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: the origins of MP3 (Read 7796 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

the origins of MP3

I'm just curious as I spent most of my time yesterday reading a few topics I missed over the last yr - since the Klemm's pc donations started. And I came across discussions where the support for MPC in portables was treated.
What was it that actually set the MP3 apart to become worldwide spread? Was the MP3 the only lossy format at the time? If yes, for how long? If not, what were the others and why was the MP3 better than those?

And I'm curious to know why u think it was so difficult for the MPC to become widely used - as a lossy format I mean, at the time the portables weren't such a market joy yet, hasn't the MPC just got time enough for a spread?

Back in the early days of MP3, how did ppl usually find about it and why didn't that also happen to the MPC?

the origins of MP3

Reply #1 –
In MP3, there's an important concept: MPEG. MP3 is MPEG I layer 3, and MPEG is an association of many industrials and editors. MP3 is therefore a standard (ISO), and their products are widely adopted.

MPC is something different. It's an encoder originally developed by one single person. Most industrial (manufacturers, software editors) have never heard about it. Information about patents are vague. Last but important point: many people don't know anything about MPC (anyway the format doesn't fill their need: small but efficient encodings, such 64-128 kbps). Consequently, the market for MPC is very, very small and therefore uninteresting for the industry.

Vorbis was more popular, and despite of not being a standard like MP3 or AAC, despite of not having the strength of a company like Microsoft behind (WMA), vorbis had an important basis, and is therefore a significant market for industrials. That's why you have more and more DAP supporting Vorbis.


To get a chance of adoption, I would say that MPC need to be attractive for a lot of people, and therefore offer an interesting quality at low bitrate, like vorbis. MPC was developed for a niche.
Wavpack Hybrid -c4hx6

the origins of MP3

Reply #2 –
thanks guruboolez for your reply. I finally went outside just to wait for an answer here and that reminded me the world that's out there. So I read a couple of articles. And I guess my only question to this topic now is: why can't the MPC become a standard as well? What is it that impedes the MPC to become a standard such as the MP3 is? Because if I'm not mistaken, it all started as a project to transfer music over a phone line and that ended up being MP3 (MPEG I Layer 3). So there wasn't exactly much of the industry efforts to develop MP3 but the work of two men. Though I don't understand exactly how the MP3 has become a standard (MPEG). Why can't the MPC do the same?

the origins of MP3

Reply #3 –
Quote
What was it that actually set the MP3 apart to become worldwide spread? Was the MP3 the only lossy format at the time? If yes, for how long?


It was developed by the Fraunhoffer IIS Institution in Enlargen, Germany under the direction if Karlheinz Brandenburg in 1987 as an ISO standard  . It become widespread with the first early peer-to-peer application like Napster in 1997 invented by Sean Fanning a Northeastern University student  and then it took off with companies like mp3.com once people realized you could compress a CD audio file down using perceptual audio coding and still achieve transperancy.


Quote
Vorbis was more popular, and despite of not being a standard like MP3 or AAC, despite of not having the strength of a company like Microsoft behind (WMA), vorbis had an important basis, and is therefore a significant market for industrials. That's why you have more and more DAP supporting Vorbis.


Vorbis was invented in 1994 by Christopher Montgomery a.k.a Monty. He didn't release source code until 1999 when Fraunhoffer starting to impose liscensing fee's on their MP3 encoder. Ironically he wasn't even going to create a lossy codec to begin with  . When he was a student at MIT he wanted to design an audio editor something very similiar to present day Audacity or Adobe Audition.  A friend urged him instead to get into perceptual audio coding. His earlier work was a lossless codec called Ogg Squish much like FLAC, but not as advanced.  I am sure he will work on it again one day  . The source code is still floating around.  Vorbis is truely the only open-source lossy audio codec out and claimed to patent free.  He has really pushed for hardware support in the last few years and helped with this process as company's started to become more interested once Tremor code was released in 2002, because of the fact that it's a fixed-point decoder.  . It found it's way into a lot of games as well, because a lot of game developers find the API easier to use 
budding I.T professional

the origins of MP3

Reply #4 –
Quote
Though I don't understand exactly how the MP3 has become a standard (MPEG). Why can't the MPC do the same?
[a href="index.php?act=findpost&pid=313436"][{POST_SNAPBACK}][/a]

A standard has a basic purpose: to avoid multiplicity (of norms, products, etc...). That's why we don't have several standard, and rarely more than two competitors (like DVD-R/DVD+R, HD-DVD/BluRay, MiniDisc/DCC, etc...).
The MPEG replaced the MP3 standard by AAC (MPEG-2 and MPEG-4). There's no place anymore for another audio format. The current standard has to compete with proprietary products like WMA, Atrac or MP3pro and with free tools like Vorbis or MPC.

