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: Need advices for listening tests (Wavpack lossy) (Read 9979 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Re: Need advices for listening tests (Wavpack lossy)

Reply #75 –
Quote from: rutra80 on
LossyWV (U & -hhx6) turned out significantly larger than LossyFLAC (U & -ep8) so I matched kbps of WavPack lossy versions to LossyFLAC by bruteforcing -b:
You did flac -b 512, but did you do wavpack --blocksize=512?
(Is that a thing for WavPack lossy?)

Re: Need advices for listening tests (Wavpack lossy)

Reply #76 –
Quote from: Porcus on
Quote from: rutra80 on
LossyWV (U & -hhx6) turned out significantly larger than LossyFLAC (U & -ep8) so I matched kbps of WavPack lossy versions to LossyFLAC by bruteforcing -b:
You did flac -b 512, but did you do wavpack --blocksize=512?
(Is that a thing for WavPack lossy?)

WavPack also requires a --merge-blocks in addition to the --blocksize=512 to do well with lossyWAV, but even then it won't quite match FLAC. But it should be kinda close.

And no, those options are not applicable to WavPack lossy and will do harm.

Those tic-tic-tic artifacts are issues at block boundaries with the extra modes. My previous work made improvements there, but I'll experiment with these samples. The fact that disabling joint stereo fixes it means there might be an easy bug responsible, but I'm not too hopeful.

Of course, this is why < 3 bps is sort of hit and miss.

Re: Need advices for listening tests (Wavpack lossy)

Reply #77 –
I did both --blocksize=512 and --merge-blocks and didn't mix LossyWAV with WavPack lossy of course.
LossyWV seems less efficient than LossyFLAC (several dozen kbps difference), looks like there's no point using LossyWV over native WavPack lossy.
 

Re: Need advices for listening tests (Wavpack lossy)

Reply #78 –
Quote from: rutra80 on
LossyWV (U & -hhx6) turned out significantly larger than LossyFLAC (U & -ep8) so I matched kbps of WavPack lossy versions to LossyFLAC by bruteforcing -b:

(...)
Thanks @rutra80 @shadowking @bryant for these additional testing and explanation about the tic-tic (minor) issue. The latter was only audible with two extreme samples at the lowest possible bitrate. It appears as an exceptional and low intensity issue.
I'm also glad to read that increasing the bitrate quickly solves the problem.
My next test will be the same 70 samples at -b3, as suggest by David two weeks ago. Average bitrate is around 270 kbps. I expect it to be much more difficult for me especially for the 55 first samples.

I would also like to remind you that these last 15 samples were specifically chosen because Wavpack produced audible defects, perceptible even at relatively high bitrates (but without the -x switch). It is highly likely that for this selection of files, WV will lag behind all others encoders. This is to be expected. Therefore, it is important to be cautious not to draw general conclusions from a comparison between two formats or encoders based on samples known to cause difficulties with one of them.
Wavpack Hybrid -c4hx6