[USELESS] offtopic split, from "A new PC for Frank" thread Options

[mdmuir]

IMHO, elements of this thread have got to be some of the saddest things I have seen since being a member of HA. This appears to be a desperate attempt to bring burnt offerings to the god of mpc! I mean, shall we have a little perspective here? Its one of many audio compression formats: it is not a universal cure for cancer or aids, nor is it the key to world peace and harmony. When is it going to be the time to let go of your
childhood teddy bear and move on?

[atici]

@mdmuir: It's the most amazing lossy codec and it adds a lot to my pleasure (and also many others) when it comes to music reproduction. We could as well give up listening to music based on the same argument After all there're lots of other things to enjoy. Or we could give up doing pure mathematics too. The reason why I feel like helping MPC is the same I invest a lot of time in HA, because eventually I get more satisfaction out of my music.

I respect Frank's efforts in any case. I don't care what he'd do with my help. It's such an unfair world where many ignorant people make a lot of easy money. I am sure how rich Frank could be, he deserves a lot more. Because he pushes forward the field of lossy compression, probably much more so than many other big corporations with lots of resources -- all by himself.

This post has been edited by atici: Feb 8 2004, 04:20

[mpcfiend]

Some of us have already committed to the format, and do not want to even consider moving to another.

[rjamorim]

I would ask him straight-out about his intentions, or lack thereof, to develop Musepack.

Why don't you just leave Frank alone?

Damn, everyone here knows I dislike Frank a lot - I am the reason he left this forum in the first place - but even I am starting to have pity on him :B

[rjamorim]

Some of us have already committed to the format, and do not want to even consider moving to another.

Erm... closed-mindedness?

[westgroveg]

I think adding the Klemm, Busch psymodel to FAAC is still a good idea. I have no idea how easy / difficult this maybe.

[Moguta]

I would ask him straight-out about his intentions, or lack thereof, to develop Musepack.

Why don't you just leave Frank alone?

Because if he has no real intent to develop Musepack, then he should go ahead and say so. THEN there wouldn't be anyone bugging him about when he plans to upgrade Musepack.

This post has been edited by Moguta: Feb 8 2004, 04:36

[rjamorim]

THEN there wouldn't be anyone bugging him about when he plans to upgrade Musepack.

You shouldn't bug him at all to start with. He owes you (myoozepack users) nothing, so what right do you think you have of bitching at him?

[Moguta]

Of course he owes us nothing. He has already had a major hand in developing & improving this format to which we should owe him something.

Did I miss something in the past? Such as Mr. Klemm officially stating he had ended his work on MPC? Because it seems a reasonable - to me - that if you leave a project in midair w/o telling anyone of the status, people will ask you about it. Just plain sense.

This post has been edited by Moguta: Feb 8 2004, 05:02

[rjamorim]

Just plain sense.

Common sense is also noticing he isn't answering about project status, and then concluding he actually doesn't want to talk about it. Why insist then?

[sld]

The thing is, one does not leave stuff in limbo. Like...


that.



Well the least Frank can do is to release the sources. I am sure there are a few developers who are more than willing to bring SV8 to completion, let alone SV7.5 . If he feels that the sources belong to him alone, then fine, just state it (I bear no responsibility for the resultant backlash though, if it occurs).


By the way, it is not, erm, wholesome for a person, in his silence, to cause others to quarrel over him.

Can we just bring this episode to an end? It is sad to see Musepack, the current best lossy codec, unnecessarily encumbered by communication issues.

This post has been edited by sld: Feb 8 2004, 05:54

[indybrett]

As good as MPC is, it isn't good enough to waste all this effort on.

It's like trying to convince that girl who left you to come back. You think that if she only knew how much you loved her, that she would realize what a big mistake she is making and come back to you.

Years later, you still call her, and she still hangs up on you.

Better to put all of that effort into something with a future...

[SenatR The Last]

What is going on? You started again just to talk about mpc wow and mpc down. It is asked absolutely clean - would you give some bucks to buy a new PC for Frank. Wanna say yes- say yes, no - no. Still there is no useful seed in current conversation.

And yes, i will give up to 20$ for that deal.

[indybrett]

Uhhhh, what??

[rjamorim]

And yes, i will give up to 20$ for that deal.

