Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: 24-bit DTS Core vs. 24-bit DTSMA downconverted to 16-bit (Read 4243 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

24-bit DTS Core vs. 24-bit DTSMA downconverted to 16-bit

Some months ago I embarked on ripping my blurays in order to put them on my NAS for easy access.  Up until just recently I'd only encountered DTSMA audio.  For my rips I'd been extracting the lossy DTS Core since it's much smaller than the lossless DTSMA and I've never heard any sound quality issues with it.

Just the other day I bought a new bluray which has uncompressed 24 bit LPCM audio.  Converting this to multichannel FLAC still resulted in a pretty big file.  After a bit of thought, I downconverted the 24-bit LPCM to 16-bit and then encoded to multichannel FLAC.  This resulted in a much smaller file and, based on everything I know, 24-bit to 16-bit conversion is totally inaudible when done with good software.  This got me thinking.  On the previous blurays I've done, rather than extracting the lossy 24-bit DTS Core, would I be better off decoding the 24-bit DTSMA track to wavs, downconverting them to 16-bit, then encoding the result to multichannel FLAC?  I tested one movie and the 24-bit DTS Core track was about the same size as the 24-bit DTSMA track downconverted to 16 bit and encoded to multichannel FLAC.  That's excellent. 

But enough backstory.  The question here is simple:  which would you think is closer to lossless, the 24-bit DTS Core track, or the DTSMA track downconverted to 16-bit.  Before you ask, yes I've tried ABXing these and I can't hear any difference.  So really this question is more of a philosophical thought experiment.  I'm inclined to think a 24-bit to 16-bit conversion is safer than any perceptual lossy codec, but I'm very interested in what other, more knowledgeable people think. 

24-bit DTS Core vs. 24-bit DTSMA downconverted to 16-bit

Reply #1 –
I would absolutely take the lossless 16 bit files.  24 bit lossy is fairly pointless.

24-bit DTS Core vs. 24-bit DTSMA downconverted to 16-bit

Reply #2 –
Well I'm not very familiar with bluray spec, but don't you want metadata of those dolby digital audio thing? (parameters for dynamic range compression, dialnorm, downmix, center/surround level or something)
Apparently these are all lost on transcoding.
If it has lossy one with all those metadata, I would happily take that one.

24-bit DTS Core vs. 24-bit DTSMA downconverted to 16-bit

Reply #3 –
Quote from: saratoga on
I would absolutely take the lossless 16 bit files.  24 bit lossy is fairly pointless.

Yeah, that is my thinking as well.  Technically the 16-bit files would be lossy too since they've been converted from the original 24-bit lossless DTSMA, but reducing bit depth is very different than going through a perceptual encoder.  I even considered trying lossywav to get the files down to 11 bits or something.  Then FLAC could really go to town.

Quote from: nu774 on
Well I'm not very familiar with bluray spec, but don't you want metadata of those dolby digital audio thing? (parameters for dynamic range compression, dialnorm, downmix, center/surround level or something)
Apparently these are all lost on transcoding.

This is a good point and something I hadn't considered.  However, searching just now, it would seem that DTSMA/DTS Core doesn't use that sort of stuff.  Well, ok, they do support dialnorm, but apparently it's not used often since it was originally a Dolby thing.  Dolby's TrueHD and AC3 can and do use dialnorm and DRC metadata.  I'm not too concerned about dialnorm as that just applies a gain to the audio and I'm ok with turning the volume knob.  DRC could be an issue, though.  Luckily, I've yet to buy a disc that has TrueHD.
 

24-bit DTS Core vs. 24-bit DTSMA downconverted to 16-bit

Reply #4 –
Theoretically, you should be right that the downsampled/reduced bit depth conversions should be more true to the original. The tiny bit of noise from dithering and likely similarly minute distortions from the resampling would be better than the discarding of frequency information done with lossy compression. And as a fringe benefit, you'd have a format with wider compatibility.