Since I started reading this forum I've tried a great many of the encoders I've read about including Wavpack and OPTIMFrog.

I think these are both great lossless encoders, I just don't understand the point of their hybrid modes. If Vorbis, Musepack, etc. give transparancy at --quality x and give respectable file sizes why use a hybrid that results in a much greater file size? Is there a benefit that I am not aware of? I know about the correction files and I think it's a great concept, but I just don't get it. I've done a site search, but I can't find an answer to this question.

(I think) one advantage of hybrid codecs over lossy codecs is that they do not apply psychoacoustics, eliminating the possibility of artifacts, which plagues every lossy codec.

I can't provide any links here, sorry, I'm useless with the search function.

The hybrids are the in-between codecs. As you've probably noticed, even lossless codecs are ~700Kbps, which some might deem too high (I do except for REALLY special albums). However, there are some that also don't feel that MPC --insane is transparent enough. If you do a search, you'll find a number of posts by Guruboolez, who has found several samples that don't work that great for MPC (at least to his hearing, which is better than most of ours).

Those who are willing to settle for 320Kbps files would do well to consider OptimFrog DualStream or WavPack Hybrid, which will be closer to transparent than any other codec.

Personally, I can't hear any of the mentioned artifacts in MPC, so I'm perfectly happy with it. But those with the warm fuzzy feeling may not.

Wavpack lossy has absolutely no issues like pre-echo, dropouts, smearing, etc.

The only artifact that can happen there is quantization noise. (AKA hissing)

Also, there are some benefits in hybrid mode:

-Storing in CDs: If you want to backup your albums in lossless format, do so in hybrid mode. After, if you lose the files in your HDD and want to get them back, just copy the lossy parts to it and leave the correction parts at the CD - no need to re-rip, re-encode.

-Online distribution/sharing: provide lossy files for people to download from your site or your FTP. If they like the music and think it's worth it, they download the correction part and have a lossless file. If they don't, they didn't waste their time downloading an entire lossless track.

-? you tell me

...and because of the lack og pre-echo, etc, my own personal testing has found that the lossy files transcode much more cleanly than mpc, vorbis, mp3 or aac at a 320 kbit bit rate. I posted a thread or two about this a while ago. This is also currently being tested in sthayashi's "little" transcoding test. (In my case the destination was ATRAC3, so I'm not sure about into others.)

Your opinion may differ.

I always took it for granted, according to logic.

Thanks, people. Very good info.

@rjamorim: I really appreciate Rarewares. Makes life a lot easier.