Jump to content

Lossless Atrac

Rate this topic


Recommended Posts

I am a MD user since 99, and the R50 is still the best unit out there.

basically with all this new format stuff, i have dabbled with ipod and decided to stick with MD. my main problem is preserving my recordings long term 20yrs down the rd without loosing the quality

initially i was considering do my live recordings in wav, keep the wavs in some lossless format and then another set in mp3/4 or whatevers the coolest out.

however i just gone off the idea of having to carry my entire collection around with me, i prefer to actually physically hold my recordings irrespective of what im going to play them on. like i dont want a hdd player for the car, then the house then the office and manage keeping everything in sync! rather have a protable medium and players.

<-------------->

so now i solved that, i dont fancy storing recordings in Wav for obvious reasons.

so im thinking is there a way to preserve the quality long term between formats.

one day when i transfered a md to pc (digitally) and did my editing and then recorded it live back to MD (digitally) i did not notice any degredation in quality (source and destination was lp2 and stored on pc in wav)

so im thinking MD records what it hears, so in theroy if we transfer our audios LINERARLY! (assuming same if not higher bitrates) we should not observe any degredation in quality???

so i propose somone record a track to pc and record back to md, all digitally. do this 15x and compare the quality to the origional track and lets see if the quality stays the same, so in a sense we have achicved lossless atrac in a sense after the initial compression

it would also be interesting if somone compared this to recoring to pc and then actually encoding it into lp2 and using the checkout. record lineraly bak to pc and encode the wav again. do this 15x and let see the diffference.

i suspect there will be no difference in quality on the linear one what so ever

but the encoded one will be messed up.

interesting implications if im correct!

Link to comment
Share on other sites

Guest NRen2k5

Mandeep, unfortunately you're wrong.

ATRAC is just like MP3 and every other lossless codec, in the sense that it doesn't have any way to "know" what it is being told to encode. It looks at what is coming in, and tries as well as it can with its restrictions to faithfully reproduce that.

The problem is that lossy files always contain some distortion resulting from their original encoding. And when you run that file through another lossy encoder, it tries to reproduce what it sees, including those distortions! And the result is more distortions, because the original distortions are difficult to encode.

Link to comment
Share on other sites

  • 3 weeks later...

Mandeep, unfortunately you're wrong.

ATRAC is just like MP3 and every other lossless codec, in the sense that it doesn't have any way to "know" what it is being told to encode. It looks at what is coming in, and tries as well as it can with its restrictions to faithfully reproduce that.

The problem is that lossy files always contain some distortion resulting from their original encoding. And when you run that file through another lossy encoder, it tries to reproduce what it sees, including those distortions! And the result is more distortions, because the original distortions are difficult to encode.

It all depends on the complexity of the incoming signal. You can create a bitstream that will compress and decompress perfectly, if you tailor it to the characteristics of the codec. Of course, I've only done this with synthesized music consisting of a small number of pure tones. Doing it for real vocals or analog instruments would be a real pain...

Link to comment
Share on other sites

ATRAC, although arguably quite better than MP3 in terms of generative degradation, still suffers from it. I used to know about this website (Can't remember where it was) that showed what happens when you ATRAC files over and over again, and it had samples for the first 20 gen copies, and another sample of every 20 gens after that, up to about 200, I think it was. First 10 times really weren't that bad, 20 had noticeable degradation, 60 was horrid, and I don't even want to mention 100...

Link to comment
Share on other sites

Guest NRen2k5

I've been thinking about that statement and I think that it isn't quite right. LP2 will suffer just as badly as MP3@128kbps.

Over a short number of generations (<10), both MP3@256 and SP should fare pretty well, by sheer virtue of their high bitrate. I imagine SP would win out over a larger number of generations, because as far as I know it's a subband-only codec, and MP3 is a subband/transform hybrid codec.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...