As a new user of WL Elements 9, I’m having trouble. So thanks for taking the time to read this and possibly reply.
This may sound stupid, but why is there a Master Section? To say it another way, why are some transformations (e.g., pitch bend, invert phase, time stretch) in the Edit window and others (e.g., equalizer, resampling) in the Master Section? What is the conceptual difference, or the problem that is solved by separating them this way?
In the Master Section, why does Final Effect/Dithering come after Resampling? Isn’t it best practice to dither before bit-depth reduction?
Is there a way to change the level of monitoring playback without changing the volume of the rendered file? I want to reduce the level of playback going to my DAC without changing the music track itself, but I can find only one level control, in the Master Section, and it seems to change both.
The master section is where all files loaded in WL go through. So you could be working on multiple files and they all get the same treatment from it. For some users that is a disadvantage, for others it saves a lot of work - setting up the same EQ, limiter, resampling, dithering all in one go.
You got that wrong somehow, dither should always come as the last step done after reducing bit depth and before saving the new file.
I don’t think it is in WL Elements, but in WL Full there is a playback section after everything else that is left out of processing. All you can do if it’s not in Elements is turn down your monitoring, or see if your audio interface can control output level.
About #2 (dither), most references I’ve seen say it should be applied “when” reducing bit depth. But when a sequence is defined, what I’ve seen says to use it just before reducing bit depth. For example, the WLE9 manual says: “Dithering changes the sample resolution, but not the sample size. For example, when dithering 24 bit to 16 bit, the file will still be 24 bit in size, although only 16 bits of information will have significance. When rendering to a 16-bit file, specify the file resolution to avoid wasting space.” And Wikipedia entry on dither says: “Dither should be added to any low-amplitude or highly periodic signal before any quantization or re-quantization process.”
About #3, thanks again. I gave up trying to find a nonexistent control in WLE and turned down the monitoring level at the DAC.
You’re welcome. But back to dither The quote from the WLE manual is about the specific case in WL, since WL (unlike most other DAWs) does NOT change the bit resolution of the resulting file unless you tell it to at the saving stage. So after processing a 24-bit file with bit reduction to 16 bits, the program will still show 24 bits, even though the last 8 have no real data. The dithering DID take place at the 16th bit, so cutting off the last 8 would normally not be considered bit reduction - but looks like it in this specific WL case. The Wikipedia entry is simply too vague to explain either way.
What dithering does though, is randomizing the signal at the lowest significant bit. So if an original 24-bit sample has all 1s (ones) at bit 16 and 17 for example, simply cutting the files 8 lowest bits to attain a 16-bit file would result in a string of 1s (ones) at the lowest significant (16th) bit. This becomes audible as distortion/noise and therefor dither is needed to create randomisation of this bit, changing it randomly in ones or zeroes. So technically that should be the last stage before saving.
That is correct. And that is why it is in the last position in the signal path in the master section, just before you reduce the bit depth. The dithering plug-in you choose will normally apply dithering and reduce the bit depth of the output signal from the plug-in according to the chosen value… but reducing the bit depth of the actual audio file is only accomplished by the choice of bit depth when you save/render the file (which of course you’d normally set to match what you chose in the dithering plug-in).
P.S. In your original post above you may be getting mixed up with resampling and bit reduction (point 2). Remember, these are completely different things.
The way I prefer to think of it is that the choice of how many bits to store is not “reducing” the bit depth - which implies some kind of process - but simply enabling you to store only the number of bits which are currently significant (i.e. 16 in the typical case after dithering).
I don’t find WaveLab’s behavior to be that strange but maybe I’m used to it.Dither and bit-depth reduction are two different things.
You can render to a 16-bit WAV file without dither and just truncate those bits above 16, though not a good idea.
You can dither to 16-bit but accidentally render to 24-bit or 32-bit floating. What you have now is 16-bit audio in a 24-bit or 32-bit floating container.
The problem is that most dithering is done from a dither plugin or dither built into a limiter, and there is really know way for WaveLab to know the desired bit-depth of the end result file unless you specify it when rendering.
Also, I think it’s been answered but Final Effect/Dithering is after the Resmpaler because changing the sample rate will increase the bit-depth to 32-bit floating and will also increase the peak level readings. You would not want to dither before more processing is going to occur.
So, it’s best to wait until all processing is done to apply a final limiter to preventing overs/clipping and then dither to the desired bit-depth.
No, not have reduced - that would be too late. Conceptually it is part of the process of reducing the bits, but WaveLab breaks that process into two parts, as do many software tools, meaning that we are left with the responsibility of ensuring that the two parts are done as a pair. Any processing after dither and before the matching shortening is an inappropriate action which destroys the integrity of the process.
Absolutely. That is important to point out. I should have written something like “The only problem is you are not going to need dithering unless you actually intend to reduce the bit depth.”
After thinking about it, Pro Tools is really the same way.
When you go to make a bounce, you have to determine the bit-depth. Best practice is to have a dither in your master section that is set to the correct bit depth of the end result file.
A few apps have dithering built into their render settings but I don’t think I’d prefer that because you usually have very little control over the dither settings like auto-black etc.
What’s not clear is if you are chaining the sample rate of your Pro Tools bounce, if that increases your bit-depth again but I’m guessing it does. Not that it matters on this forum but just thinking. I wouldn’t use the Pro Tools SRC anyway but I know lots of people likely do.
I think the way it’s set up in WaveLab is great, you just have to be aware of what you’re doing.
Just a final follow-up to all who answered. It was very helpful! I feel I’ve got a pretty good grasp of WaveLab (Elements) by now, and the workflow seems pretty natural.
The quote from the WLE manual is about the specific case in WL, since WL (unlike most other DAWs) does NOT change the bit resolution of the resulting file unless you tell it to at the saving stage. So after processing a 24-bit file with bit reduction to 16 bits, the program will still show 24 bits, even though the last 8 have no real data. The dithering DID take place at the 16th bit, so cutting off the last 8 would normally not be considered bit reduction - but looks like it in this specific WL case. The Wikipedia entry is simply too vague to explain either way.