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: so what happened? (Read 30576 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

so what happened?

Reply #25 –
Reading an old post from Christian, Frank himself believe/d  that the psy model can be transferred.  Apparently he's tried to teach his psy model to the other formats people without success so presumably [the speculations are that] now he has to take time to learn another format and do it himself.

Refer to Christian's conversation with Frank (nice thread but bear in mind it is from 1 year ago so it won't help us to know if his plans have changed), maybe Citay can speak to Frank  :
http://www.hydrogenaudio.org/forums/index....ST&f=11&t=15384

Quote
He told me he is maybe interested in making a proper AC3 encoder,
> using his own psy model, that in principal can be transferred from one
> encoder to another.

so what happened?

Reply #26 –
Quote
Refer to Christian's conversation with Frank (nice thread but bear in mind it is from 1 year ago so it won't help us to know if his plans have changed), maybe Christian can speak to Frank again   :
http://www.hydrogenaudio.org/forums/index....ST&f=11&t=15384


I seriously dont think this is a good idea. Whenever i was calling Frank on the phone, after the donation, i had this feeling i was going on his nerves, or he felt at least uncomfortable. It was like his boss calling him ( you know he is writing software for his living ) asking him when the new microscope control code will be ready, and to tell him he is behind shedule.

It took me a while to understand that this is counter productive, so i stopped it. CiTay has built up a very good relationship to Frank IMO, they had contact on almost a daily basis during the PC setup, so i guess he might be the better person to try to find out if Frank has any plans to start working on audio compression again.

Sorry, this is not to steal myself out of responsibility, i just believe it wont any good if i will phone Frank up to ask if he will start coding soon  .....

so what happened?

Reply #27 –
I still don't think, that it is any good, to talk/speculate at the moment about further improvements on the psy-model or a portable psy-model. If Frank should have any time left, he should care only about the so called sv7.5, which means technical stream fixing. Maybe this is possible for him without investing dozen of dozens of hours.
iirc, this solution would already mean, video support with mpc is possible then ?
The 1.14 and 1.15r encoders are still top, see the 128k tests.
For wider spreading of this superior format, we don't need to improve the encoders regarding psy-model, only the stream fix for possible video would be fine

so what happened?

Reply #28 –
Quote
For wider spreading of this superior format, we don't need to improve the encoders regarding psy-model, only the stream fix for possible video would be fine
[a href="index.php?act=findpost&pid=253767"][{POST_SNAPBACK}][/a]


Sorry. Even if i may have a lot of enemies after stating this, i still think it should be said. If i understood Frank correctly MPC, as a format, is NOT superior to Vorbis or AAC, maybe not even to AC3.

The superior sound quality of the existing MPC encoder, and thats the summary of all my telephone conversations with him, is only the result of the PSY model that is being used. Now, if this model was transported into Vorbis, we all could benefit from an opensource, patent free audio compression format with superior sound quality compared to anything else existing, probably even better than current MPC.

Again, i am convinced this is the main reason to understand why Frank cant motivate himself to continue working on MPC. He is of the opinion that MPC would require a lot more changes from SV7 to SV8 than just changing the framing. We talked about new features that should be implemented into next generations MPC, and this would be a LOT of work, its not like tweaking the existing SV7 encoder, but more like creating a complete new compression format with compatibility for older, subband coding based files. Maybe Frank gave up on this because he realized he would never be able to invest the necessary time to make this happen.

As a result of this, i am still convinced if he just had invested the time to make the SV8 encoder using his PSY model API, allowing to reuse his PSY model in other encoder like FAAC or Vorbisenc, the opensource community would have the biggest benefit. Maybe he will find the time to do that one day. Or somebody else will try to dig his way through the existing code for the MPC encoder, and find a way to reuse this great PSY model, we will see.

so what happened?

Reply #29 –
Quote
Sorry. Even if i may have a lot of enemies after stating this, i still think it should be said. If i understood Frank correctly MPC, as a format, is NOT superior to Vorbis or AAC, maybe not even to AC3.

By looking at the source code, I can confirm the the MPC format is inferior to Vorbis I or AAC. It is superior to MP2 and AC3. Regarding MP3, on some points MPC is better, and on some other ones it is inferior.

so what happened?

Reply #30 –
Quote
If you're going to make such bold statements, please don't hide them with rhetoric.
[a href="index.php?act=findpost&pid=253690"][{POST_SNAPBACK}][/a]


Lighten up, dude. They were obviously a joke (the sentences don't even make much sense)

so what happened?

