Dither in Album Publishing

[Silas]

Hello to All,

Quick question - Pow-R Dither is not available in Album Publishing but only in the mix bus, correct? Any idea why not?

Thanks,
Silas

[Perfect Record]

Good point. When possible, I only use Pow-r.
The other thing I'm hazy about in Album Publishing is whether dither is a separate process following sample frequency conversion.

[fl]

Remember that Album publishing is a two step process. The first step, when you choose "Digital Release" as the output in the "Generate CD Image/SACD Edited Master" window, is the generation of a single "master" MTFF file which is used as the source for all subsequent generations (to FLAC, MP3, etc.).

Dither will only applied at this first stage if there is going to be sample rate/word length changes and then the final product generated, as would be the case when you choose to output a "Redbook CD Image" to make a 16/44 PMI file for CD burning/DDP creation. When you choose to make a "Digital Release", the Dithering options are unavailable, and the manual suggests that any dithering options you set on the Mixer's main output Bus will be ignored, so you will generate an undithered 24 bit MTFF at the same sample rate as your Project. Remember that you only want to dither your output once, at the final stage, which in the case of a Digital Release, is not yet.

It's only when you proceed to the second, or "Album Publishing" window that you get the opportunity to choose the dither settings for each of the output formats you select. If you choose to make an output set of 44/16 WAV files, you'll be able to specify the dithering options for this particular set of files. In this way, it's possible to make more than one set of the same file type, such as a set of 96/24 FLACs - AND - a set of 44/16 FLAC files, each with the appropriate amount and "bittedness" of dithering. Look in the manual starting on page 618, but especially on pages 622-3.

If you're clever with the use of the "\" character and the wild cards when you are setting the output destination for your published files on the second or "Album Publishing' page, you can wind up with separate appropriately named folders and files for each of your file sets. An example would be to output to

Y:\(name of Project File Folder)\Published Files\<DiscTitle> <FileFormat> <SampRate>-<Wordlength>\## <TrackTitle> >FileFormat> <SampRate>-<Wordlength>

Which would yield a folder named something like
"Disintegration FLAC 44-16"
which contains a number of numbered files including
"01 Plainsong FLAC 44-16.flac"
and so on. It was a happy day when I learned about this.

It's my guess that the omission of Pow-R here is because only until recently it was a separate, paid upgrade to the main program, and was probably only implemented in the Mixer Bus structure. It's possible that adding it to the Publishing could entail an additional licensing cost which Merging is reluctant to incur, or maybe they just haven't gotten around to making this addition. Perhaps some encouragement or inquiry directed to Merging support would be in order, or at the very least, to the Pyramix Suggestions forum. On the other hand, Pyramix's Apodizing Dither, shaped or not, is as good as it gets.

[Perfect Record]

A good reply Frank, but it only glances at the question at hand, and what I'm wondering about.

Yes, the lack of Pow-r in album publishing matters. And what I'm pondering is whether Pyramix completes the SFC, then applies unspecified dither?

I'm very fussy about dither and SFC. Fortunately Pyramix SFC is the best I've heard thus far.

[fl]

I'm just assuming that Dither is always applied as the final process, so in this case, after any and all word length and sample rate conversion.

[alesC]

Any news of PowR dithering in Album Publishing ?
I'm agree with "Perfect Record", PowR is better and it would be great if it was available in Album Publishing !

[per]

Hi All.

I'm amazed (and a bit envious) that you guys can hear something happening more than 100 dB below your music. I would be very happy if you could give some advise on how I can test what different ditherings "sound" like and what I should listen for. When I demonstrate dithering to my signal processing students, I re-convert e.g. a fading piano (recorded with 24 bits) with a 4-bit AD and then we listen without dithering (where the quantization distorsion is very present) and then with different types of dithering (which has a linearizing effect). Very interesting!

In a real recording situation, I would say that room noise and pre-amp noise probably set the noise floor for the recording. The theoretical dynamic range for a 24 bit conversion is 144 dB, which is much more than any mic pre-amp. Let's say you have a super mic pre with a dynamic range of 130 dB. If your recorded sound levels top at 130 dBSPL and you have a 10 dB headroom, then you should be able to record sound levels as weak as 10 dBSPL. There are not many studios that have a noise floor lower then 15-20 dBA (and probably at least 30 dBZ), which would suggest that room noise would be the dominant "dithering noise", at least 20 dB louder than the dither noise level.

All the best
Per

[per]

Frank,

what you write about album publishing and specifying folders is absolutely amazing. I had no idea!
One question though: Do you have to create the different folders first, or are they created in the process?

Respect
Per

[fl]

Per,

They are created as part of the process.