Choose a dither, or truncate in 20-24 bits?

[pipefan413]

Hah, that's interesting. My ears have actually gone a little worse over the years, but I can still easily hear the 16 kHz sound. I wonder if I would hear that sound.

Only one way to find out I'd say - test it. I think the thing is that Laserdiscs (and probably also CDs) already use dithering of their own, so dithering again probably increases the noise floor even further. But I'd just do a few experiments to see if it's audible at all. Crank up the volume in some quiet part where there is nothing but background noise and compare.

That's precisely what I've done to see if I can tell much difference but it's at that point where I'm doubting my senses. This is incidentally also why I'm obsessive about sync: if I don't get things as spot on as I possibly can, I'll always have that thought in the back of my mind... "is this actually slightly out of sync, or is it fine and my brain is being paranoid?" Similarly, I'm sitting with the two tracks side by side going "... I think it's the same. Is that noise there actually? Did I imagine that?" Plus, like I said, my equipment is very, very cheap because my income is crap and I spend all my money on Blu-rays.

Subjectivity is a pedant's worst nightmare...

[deleted user]

Haha yeah I have the same issue with sync. Maybe it's because there's no such thing as perfect sync and one can always imagine to be hearing a desync? Must be some science on that somewhere.

Oh I see where you're coming from. I kinda trust my setup somewhat well. My headphones are AKG K701 ones, not too expensive in Germany, but for strange reasons they have incredibly high prices in the US. Plus an RME interface for the headphone amp, but I'm sure there's actually better solutions for that that are also cheaper.

[pipefan413]

Haha yeah I have the same issue with sync. Maybe it's because there's no such thing as perfect sync and one can always imagine to be hearing a desync? Must be some science on that somewhere.

Oh I see where you're coming from. I kinda trust my setup somewhat well. My headphones are AKG K701 ones, not too expensive in Germany, but for strange reasons they have incredibly high prices in the US. Plus an RME interface for the headphone amp, but I'm sure there's actually better solutions for that that are also cheaper.

Indeed.

When I resynced Near Dark, the audio tracks on the same DVD were wildly* out of sync with each other, and the Blu-ray audio was an absolute mess, so picking reference points and so on was always going to be a nightmare. I also didn't want to re-encode anything because that was all lossy, which meant I was restricted by audio frame boundaries. Nonetheless, I got it down from something like 32 ms between the earliest and latest DVD audio tracks to 8 ms or something, off the top of my head.

Of course, at no point did I feel like any of that was actually noticeable, but I wanted to be as sure as reasonably possible that it wouldn't be "wrong".

Crazy person logic.

* less than 1 video frame at 24000/1001 fps

[Falcon]

Hi, after slowing down my DTS Cinema tracks on izotope and exporting in 24 bits, I tried to see what happened if I reduced to 16 bits with EAC3to:
It reduced the track to 16 bits without a 2nd pass.
This means that it did not detect any clipping during the reduction to 16 bits.
I suppose I can truncate in 16 bits directly on izotope?

[deleted user]

I could be wrong, but I think the second pass is only volume and clipping related, not dithering related. Truncating bypasses dither, whereas a single pass in eac3to probably (presumably?) still applies dither.

[Falcon]

You're right, -down16 applies a dither, madshi talked about it

[Falcon]

I'm bringing this thread up again to discuss the habits I've picked up:
I always reduce to 24 bits for stereo, and 20 bits for 5.1.
The choice of dither is not important in these resolutions because TPDF or Mbit+ stay quiet.

But for a reduction in 16 bits, TPDF seemed to me the best choice in all circumstances, because Mbit+ adds too much noise in the top of the spectrum (less than TPDF at the bottom, but much more at the top).

[Bilbofett]

if your final project is a disc, then you should dither to 24-bit, not 20-bit

[Falcon]

if your final project is a disc, then you should dither to 24-bit, not 20-bit

Lot of DTS-HD MA and Dolby TrueHD on BD are in various bit depth. 
The container show 24 bits, but you can have inside 18 or 20 bits...
The Dolby TrueHD encoder have a bit optimization option. It reduce the audio to variable bits depending on the sequence.
But I prefer to reduce myself to 20 bits entirely, it's the best compromise.