Reply #31 –
I haven't meant the format's definitions, I meant by "format" above, the mpc 1.14, 1.15r, in complete, including the psy-model. And the usage of these encoders is imo superior to aac, ogg, ac3, mp3 regarding high quality, transparency, speed.
I agree with Christian, who wrote probably the point, that Frank thinks, he has or better wants to do a complete new format for sv8, or that this is needed by the open source community, but that would mean way too much work for him.
Of course, spreading his psy-model would be an all-in-one solution.
But maybe CiTay could talk again with Frank, that Frank does not work on sv8 or whatever, instead, Frank makes only the sv7.5, which should be the least possible amount of work for him ?
(and sv7.5 would allow video&mpc) ?

so what happened?

Reply #32 –
Quote
Lighten up, dude. They were obviously a joke (the sentences don't even make much sense)[a href="index.php?act=findpost&pid=253805"][{POST_SNAPBACK}][/a]


Joke or not, I consider it incredibly disrespectful to the organizers of the donation, the contributors to the donation and to Frank himself to refer to the pc donation as either a scam or a fraud.

There are many more sophisticated ways to get a few laughs than making back-handed disparaging comments. I'll refrain from ranting about how lack of respect and integrity are very fundamental societal problems.

so what happened?

Reply #33 –
Quote
Quote
Sorry. Even if i may have a lot of enemies after stating this, i still think it should be said. If i understood Frank correctly MPC, as a format, is NOT superior to Vorbis or AAC, maybe not even to AC3.

By looking at the source code, I can confirm the the MPC format is inferior to Vorbis I or AAC. It is superior to MP2 and AC3. Regarding MP3, on some points MPC is better, and on some other ones it is inferior.
[a href="index.php?act=findpost&pid=253797"][{POST_SNAPBACK}][/a]

I am very curius how come that format which should be inferior to vorbis and aac maybe even mp3 is performing so good?
I ripped tons of cds to mpc and as a encoder is a much faster too then thoise superior formats??It decode faster too.is that becose that psu model which are you talking about,or some other optimizations?

so what happened?

Reply #34 –
Quote
I am very curius how come that format which should be inferior to vorbis and aac maybe even mp3 is performing so good?
[a href="index.php?act=findpost&pid=253856"][{POST_SNAPBACK}][/a]


The answer has already been said you must have missed it...

so what happened?

Reply #35 –
Quote
The answer has already been said you must have missed it...
[a href="index.php?act=findpost&pid=253868"][{POST_SNAPBACK}][/a]

I assume you mean the fact that the PSY model is much superior than the other formats' right now.  However, I was wondering this too--how (specifically) are Vorbis I and AAC superior, and in what respects are LAME better and worse than MPC as a format?  I understand that the PSY model is better in MPC but I don't think the specifics of the format differences were discussed.  Maybe this is too much of a derail though...

so what happened?

Reply #36 –
I think people should be talking to Monty about the the advantages of the MPC psy model, if it's patent free, etc. because if I recall Monty was going to create a new pys model for ogg.

so what happened?

Reply #37 –
Quote
I think people should be talking to Monty about the the advantages of the MPC psy model, if it's patent free, etc. because if I recall Monty was going to create a new pys model for ogg.
[a href="index.php?act=findpost&pid=253934"][{POST_SNAPBACK}][/a]


Well, last time I asked him about the ATH values from the MPC psy model, Monty said that at least for Vorbis, ATH values were not really that important.  So we can cross that one off the list and move onto the next psy model parameter...

so what happened?

Reply #38 –
Quote
Quote
The answer has already been said you must have missed it...
[a href="index.php?act=findpost&pid=253868"][{POST_SNAPBACK}][/a]

I assume you mean the fact that the PSY model is much superior than the other formats' right now.  However, I was wondering this too--how (specifically) are Vorbis I and AAC superior, and in what respects are LAME better and worse than MPC as a format?  I understand that the PSY model is better in MPC but I don't think the specifics of the format differences were discussed.  Maybe this is too much of a derail though...
[a href="index.php?act=findpost&pid=253910"][{POST_SNAPBACK}][/a]