See? You guys are already talking about a "deal"

I can only imagine Frank will be screwed if he accepts the "deal"

[rutra80]

What's this topic about? If it's about "how much & how many people would pay" lets just make a poll instead of another soap-opera...

[Canar]

IMHO, elements of this thread have got to be some of the saddest things I have seen since being a member of HA.  This appears to be a desperate attempt to bring burnt offerings to the god of mpc! I mean, shall we have a little perspective here? Its one of many audio compression formats: it is not a universal cure for cancer or aids, nor is it the key to world peace and harmony. When is it going to be the time to let go of your
childhood teddy bear and move on?  

If audio compression is so meaningless to you, why stick around?

There's some level of meaning to it; it's something that I enjoy greatly. It's technically intriguing, as well as pragmatically satisfying in knowing that I'm getting a great compression ratio yet great quality too.

What do you suggest we move on from this "childhood teddy bear" to? AAC? Vorbis? Neither is tuned well enough to compete with Musepack at the higher end, as far as I'm aware of. AAC is being hawked as a mid-bitrate solution (ie. iTunes), and there's been no official high-end tuning of Vorbis since Garf left us with the GT3 series. I'm aware there is some third-party tuning going on right now, but we've yet to see anything substantial come from that direction.

If some people who greatly appreciate the work Frank and co. have done on Musepack want to help him get a new computer to assist in development, let them do so.

It's no teddy bear; it's the highest quality lossy compression format out there. The key to world peace and harmony is to get everybody to agree on a common set of morals and ethics. The cure for cancer is being worked on by biologists, not computer scientists and hobbyists. And until there's a better format, why should we need to move on?

There are a thousand things we could all do that would be more "productive" than discussing audio compression. But that doesn't mean that the discussion is wrong.

[rjamorim]

If audio compression is so meaningless to you, why stick around?

Duh. Where the heck did you see him saying it is meaningless? Don't try to put thing he didn't say in his mouth, you'll only make a fool out of yourself.

He was just complaining people are taking it too damn seriously here. The mask seem to have fit pretty well on you, actually.

What do you suggest we move on from this "childhood teddy bear" to? AAC? Vorbis? Neither is tuned well enough to compete with Musepack at the higher end, as far as I'm aware of.

Gimme ABX.

This post has been edited by rjamorim: Feb 8 2004, 07:09

[rinseaid]

It's not as if it's not commonly accepted around here that MPC is the most transparent lossy encoder at high bitrates. Let's not be facetious in the quest of winning an argument or having the last word. IMO, this is getting way out of hand and offtopic. If people want to donate, just post saying you want to donate. If not, just leave the thread alone and allow those truly interested to have a civilized discussion. FWIW, I would be willing to donate money if Frank truly thought this was a good gesture and I would have no expectations of Frank ever finishing SV8. I believe that I personally already owe enough to him for what he's done.

rjamorim: This should satisfy your request.

This post has been edited by rinseaid: Feb 8 2004, 08:56

[Canar]

Duh. Where the heck did you see him saying it is meaningless? Don't try to put thing he didn't say in his mouth, you'll only make a fool out of yourself.

He was just complaining people are taking it too damn seriously here. The mask seem to have fit pretty well on you, actually.

So, instead of letting people act out of generosity (or even just act in a generous manner) we should berate them for feeling strongly about something. mdmuir was criticizing the others ("elements of this thread have got to be some of the saddest things I have seen") for doing this. That seems more sad to me than enjoying an audio format.

I'd like to see work continue on SV8, even just so that it can reach a point where it can be successfully open-sourced and Frank can do what he likes after that. If a contribution of cash or parts can assist in reaching that end, even in some utterly trivial inconsequential manner, I'm for it. If that's taking it too seriously, then so be it.

The implication that he made was that the fact that we cared to see MPC reach SV8 had little relevance. But, taking a step back, how meaningful is audio compression anyhow? Attacking audio compression based on its meaningfulness is like attacking an old 8-bit Nintendo game based on its graphics quality; you're attacking something that isn't intrinsic to the object anyhow.

Gimme ABX.

Erm... Which codec has more catastrophic failures? It's taken more as a given rather than a proven that Musepack is better, namely because so much has been encoded with it, so many people have tested it, and it's fallen short on so few files in so few instances. I know you favour AAC, and I know you hate to just let givens be, but Musepack is perfectly transparent to me, which is something I couldn't say about any of the other formats I've tried (which, at present, stand at Vorbis (~Q4/5), VQF, and (non-aps) MP3). A single ABX will do nothing to prove things one way or the other and you know that. It'd take a lot more work than one man can do to "prove" this. Nice attempt at a troll, anyhow.

