc. are far superior to any of the converters found onboard DAT recorders - and even the vast majority of contemporary sound cards converters spec out better. Still - unless you are mastering using analog processing (which I do for most material) there is no reason I can think of for doing a loopback out DA and back in the AD just to requanitize.

While there are some AD's that have proprietary dithering options for their 16bit captures (such as Mytek's "Super Shaper HR")
my usual procedure in mastering (and I'm sure you'll find most mastering engineers) is to capture back from the analog process chain at 24bit and then to add dither in the digital realm as the final stage after all other digital processing (including fades and brickwall peak limiting).

A good article to read on the basics of dithering is here:
http://www.users.qwest.net/%7Evolt42/cadenzarecording/DitherExplained.pdf

Also I recommend checking out the Ozone dither guide -
http://www.izotope.com/products/audio/ozone/dither_files.zip
and listening to the included files. Compare the original 24bit wav file and the truncated 16bit file (you'll need to listen in a quiet room with levels fairly cranked up) so that you can hear the distortions that truncation actually causes. Personally when requantizing from 24bit to 16bit I think always think dither should be added.

So far in my tests by far the best dither option for SAW users is the Sonoris Dither plugin -
http://www.sonoris.nl/nl/sdtr.asp
and I highly recommend it if you are searching for the way to best retain the sound quality of your 24bit mixes in your 16bit masters.

If you are sending your mixes out to be professionally mastered my own recommendation is to send a data disc containing 24bit wav files at the same sample rate that you used for the tracks.

Best regards,
Steve Berson

[Carlos Mills]

Hi Steve,


So far in my tests by far the best dither option for SAW users is the Sonoris Dither plugin -
http://www.sonoris.nl/nl/sdtr.asp
and I highly recommend it if you are searching for the way to best retain the sound quality of your 24bit mixes in your 16bit masters.

Just bought Sonoris Dither... What would be your recommended "average" settings, specially regarding this:
- TPDF (Triangular Probability Densitity Function) x HighPass TPDF
- curve A: 5th order
- curve B,C: 9th order

Thanks,

[PieterS]

Hi Carlos,

there is no "best" setting. As long as you use tpdf or highpass tpdf there is no non-linear distortion as well as no modulation artifacts.
The type of noiseshaping doesn't influence the effectiveness of the dithering basically, but it depends on the source what you like best. The higher order noiseshaping algorithms result in less noise at ear sensitive frequencies, but more noise at the higher frequencies where the ear is less sensitive.
Generally speaking, type A is a broader type of noise that can be used all around, but if you have a more pristine source, maybe a higher order can help minimizing the masking of low level details due to the noise, at the expense of a brighter noise type.
When the end resolution is 20 bits, you may not need noiseshaping at all.

regards,

[tomasino]

Do it all in SawStudio baby yea!

p.s. I have two belly buttons and I keep one in the closet right next to the tapes. Baaaghhhh!

[TotalSonic]

Hi Steve,



Just bought Sonoris Dither... What would be your recommended "average" settings, specially regarding this:
- TPDF (Triangular Probability Densitity Function) x HighPass TPDF
- curve A: 5th order
- curve B,C: 9th order

Thanks,

For most material I prefer regular TPDF with the curve B noise shaping. Best thing to do is to do some listening tests yourself to decide what you like best.
I think for most cases dither choice really isn't that critical of an issue.

I finally have a file that I think should be fairly revealing that I am going to use for my planned dither shoot out - files should be posted for download and blind comparisons in a week or 2.

Best regards,
Steve Berson