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: Uses for wavpack's correction file? (Read 6406 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Uses for wavpack's correction file?

I was looking at wavpack, and I saw the part about the correction file and thought "Wow that's the coolest thing since sliced bread!".  However I thought about it a bit more, and I can't seem to think of a use for it.  The only thing I could think of is not having to transcode if you want a smaller set of your song collection.  Anyone have any other uses for hybrid mode?

Uses for wavpack's correction file?

Reply #1
No I think that's the one good point about it.
Actually it should be the other way around. You have lossless files. But you can play them even if they are smaller (share or put on a smaller system).

I hope small hardware device will have support for the lossy part of such files ASAP.

Uses for wavpack's correction file?

Reply #2
I have a DAT DDS3 tape drive in my PC and a small number of tapes.  Space on my hard drive is limited but so is space on the tapes, so I use OptimFROG DualStream (--mode fast --quality 3) and copy the correction file to tape while keeping the ofs file on my PC.  At some point in the future when I have more abundant disk space I can easily restore the correction files from backup and recreate lossless files without having to re-rip from the CD, which of course I may have lost or scratched in the meantime.

Until that time I get good quality files to listen to that take about two-thirds of the size of lossless.  And the resulting size of both the ofs and ofc correction file combined is on average (very slightly) less than a flac -8 of the same track, and not significantly larger than a standard OptimFROG ofr file.

I guess this is fairly specific to me as I happen to have the tape drive already for other reasons.  But I can't see any reason why a similar strategy could not be used with other backup media where space is limited on both the hard disk and backup storage.

EDIT: typo.

 

Uses for wavpack's correction file?

Reply #3
I wondered the same thing (uses for correction file). I switched to Wavpack because of the hybrid mode & correction file, but then found I didn't have much use for the correction file.

One reason is because I would like to use Foobar to transcode into whatever lossy format, but there's no way for it to combine the correction file (that I know of, anyway). So it was two steps: join the two files with the front end, then use Foobar to compress the wav. It was easier for me to just have the one lossless Wavpack file and have Foobar transcode it in one step.

Still, Wavpack rocks, I love it and look forward to the new final 4.0.


Uses for wavpack's correction file?

Reply #5
Quote
One reason is because I would like to use Foobar to transcode into whatever lossy format, but there's no way for it to combine the correction file (that I know of, anyway). So it was two steps: join the two files with the front end, then use Foobar to compress the wav. It was easier for me to just have the one lossless Wavpack file and have Foobar transcode it in one step.
[a href="index.php?act=findpost&pid=225607"][{POST_SNAPBACK}][/a]

AFAIK, every version of the foobar plugin does use the correction file if it is the same folder as the wv file. This is also true for dBpowerAMP and winamp. The winamp plugin has an option to disable this if you want to compare or reduce disk seeking or network traffic. I found it difficult to argue that having two files was better than a single lossless file if it didn't work this way. Did you try this and find it didn't work? You can easily tell when playing in foobar by looking at the bitrate or under file properties it will say "hybrid lossy" or "hybrid lossless".

BTW, thank's for the support! 

Quote
I hope fb2k will be able to use the correction part when it's in the same Matroska file...
Me too! 

Uses for wavpack's correction file?

Reply #6
What approximate bitrate are OptimFROG and WavPack's lossy files? How do they compare in terms of audio quality to, say, LAME --alt-preset standard?

Uses for wavpack's correction file?

Reply #7
Quote
What approximate bitrate are OptimFROG and WavPack's lossy files? How do they compare in terms of audio quality to, say, LAME --alt-preset standard?
[a href="index.php?act=findpost&pid=225676"][{POST_SNAPBACK}][/a]

The first step is around 200 kbps, which is close to lame --preset standard.
About quality: I suggest you to make your own opinion. This kind of encoders suffers from different problem than transform/perceptual encoders. Noise is the main concern, especially at ~200 kbps. In my opinion, quality is not comparable to a modern transform/subband encoder at the same bitrate: they are too noisy, whereas usual lossy format could reach transparency or near-transparency on most samples.
DualStream/Wavpack lossy are more interesting at higher bitrate (250...350 kbps, depending on sensibility).

Uses for wavpack's correction file?

Reply #8
Quote
What approximate bitrate are OptimFROG and WavPack's lossy files? How do they compare in terms of audio quality to, say, LAME --alt-preset standard?
[{POST_SNAPBACK}][/a]


From the [a href="http://losslessaudiocompression.com/DualStream.php]OptimFROG DualStream webpage[/url]

Quote
Quality (--quality) usage guidelines:
  0: ranging from 183 - 265, on average 236 kbps
     should be distinguishable from the original in ABX tests
  1: ranging from 209 - 295, on average 268 kbps
     possibly distinguishable from the original in ABX tests
  2: ranging from 242 - 326, on average 302 kbps
     possibly distinguishable from the original in ABX tests
  3: ranging from 277 - 362, on average 339 kbps, default
     normally undistinguishable from the original in ABX tests
  5: ranging from 351 - 443, on average 418 kbps
     transparent, suitable for archiving and transcoding
  6: ranging from 392 - 486, on average 458 kbps
     transparent, on average 3:1 compression for CD audio


So for DualStream for a --preset standard equivalent you're looking at probably quality 3 as a minimum: average 339 kbps.

Uses for wavpack's correction file?

Reply #9
The hybrids have more advantages than most realise. For one you can create lossless and lossy files at once. If you want you may keep the lossy on the hard drive and the correction file somewhere else for extra security without eating space.

Another compelling reason is lossless performance at half the bitrates.  Many people struggle between choosing a perceptual lossy and true lossless so that they can have the best quality for listening and transcoding. Unfortunately the size gap is huge 190k vs. 870k. A hybrid lossy at around 400k is more attractive. Thats 1200 albums on a 200gb hdd, 30 albums on dvd, 5 on CD-R

Try to abx dualstream at Q5 - I don't think so. Try to transcode it to another lossy and spot the difference - very hard. Den mentioned this in another post and he has a big point.

The new wavpack is also higher in quality, maybe comparable to DS ?
I am seriously considering these hybrids. You can always use correction files for the ultimate security if required so there is no real risk.