Meh... I expect this thread doesn't have much more time to live anyhow. Seems to be attracting strong emotion from both sides.

This post has been edited by Canar: Feb 8 2004, 08:09

[ChristianHJW]

What's this topic about? If it's about "how much & how many people would pay" lets just make a poll instead of another soap-opera...

Excellent reply, couldnt have said it better. I really wonder why so many AAC and Vorbis users feel inclined to post here in this thread, as they werent asked at all to give their opinion on what i am trying to do at all ..... maybe they are panicking that SV8 might really become even better than where SV7 is now already ??

Dont worry guys, Frank's SV8 block diagram ( i forwarded it to the mpc-devel ML once ) is showing that one of the major backbones of it is the 'pluggable' psy model. And yes, before you ask, it should be possible to tune both Vorbis and FAAC with it, Frank has confirmed this already several times. He even admitted that, with his psy model, both formats should perform better than MPC could do.

So ..... going to put up a poll now, this is leading nowhere ......

[mpcfiend]

Erm... closed-mindedness?

No, I have a collection in Musepack that I've already committed to. I have some hard-to-rerip (ahem...) albums that I would like to keep. I really don't want to do a conversion, and I'm happy with Musepack the way it is. But if development looks dead, there won't be any plugins released for e.g. Noatun (The closest player I can get to fb2k on Linux) or other media players.

Long story short, I'm lazy; I don't know how to code; I need to rely on the fruits of someone else's labours in order to play my Musepack files. Without new development on the front, the format dies.

[mpcfiend]

Meh... I expect this thread doesn't have much more time to live anyhow. Seems to be attracting strong emotion from both sides.

Which is funny, since the 'other side' has nothing to input into this. Lets break it down, shall we?

1) Christian floated the idea of trying to 'entice' Frank into continuing work on the format.

2) WTF?

C'mon, why the reaction?

[ScorLibran]

Gimme ABX.

Erm... Which codec has more catastrophic failures? It's taken more as a given rather than a proven that Musepack is better, namely because so much has been encoded with it, so many people have tested it, and it's fallen short on so few files in so few instances. I know you favour AAC, and I know you hate to just let givens be, but Musepack is perfectly transparent to me, which is something I couldn't say about any of the other formats I've tried (which, at present, stand at Vorbis (~Q4/5), VQF, and (non-aps) MP3). A single ABX will do nothing to prove things one way or the other and you know that. It'd take a lot more work than one man can do to "prove" this. Nice attempt at a troll, anyhow.

I think what Roberto is referring to with "gimme ABX" is the threshold of perceptual transparency rather than the incidence of "catastrophic failures".

And all psychoacoustic formats "major" enough to have their own forum at HA can provide perceptual transparency at a point below their maximum. A general statement that MPC is better is not a safe one, in my opinion. For instance, in last year's 128kbps Extension Test, MPC statistically tied AAC, Vorbis and WMA Pro. And I haven't seen tests that show MPC to be "better" at higher rates, though there may be a commonly accepted view at HA that MPC is the better performer at mid-high bitrates.

But another consideration is that the transparency threshold differs across different individual audio samples (even within the same "type" of music), and with different people, for each codec.

That MPC is "better tuned" may be technically valid, but I haven't seen significant benefits in perceived sound quality that this provides over the other "world class" psychoacoustic encoding formats, to be honest. The most I've seen of MPC testing with documented evidence is a tie with 3 other formats in a public listening test, and specific sound quality problems discovered by guruboolez.

I agree that there should be more tests to point out the benefits and limitations of each format, but until we have those results, I don't think anyone can say "MPC is better". More easily supported arguments would be whether the format has hardware support, active development, and significantly increasing popularity. MPC falls down on these points.

So, accordingly, I vote "no" on buying Frank Klemm a new computer.

[Pamel]

For instance, in last year's 128kbps Extension Test, MPC statistically tied AAC, Vorbis and WMA Pro.

Yes, it scored the same. But even the author admits the other formats are better, with MPC excelling simply from its psy model. I think that the question is whether one of these formats would produce better results using a pluggable version of MPC's psy model.