MPC could become a standard or at least a semi-one if:
- the MPEG group standardize it (no chance to happen)
- a big company (Apple, Creative...) buy or promote it (chance are infinitesimal)
- a lot of people decide to use MPC and then become a target for manufacturers (chance are really small).

Now it's clearly too late for MPC. It's very hard for any new audio format to capture the market against established format as MP3 and WMA. Even AAC (ISO!) has trouble to conquer an important place in the market (outside Apple, few manufacturers are supporting AAC over MP3 and WMA).
Quality is not enough to catch interest. At best MPC could survive as a niche [just for fun, type "niche" in the search engine: most of the occurence of this word is associated to MPC]. I give more chance to Fidel Castro to be elected as next President of United States of America than to MPC for becoming an industrial standard in audio coding
Wavpack Hybrid -c4hx6

the origins of MP3

Reply #5 –
Perhaps Ahead could consider it, they've probably got the capacity to do it (fund MPC development or something)... or perhaps Real via their Helix Community.

Yes, I'm being unrealistic.

Ruairi
rc55.com - nothing going on

the origins of MP3

Reply #6 –
MPC is in most ways pretty limited compared to AAC, don't forget. 2-channel 44.1kHz, high-bitrate recordings are important to almost no one outside HA.org. That is ALL MPC can do! AAC can do efficient low-bitrate, supports multi-channel and DRM, and has multiple profiles for different applications (low and high complexity, high efficiency, etc.)

Yes, MPC is well-tuned for transparency and encoding speed, but AAC will be eventually as well. Even Frank Klemm has hinted in previous threads that MPC will eventually have to roll over in favour of AAC.

the origins of MP3

Reply #7 –
Quote
Quote
What was it that actually set the MP3 apart to become worldwide spread? Was the MP3 the only lossy format at the time? If yes, for how long?


It was developed by the Fraunhoffer IIS Institution in Enlargen, Germany under the direction if Karlheinz Brandenburg in 1987 as an ISO standard  . It become widespread with the first early peer-to-peer application like Napster in 1997 invented by Sean Fanning a Northeastern University student  and then it took off with companies like mp3.com once people realized you could compress a CD audio file down using perceptual audio coding and still achieve transperancy.
[a href="index.php?act=findpost&pid=313441"][{POST_SNAPBACK}][/a]


I'm sorry, but MP3 was invented to a great part in Murray Hill, New Jersey, where Karlheinz Brandenburg was a post-doc of James D. Johnston (the jj of internet infamy).  It seems that the two of them hit upon the same principle, and discovered this when they both presented papers at a NY ICASSP (in a hot, sweaty, disgusting hallway) that they could have traded and still presented properly.

After some corporate to-do, they managed to bring Karlheinz to Murray Hill for about two years (or a year, or something like that, ask the principles for a better answer), during which they submitted ASPEC to MPEG, then created Layer 3 after the politics buried ASPEC.

This is all a matter of public record as well as patent record, you know. you could easily find this out on your own.

I have no idea why you guys keep giving Fraunhofer all the credit when the main people were at Bell Labs when it all happened.

Franhaufer does, I think, deserve a great big cheer for actually promoting MP3, instead of pretending that it didn't exist like AT&T did when Bell Labs tried to make something of it.

For that matter, MPEG-2 AAC is just about all AT&T's "PAC" algorithm from before the Lucent/AT&T split.  I think you can thank Johnston and Anibal Ferreirra for that.
-----
J. D. (jj) Johnston

the origins of MP3

Reply #8 –
Quote
I'm sorry, but MP3 was invented to a great part in Murray Hill, New Jersey, where Karlheinz Brandenburg was a post-doc of James D. Johnston (the jj of internet infamy).