I think the 2 most repeated things on ha forum is that his psu model is specific and the best and that lame3.96.1 has to go thru same testing torture as 3.90.3 even if half of those people who testing are using another lossy format and another half switched to lossless in meantime 
Ok,seriosly i see your point in telling me that your and proper answer is that his tweaked model of psu is the answer for my question.But;
I ripped all my cd collection to mpc-q5 with 1.14b encoder which was then and still now recomended mpc encoder.That was around 2 years ago.
Since then Ahead-Nero was pushing aac like its the only codec around,also there is apple....dolby too,some smaller projects before and after that like faac and psytel and few other which cant remember the name exactly but they were good encoders too 
I remember that Roberto who is very knowledgeble posted that Microsoft hired few maybe brightest minds in industry to develop further wma and.......They come out with wma8 which is failing miserably everywhere except on their """internal"""tests where they comparing it to mp3 with some obsolete versions of old Lame like 3.88 or so.
We got mp3 which is developed since 1987-thats when fhg started their work on mp3 afaik,went thru many developments like ..1987???That was 17 years ago????
We had it tweaked extensively here too with 3.90.1-2 and 3 encoders which were all top of the line btw,also we got a excelent Lame team..
Also we got a Sony one of the most powerfull  tehno giants and they are subject for fun with their atrac crap,and dont even think about smaller projects like real which is developing their format even before first cd burner was made and they are still nowhere..
So on one side we got all this and on another we got a one man project.guy who made it all at home in his spare time in a period of year or so???Well,he was either lucky in his attempts to enhance the quality of encoder or he`s more then a genius couse he made from "almost"inferior format to most superior in a incredible short time.
By most superior i meant much faster encoding-decoding comparing to competion like aac,ogg vorbis,mp3;low cpu usage and transparency at same or lower bitrate then others.
I do belive that his psu model is the answer why mpc skyrocketted,its just the facts i posted about now makes me feel weird aboit it

so what happened?

Reply #39 –
Quote
... So on one side we got all this and on another we got a one man project.guy who made it all at home in his spare time in a period of year or so???Well,he was either lucky in his attempts to enhance the quality of encoder or he`s more then a genius couse he made from "almost"inferior format to most superior in a incredible short time....

If I remember correct, Andrée Buschmann--the original developer of mpc--was a student at "Institut für Theoretische Nachrichtentechnik und Informationsverarbeitung" in Hannover Germany. There is a big knowledge reservoir in audio and video coding, just take a look at some research topics, publications and former staff.

so what happened?

Reply #40 –
The difference between MPC and RA development is very simple. RA aimed to sound decent at low bitrates. MPC aimed to sound perfect at as low a bitrate as possible for perfection. Transparency was MPC's goal all along (as far as I can recall anyhow; the early days may have been different), not small file size.

Furthermore, it's a subband encoder, so it's easier to tune.

so what happened?

Reply #41 –
Quote
The difference between MPC and RA development is very simple. RA aimed to sound decent at low bitrates. MPC aimed to sound perfect at as low a bitrate as possible for perfection. Transparency was MPC's goal all along (as far as I can recall anyhow; the early days may have been different), not small file size.

Furthermore, it's a subband encoder, so it's easier to tune.
[a href="index.php?act=findpost&pid=253970"][{POST_SNAPBACK}][/a]


That is correct, it was the goal from the beginning. Andree was unsatisfied with the performance of MP3 at the time, so he started MPEGPlus (MP+). He decided to make it a subband encoder because it's less critical in time resolution and to avoid block switching patents. He also used some simpler quantization methods with huffman tables because of existing patents. And he actually made contracts with MP2 patent holders.

so what happened?

Reply #42 –
Quote
I remember that Roberto who is very knowledgeble posted that Microsoft hired few maybe brightest minds in industry to develop further wma and.......They come out with wma8 which is failing miserably everywhere except on their """internal"""tests where they comparing it to mp3 with some obsolete versions of old Lame like 3.88 or so.[a href="index.php?act=findpost&pid=253955"][{POST_SNAPBACK}][/a]


Actually WMA Standard was mostly developed by only one bright mind - Henrique Malvar, one of the creators of MDCT.

WMA Pro, OTOH, was developed by a bigger team of experts, starting with the famous J. Johnston, of AT&T. And that shows - WMA Pro, at its very first version, got tied to other excellent codecs at my first multiformat test.

As I see it, WMA standard was hurried out to compete with MP3 as soon as possible. That's why it's not a high quality codec.

so what happened?

Reply #43 –
Quote
Actually WMA Standard was mostly developed by only one bright mind - Henrique Malvar, one of the creators of MDCT.

