h a l f b a k e r yIs it soup yet?
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
An audio compression format that works the same way as progressive jpegs, where the downloaded file becomes progressively "clearer" the more of it you download.
This way, if you had only downloaded 2MB of a 4MB audio file you would have the entire song, but it would be at 64kbps quality rather
than 128kbps.
"I don't curse what I can't change" 11.6 seconds @ 20kbps MP3 (11KHz s/r)
http://www.flash.ne...tastranger11mp3.wav
From "I am not a stanger to the rain", Children of Eden, by Stephen Schwartz [supercat, Feb 04 2001]
"And for a boy..." 12.3 seconds @ 24kbps MP3 (22KHz s/r)
http://www.flash.ne...andforaboy11mp3.wav
From "I am not a stanger to the rain", Children of Eden, by Stephen Schwartz [supercat, Feb 04 2001]
Ogg Vorbis bitrate peeling
http://www.xiph.org...is/200106/0023.html
A way of storing pieces of the stream so that a sender can downgrade the quality without having to re-encode. [jutta, Feb 04 2001]
[link]
|
| |
Use of such a format would of
course preclude streaming. |
|
| |
I think this is a bad idea, because it would lead to the proliferation of poor-quality MP3s. |
|
| |
If you're going to do away with streaming, there are much better things that can be done than progressive audio. You could look for long-range correlations which would dramatically improve the compression ratio. For example, the average pop song has a chorus which is repeated several times. By storing only the (small) differences between the different versions of the choruses you could get much better compression, in theory. |
|
| |
I don't see it leading to the proliferation of poor-quality MP3s - you'd just be replacing all the incomplete songs you find on Napster with lower-bitrate (but complete) songs.
If you wanted the song at 128kbps then you'd set Napster to screen out everything lower than 128kbps, just as you do now. |
|
| |
Yeah, but more people would be satisfied with lower-bitrate MP3s than MP3s with the ends missing, so they'd not bother to go and search out a better version. |
|
| |
Incomplete MP3s on Napster result from people cancelling (usually outgoing) transfers before they're finished, so "screening out" wouldn't help. |
|
| |
The title of this idea sent a shudder down my spine. No-one in their right mind could possibly be suggesting MP3s of Hawkwind or Tangerine Dream. Besides the general awful mind-numbing effects, would anyone have a hard disk big enough? |
|
| |
Then I read the rest of the idea and a wave of relief washed over me. Not a bad idea really. |
|
| |
[aj] "more people would be satisfied with lower-bitrate
MP3s than MP3s with the ends missing" - that's the whole
point of this idea - nobody wants an MP3 with the end
missing, but you still wouldn't download an MP3 which was
worse quality than you wanted, just like now.| — | Jim,
Feb 05 2001, last modified Sep 10 2001 |
|
|
| |
When using raw sampling, the transition from 44KHz to 22KHz is not objectionable for most material. The difference between 22KHz and 11Khz is clearly noticeable though if the audio is pre-filtered (preferably with a brick-wall filter at 5KHz) the results are not too bad. Dropping by half again yields pretty crummy, though intellegible results, but dropping by half again will render audio totally unintellegible. |
|
| |
Personally, I have found that sample resolution (16-bit vs 8-bit) is often more important to audio quality than sample rate, at least on the high end; for most material I've found that 11KHz 16-bit sounds better than 22KHz 8-bit. Of course, at the low end 5.5KHz 8-bit is going to sound much better than 2.75KHz 16-bit. |
|
| |
Compression, however, can do much better than raw wave storage. I personally am amazed at how well .mp3 works at low bit rates; check out the two clips from Children of Eden (~12 seconds each) and you'll notice the audio quality is quite pleasant despite the fact that they only take about 3KB/second. |
|
| |
(Of course, since WAV files are so large, this "progressive WAV" format would be sent compressed.) |
|
| |
In which case we now need a way to compress .WAV's considerably! Algorithms to do this just don't exist! |
|
| |
This sounds a lot like the bitrate peeling feature in Ogg Vorbis. |
|
| |
A little, yeah, if you think of the whole song as one big packet. ("Ogg Vorbis" sounds like fake Latin from a Terry Pratchett novel, but is actually a free-as-in-software replacement for MP3. See link.) |
|
| |
"Algorithms to [compress .wav considerably] just don't exist!" -- zapped
In this discussion, .wav has been used to mean "raw PCM audio", similar to what [ravenswood] explained. Given this meaning, and given the plethora of excellent speech and music compression algorithms out there, that's an astoundingly false statement. |
|
| |
(".Wav" is actually an encapsulation format that can contain any number of different encodings, some of which use sophisticated compression algorithms. In the general sense, you can't rely on being able to further compress .wav since it might already be very strongly compressed.) |
|
| |