I have heard the name Johnston before and I have seen some other research papers he has written in reguard to perceptual audio coding (if this is the same Johnston), etc but I have never come across the name Murray Hill 

Quote
Franhaufer does, I think, deserve a great big cheer for actually promoting MP3, instead of pretending that it didn't exist like AT&T did when Bell Labs tried to make something of it.


Well yes it's Fraunhoffer that's consistently promoted it .  I thought they were the ones who took credit for inventing it as well I guess not though  . Not to deviate from the topic, but Edler also has a large role in a lot of the prior research and current MPEG-4 audio stuff. His paper on "Use of Multirate Filter Banks for High Quality Digital Audio" is very interesting read, but it's a little too technical for me (it really get's down to the barebones of the algorithm describing the twiddle factors and such) 
budding I.T professional

the origins of MP3

Reply #9 –
note: Vorbis was not invented but designed.

Regarding mp3, some parts are from Thomson/FhG, and some other parts are from AT&T.
The key difference is probably that Thomson and FhG were more into research with short term applications at this time, while Bell lab (AT&T) was more into theorical research in audio coding at this time. That explains why 1 side aggressively pushed it.

the origins of MP3

Reply #10 –
Quote
I have heard the name Johnston before and I have seen some other research papers he has written in reguard to perceptual audio coding (if this is the same Johnston), etc but I have never come across the name Murray Hill 

well, we should introduce you two

(Unfortunately, the building is not very occupied any more.  )
Quote
Well yes it's Fraunhoffer that's consistently promoted it .  I thought they were the ones who took credit for inventing it as well I guess not though  .

Well, yes, their web site nearly but not quite says they did it all alone.  That summary, in fact, is rather incorrect, if you'd like to, oh, say, ask Karlheinz, you might get a new view on it.
Quote
Not to deviate from the topic, but Edler also has a large role in a lot of the prior research and current MPEG-4 audio stuff.

Bernd yas quite a bit to contribute, as well.
Quote
His paper on "Use of Multirate Filter Banks for High Quality Digital Audio" is very interesting read, but it's a little too technical for me (it really get's down to the barebones of the algorithm describing the twiddle factors and such) 
[a href="index.php?act=findpost&pid=313490"][{POST_SNAPBACK}][/a]


There are many people with a contribution. I think Johnston and Brandenburg (in either order, they were sitting, as I understand, in the same room, using the same (mainframe) computer, when the work happened) are the primary inventors.
-----
J. D. (jj) Johnston

the origins of MP3

Reply #11 –
Quote
MPC is in most ways pretty limited compared to AAC, don't forget. 2-channel 44.1kHz, high-bitrate recordings are important to almost no one outside HA.org.
[a href="index.php?act=findpost&pid=313479"][{POST_SNAPBACK}][/a]


Who has anything other than two channel, 44.1 khz recordings? Audio engineers, sure, but not people ripping their cds.

the origins of MP3

Reply #12 –
Quote
MPC is in most ways pretty limited compared to AAC, don't forget. 2-channel 44.1kHz, high-bitrate recordings are important to almost no one outside HA.org. That is ALL MPC can do! AAC can do efficient low-bitrate, supports multi-channel and DRM, and has multiple profiles for different applications (low and high complexity, high efficiency, etc.)

Yes, MPC is well-tuned for transparency and encoding speed, but AAC will be eventually as well. Even Frank Klemm has hinted in previous threads that MPC will eventually have to roll over in favour of AAC.
[a href="index.php?act=findpost&pid=313479"][{POST_SNAPBACK}][/a]


Not true.  Musepack isn't quite that limited.  No more than MPEG-1 layer III.  Actually it supports one more samplerate and more channels than mp3 does.

Quote
Currently only 32, 37.8, 44.1 and 48 kHz, 1...8 channels, 8...32 bit linear PCM is supported.
Nero AAC 1.5.1.0: -q0.45

the origins of MP3

Reply #13 –
Quote
Not true.  Musepack isn't quite that limited.  No more than MPEG-1 layer III.  Actually it supports one more samplerate and more channels than mp3 does.

Quote
Currently only 32, 37.8, 44.1 and 48 kHz, 1...8 channels, 8...32 bit linear PCM is supported.