WMA Pro, OTOH, was developed by a bigger team of experts, starting with the famous J. Johnston, of AT&T. And that shows - WMA Pro, at its very first version, got tied to other excellent codecs at my first multiformat test.

As I see it, WMA standard was hurried out to compete with MP3 as soon as possible. That's why it's not a high quality codec.
[a href="index.php?act=findpost&pid=254007"][{POST_SNAPBACK}][/a]

Uuups,i got it half-right 
Thanks for clarification

so what happened?

Reply #44 –
  Nice topic and discussion guys, really.. But I suggest stop blithering about what should Frank do or not. Do some work yourself, earn money, buy new HDD. It's simple, quick, efficient, and very realiable way to be able to increase quiality of your music. geeez   

Frank did his work and did it well, now it's our turn.

so what happened?

Reply #45 –
Why isn't someone else joining the work on MPC, or perhaps, taking the good from MPC and applying it to other formats?

And actually, why is LAME still being developed instead of, say, Gabriel and the others concentrating their efforts on more modern encoders like Vorbis?

so what happened?

Reply #46 –
Quote
Why isn't someone else joining the work on MPC, or perhaps, taking the good from MPC and applying it to other formats?
you seem to have too much free time, go ahead then.
Nothing but a Heartache - Since I found my Baby ;)

so what happened?

Reply #47 –
Quote
Why isn't someone else joining the work on MPC, or perhaps, taking the good from MPC and applying it to other formats?[a href="index.php?act=findpost&pid=255343"][{POST_SNAPBACK}][/a]


I think this has been said already. There are very few people that are actually qualified to work on a codec. And even less that are willing to do it for free.

Quote
And actually, why is LAME still being developed instead of, say, Gabriel and the others concentrating their efforts on more modern encoders like Vorbis?[a href="index.php?act=findpost&pid=255343"][{POST_SNAPBACK}][/a]


Well, that is simple. mp3 is still the most popular audio codec. And I'm pretty sure, that it won't change in the near future.

so what happened?

Reply #48 –
Quote
you seem to have too much free time, go ahead then.[a href="index.php?act=findpost&pid=255345"][{POST_SNAPBACK}][/a]

But not the knowledge.

Quote
I think this has been said already. There are very few people that are actually qualified to work on a codec. And even less that are willing to do it for free.[a href="index.php?act=findpost&pid=255361"][{POST_SNAPBACK}][/a]

Some people *are* already involved in this kind of development, so the qualification and willingness are there.

Quote
Well, that is simple. mp3 is still the most popular audio codec. And I'm pretty sure, that it won't change in the near future.[a href="index.php?act=findpost&pid=255361"][{POST_SNAPBACK}][/a]

MP3 is popular, but why spend more time optimizing an already optimized codec that hasn't got much room for improvment?  What big breakthroughs can you expect from MP3 nowadays?

Modern formats have much more headroom and are more exciting.  I'd still love to have transparency at 128kbit as was initially promised. ;)

so what happened?

Reply #49 –
Quote
Quote
you seem to have too much free time, go ahead then.[a href="index.php?act=findpost&pid=255345"][{POST_SNAPBACK}][/a]

But not the knowledge.
[a href="index.php?act=findpost&pid=256649"][{POST_SNAPBACK}][/a]


Knowledge can be gained. And since you obviously have the time...

Quote
Quote
I think this has been said already. There are very few people that are actually qualified to work on a codec. And even less that are willing to do it for free.[a href="index.php?act=findpost&pid=255361"][{POST_SNAPBACK}][/a]

Some people *are* already involved in this kind of development, so the qualification and willingness are there.
[a href="index.php?act=findpost&pid=256649"][{POST_SNAPBACK}][/a]


Oh, you seem to know something we don't. As far as I know... There hasn't been any real mpc development for almost 2 years. With all the respect to the LAME development team, there haven't been any significant quality changes in LAME for 3 years, since the introduction of Dibrom's presets. And Vorbis? How long have we been waiting for some development? True, there is finally some now, but only thanks to an external developer.

Quote
Modern formats have much more headroom and are more exciting.  I'd still love to have transparency at 128kbit as was initially promised.
[a href="index.php?act=findpost&pid=256649"][{POST_SNAPBACK}][/a]


Why transparency at 128kbps, when we have transparency at 64kbps nowadays?
Well, we do all believe Microsoft, don't we?