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: LossyFlac vs Opus 256 (Read 1128 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

LossyFlac vs Opus 256

Hi all,

Long time lurker first time poster.

I have 2 hard drives full with flac files from my CD collection, i would like to listen to some of these on the go on my ipod (rockbox) but i'm a little conflicted regarding which format to use.

On some samples with my PHILIPS Fidelio X2HR headphones, although entry level headphones, i feel like i'm able to hear small differences in background drums / cymbals on some tracks all the way up to Opus 256kbps. I know that even Xiph state that 128kbps should be transparent but it doesn't feel that way to me (is it psychological?).

I would like to have the best quality with all the background parts of the song easy to hear as per the flacs but i would also like a compromise in size vs quality.

This has lead me to consider LossyFlac (extraportable) which i won't claim to fully understand how the loss in quality for the bitrate happens as the Spek shows it to look identical to the untouched flac, or Opus at 256kbps.

My question is as both these options render a similar file size which would give the best transparent sounding files?

I know that Opus uses high pass filtering where as the flacs bit depth is altered with LossyWAV>Flac .

So far with ABX tests it seems both are identical to my ears so which one would be closer to the original flac?.
 
 Any other suggestion welcomed also, I've heard but never used xHE-AAC or WavPack are these good alternatives?

also is there a nice quick way to convert LossyWav to ALAC instead of flac?



Re: LossyFlac vs Opus 256

Reply #1
>i feel like i'm able to hear small differences in background drums / cymbals on some tracks all the way up to Opus 256kbps.
These feelings are common. It's not impossible to hear a difference but it's most likely a psychological effect. If you're a long time lurker you probably heard about ABX test. A blind test is the only way to be sure if what you hear is really what you feel and to remove all "placebo" effect.
Opus at 192 kbps should be transparent with almost everything and for almost all listeners:
https://hydrogenaud.io/index.php/topic,120007.0.html

>My question is as both these options render a similar file size which would give the best transparent sounding files?
I'd first say that LossyFLAC's bitrate is significantly higher (~310 kbps) than Opus 256.
-extraportable is the lowest LossyFLAC preset. Quality seems really high but I don't think it's supposed to be fully transparent.
My bet is that Opus 256 kbps is smaller than lossyFLAC and probably a bit closer to full transparency.

>So far with ABX tests it seems both are identical to my ears so which one would be closer to the original flac?
I don't even sure the question makes sense. Take a FLAC file, increase the gain by .5dB: difference is audible but the modified FLAC is objectively and mathematically very close to the original. A transform codec will change much more things but it'll sound identical to the reference.
What would you keep (quality wise): a lossless copy lowered by .5 dB or a 192 kbps MP3 encoding that sound identical to the reference? Look first for transparency, then efficiency and compatibility.

>Any other suggestion welcomed also, I've heard but never used xHE-AAC or WavPack are these good alternatives?
xHE-AAC is a competitor to OPUS (=modern transform codec). It's very good but at 256 kbps there's no real benefit. WavPack is technically closer to LossyFLAC.

A possible way to go would be to encode your library in WavPack Hybrid. Then you'll get two files for each track: a lossy one and a correction one. The combination of both file has almost the same size than a pure lossless encoding. With a good explorer software you'll be able to copy and paste your folders to your portable device and it'll only copy the lossy part of your encodings. It's a very clever solution if you want high bitrate files on your portable devices and keep lossless on your main hard drives. And you don't have to handle to separate libraries. You really should try it ;)
Wavpack Hybrid: one encoder for all scenarios
WavPack -c4.5hx6 (44100Hz & 48000Hz) ≈ 390 kbps + correction file
WavPack -c4hx6 (96000Hz) ≈ 768 kbps + correction file
WavPack -h (SACD & DSD) ≈ 2400 kbps at 2.8224 MHz

Re: LossyFlac vs Opus 256

Reply #2
Note that the WavPack hybrid solution requires players that support the whole thing. foobar2000 on desktop and on mobile will do. On Android you might need to use either an SD card or a player app's directory - explained at https://hydrogenaud.io/index.php/topic,122768.msg1013785.html#msg1013785 . Don't know how it works on iOS.
On desktop, beware that not all players that support WavPack, will support hybrid. VLC didn't the last time I checked.

As to the question on ALAC: No, ALAC cannot utilize LossyWAV. LossyWAV works by selectively zeroing out the lower bits, and FLAC, WavPack and TAK (and OptimFROG, but you won't be playing that on portables) can know that "we can just compress as N bits and pad up with 16-N afterwards" (or 24-N or whatever) - but the ALAC format doesn't support that trick.

Re: LossyFlac vs Opus 256

Reply #3
-extraportable is the lowest LossyFLAC preset. Quality seems really high but I don't think it's supposed to be fully transparent.
lossyWAV Beta 1.4.3c includes, for testing at this time, four more quality options below extraportable (-q -5): "unadvised" (-q -10), "dubious" (-q -8.75), "questionable" (-q -7.5) and "aggressive" (-q -6.25). These will yield lower resultant lossyFLAC bitrates, with the consequent increased risk of noticeable added noise as the quality numeral reduces.

Re: LossyFlac vs Opus 256

Reply #4
I'd recommend you to find Gurus listening tests and go for a robust, very compatible (maybe you want to use your converted files elsewhere) and space saving format: Apple LC AAC. Become more opportunistic and don't waste too much energy into theoretical problems. Example: In one car, I can only use 16 GB sticks, so I encode to Q64 or Q73 (say, q64, depending on the audio, ranges from ~130 down to 80 kbps, pads, lowpassed, piano music, in general few transients seem to qualify for low bitrates). Totally perfect. In another car, I use LAME mp3 @V2 - just because that one does not display embedded album art from the AAC files.
Good headphone, I personally would probably go for Apple AAC @ q82 or q91
And don't hunt for problems, it's the beginning of the end. It distracts. Enjoy the music. Don't forget, your cat walking by has a bigger impact on your audible experience than the almost academic problems we tend to find.

I'm saying that, because in the past I perfectly heard if magnetic tape heads were not correctly adjusted. Hence I couldn't 100% enjoy the music, everyone else could.

Learn to make things right but then free yourself from over-critical listening ("i feel like i'm able to hear small differences in background drums"). It makes few sense. In a concert, there is huge differences at every moment because of the audience.