[a href="index.php?act=findpost&pid=314035"][{POST_SNAPBACK}][/a]

That's only for input. Musepack output only supports stereo, 2 channels. It doesn't even support mono.

the origins of MP3

Reply #14 –
And according to Gabriel, MP3 specifications allows multichannel encoding (but no encoder/decoder available yet).
We could also add other limitations:
- error robustness
- slow seeking
- no CBR/ABR (useful for streaming)
- not compatible with current video containers

MPC is fine for CD ripping, but that's all.
Wavpack Hybrid -c4hx6

the origins of MP3

Reply #15 –
Quote
It found it's way into a lot of games as well, because a lot of game developers find the API easier to use 
[{POST_SNAPBACK}][/a]
It also saves them having to throw $2500 a title into the [a href="http://www.mp3licensing.com/royalty/index.html]Frauncoffers[/url].

Quote
We could also add other limitations:
- error robustness
- slow seeking
- no CBR/ABR (useful for streaming)
- not compatible with current video containers[{POST_SNAPBACK}][/a]
There's a whole bunch of them [a href="http://www.mp3-tech.org/content/?Mp3%20Limitations]here[/url], in an article written by Gabriel.

the origins of MP3

Reply #16 –
Sounds like what was my impression already in another thread: vorbis may be the only lossy format which has a chance of competing with the "standards". It would be interesting to know where vorbis would be now if it would have more dev-manpower behind it. Currently, the "official" devs are busy with hardware-support and politics - and it is only developed by a handful of skilled 3rd-party devs. Meanwhile, the competition doesn't sleep. If vorbis wants to become a bigger player in the marketshare-game, then imho it's team needs some reorganization and reinforcements.
I am arrogant and I can afford it because I deliver.

the origins of MP3

Reply #17 –
Quote
Sounds like what was my impression already in another thread: vorbis may be the only lossy format which has a chance of competing with the "standards". It would be interesting to know where vorbis would be now if it would have more dev-manpower behind it. Currently, the "official" devs are busy with hardware-support and politics - and it is only developed by a handful of skilled 3rd-party devs. Meanwhile, the competition doesn't sleep. If vorbis wants to become a bigger player in the marketshare-game, then imho it's team needs some reorganization and reinforcements.
[a href="index.php?act=findpost&pid=314131"][{POST_SNAPBACK}][/a]
I think the best thing that Xiph could do for Vorbis right now is hand it over to the 3rd party developers.

the origins of MP3

Reply #18 –
Quote
I think the best thing that Xiph could do for Vorbis right now is hand it over to the 3rd party developers.
[a href="index.php?act=findpost&pid=314135"][{POST_SNAPBACK}][/a]


They published it under BSD, that pretty much amounts to that. Or do you have something differently in mind?

the origins of MP3

Reply #19 –
Quote
They published it under BSD, that pretty much amounts to that. Or do you have something differently in mind?
[a href="index.php?act=findpost&pid=314139"][{POST_SNAPBACK}][/a]
I was thinking more along the lines of letting the 3rd party developers manage the project themselves.

It's a shame Xiph aren't working on their Tarkin project yet, because the big electronics names are looking at wavelet based video compression for usage in digital cinemas. At the moment it seems motion JPEG2000 is going to be the codec of choice.

the origins of MP3

Reply #20 –
Quote
I think the best thing that Xiph could do for Vorbis right now is hand it over to the 3rd party developers.
[a href="index.php?act=findpost&pid=314135"][{POST_SNAPBACK}][/a]

Why bother even 'handing it' over?
These great 3rd party devs could just do it now, club together and have their 'own' branch of Ogg Vorbis.
Granted this would not be an ideal way; it would also likely get more industry recognition if Xiph continued development of it, but we haven't seen much from then in quite a while.

TBH I don't see this happening (the 3rd party devs getting together and having a branch of their own that is)... but you know, if they did it might kick start Xiph into action??

Plus, if Xiph (or Monty or whoever it is) start Vorbis II, and it's not gonna be back-compatible (crazy IMO) then Vorbis I could very well have a place with the 3rd party devs... and suddenly it all makes sense 


EDIT: tsk, Defsac replied before I'd finished my post