Interest whetted by the results of the 64kbps test and announcement of a new MP3Pro playback plugin for Winamp, I played with encoding MP3Pro at 80kbps and am impressed by results. It seems they have a useful tweak. Does it only work with fixed rate encoding? Since it isn't going to make a mint for them, couldn't someone from Xiph coax them to offer it for Ogg? Kinda like the way Motif gave up and offered it to the Linux community...
And, as a general issue, I'm puzzled how rights to psychoacoustic reduction techniques are established. After all, hearing belongs to all who have it...If AMD can clean-room-design Intel work-alikes, why can't reduction codecs be coded to address the same listener perception quirks? Is it mainly a matter of losing regardless of rightness just from the enormous costs of defending a copyright infringement suit?
Full Version: Rights in codec tricks
QUOTE
Originally posted by ckjnigel
It seems they have a useful tweak. Does it only work with fixed rate encoding?
It seems they have a useful tweak. Does it only work with fixed rate encoding?
Are you talking about SBR?
No, it works for VBR encoding as well.
QUOTE
Since it isn't going to make a mint for them, couldn't someone from Xiph coax them to offer it for Ogg?
"make a mint" = make money?
In this case, they are making lots of mint from SBR.
They license the technology. (For companies like Ahead, Steinberg, Syntrillium...), and now SBR will become part of MPEG4v3 audio standard - what means more revenues for them.
Besides, the technology is packed with patents, so Xiph wouldn't be able to use it anyway.
Regards;
Roberto.
QUOTE
couldn't someone from Xiph coax them to offer it for Ogg?
Why would they want to? probably patented like Roberto said as well. Monty comes up with some good experimental research on his own. I am sure he will come up with something intuitive in the coming days. I am satisfied for the time being though.
Could someone (roberto!!) explain SBR to me?
It sounds intriguing if it would work properly with AAC. By properly I mean MP3Pro wasn't properly compatible with mp3. It was just a 22khz mp3 until you got a plugin for it, which isn't available on hardware players
(i'm getting out of my ;o) smiling habit, slowly!)
It sounds intriguing if it would work properly with AAC. By properly I mean MP3Pro wasn't properly compatible with mp3. It was just a 22khz mp3 until you got a plugin for it, which isn't available on hardware players
(i'm getting out of my ;o) smiling habit, slowly!)
QUOTE
Originally posted by Mac
Could someone (roberto!!) explain SBR to me?
Could someone (roberto!!) explain SBR to me?
Here is the info I got from AudioCoding's Wiki:
(I didn't link to the article because the page is still messy)
QUOTE
During the encoding stage, the SBR encoder stores some information in order to try reconstructing properly during the decoding stage. It stores from which part of the original band-limited signal the upper frequencies should be replicated, and also the scale of those frequencies. That information is very small compared to a conventional encoding.
SBR is guessing the nontransmitted higher frequency range of a compressed audio file by some helper bits (transmiited with the stream) and the transmitted base band. SBR allows a restoration (not reconstruction) of the upper frequency range without lots of bits. Of course, this might lead to audible artifacts. It was developed by Coding Technology, and is useful for medium and high quality coding at low data rates. It is used by Digital Radio Mondiale and MP3 Pro.
Basically, it can be defined as replacing subbands or scale factor bands by
noise and some sine components.
Hope that helped.
Regards;
Roberto.
www.drm.org . Maybe they will make a mint!
QUOTE
Originally posted by ckjnigel
www.drm.org . Maybe they will make a mint!
www.drm.org . Maybe they will make a mint!
They who? DRM or CT?
As I said before, CT is indeed making lots of money from SBR trechnology. So, I doubt they would put it in the public domain. (I.E: Without license fees as well as open specs and source distribution)
BTW: DRM uses AAC+ (AAC+SBR)
I take it AAC + SBR has already been developed by one of the big companies then?
And also, listening to the samples, if those were seriously samples of AAC + SBR at ~20-25kb/s, then that will be so awesome!!!! I hope Ivan's does better as it should!!
And also, listening to the samples, if those were seriously samples of AAC + SBR at ~20-25kb/s, then that will be so awesome!!!!
I hope Ivan's does better as it should!!
QUOTE
Originally posted by Mac
I take it AAC + SBR has already been developed by one of the big companies then?
I take it AAC + SBR has already been developed by one of the big companies then?
Yes. But, obviously, there's no encoder publicly available.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.