I think it's safe to say that it's not a matter of the engineers caring or not caring.
It's a political decision, made by the marketing and managment department.

It's irrational to implement Musepack. It's a dated codec and not part of any platform or strategy. It offers virtually no advantages to anyone but it's very limited niche userbase. It offers compressionrates similar or worse than industry standard MP3.

Let me ask you all. What benefit does x-manufacturer get from implementing Musepack?

A sales increase of what, 10-20 units? I'll bet that if the Neuros got Musepack support, 80% of the people begging would suddenly withdraw their interessest and claim they're waiting for x-product or x-manufacturer, because the Neuros just doesn't have the design, features, size, durability or similar to satisfy their needs.

Add in the fact that Musepack is backed by a single individual, who depend on getting hardware donations.

Benefits for x-company to support musepack:
- Possible insignificant increase of sales
- Possible media exposure of brandname which could boost sales for non-musepack interested people.
- Possible mainstream interrest in musepack, placing x-company in the lead, having musepack-ready products available.

Disadvantages:
- Costs of educating support team
- Possible costs of marketing the format (including educating and developing salesmaterial for distribution channels)
- Costs of developing and adjusting firmware
- Costs of replacing/servicing units from idiots who made an unsuccesful firmware flash
- Additional hardware requirements, additional loadtime or battery consumption.
- Costs of developing and adjusting bundled PC software
- Costs of examining and verifying legal situation
- Possible costs of acquirement patents/licenses.

If I was a business manager, I would find it very irrational to spend any time or effort on examining and implementing musepack.

I think you can see that.

If you want portable musepack players, why not start your own company?
Good luck selling them

One might ask, how did Vorbis make it soo far then. Frankly, I don't really know. But it does have some sort of strategy called Linux. It's the standard audioformat on Linux, and Linux does have a plan backed by many computer industry giants. Vorbis is also supported by Xiph.org oppossed to a sparetime individual. Vorbis produces fairly impressive compressionrates, and many people are perfectly satisfied with Vorbis at 60-90 kbit rates.
There are supposedly no legal arguments to examine, and the succes of Vorbis would provide a liberation from the mp3-license. There's probably more reasons, why iRiver and Rio decided to implement Ogg Vorbis.

sispawn, yes you made some valid points. sadly. so sadly.

-Inherintly gapless
-Excellent metadata spec
-Very fast encoding
-Extremely fast decoding (would probably be very easy on batteries and wouldn't require a powerful processor)
-Open source (ok that wouldn't matter to too many people I guess)

Also whether it is dated or not it is still a contender in the 128 - 160kbit/s bitrate range, which is, I would think, the most commonly used range for average users.

For an HD based system IMHO there would be no better compression available then Musepack at --quality 5.

I do agree with your other comments, in general, and you are probably right about it not making economic sense to support Musepack. However, unless I am seeing it the wrong way, I think that marketing divisions could claim best battery life and best sound quality, over all other portables if they chose to support Musepack. These might be significant selling features.

edit: Although the Vorbis community is probably at least 10 times the size of the Musepack community (note: there are sizeable Musepack communities outside of HA.org, I can't link to them because they all, I think all of them, anyway, relate to Musepack filesharing) it is, I believe, significantly harder, and thus more expensive, to impliment Vorbis then MPC would be. Yet Vorbis has gained some considerable hardware support. Couldn't MPC at least get one player (outside of China)???

edit 2: typo

And on that note wouldn't it be better if the import utilities supplied with the hardware players, which allow for import from CD->Hardware player used Musepak. Like was said just above, lower battery consumption, better quality. It seems kind of like the impossible dream.

Great input folks.

Wrong. Musepack achieves transparency at bitrates ~20% lower than mp3, aac, or vorbis.

Rule 8 at you, foo!

oh come on, lots of listening tests on this board support this, as --standard is considered transparent and is typically 20% smaller than LAME --preset standard, which is also considered transparent (and I doubt you require a listening test to back that up, do you?)

You must be blind. He wasn't talking about MP3 only. He was talking about Vorbis and AAC too.

And my former listening test at 128kbps proves that, actually, Musepack is on par with AAC at that bitrate, and Vorbis, although being worse, got much improved by independent tuning, meanwhile Musepack stalls.

You should really consider reading with attention before posting.

Robert, has anyone told you that you're a party pooper?

Detective John Kimble: It's not a tumor!
(Sorry couldn't resist.)

Eitherway some good stuff so far. I hope this thread is far from dead, but thanks for all the great input up till now.

Rock on HA!

Why? Just because I don't stand people deliberately saying bullshit in order to attempt to promote their format of choice?

Rmajorim,

If you want to be funny, musepack, as seen in this thread could be considered transparent with some samples whereas aac/vorbis/mp3 are easily abxable. So we can't even mesure % as others codecs don't obtain transparency.

Talking about your former test is useless as 128kbps has never obtained transparency yet for any codecs, and so is irrelevant in this subject.

Thanks you, and let's get back on topic, which is hardware decoding for musepack.

I guess the truth hurts for some.

Honestly, with the kind of quality AAC and Vorbis archieve these days, I don't see any reason why the manufacturers would choose MPC over those two. The adoption rate of AAC and Vorbis is already slow enough as it is.

So what? Because there are some samples where musepack performs well and other encoders don't doesn't necessarily mean Musepack reaches transparency first.



Then, I suppose you conduced a listening test where codecs reach transparency? Oh, no? How foolish of you! In this case, my test is the best reference we have, no matter how much you consider it useless just because it didn't confirm your dreams about your codec being better than everything else



No worries, I'll buy some boxes of Midol for Lefungus.

oh yeah? well mine is waaay bigger than that!

eheh...

stop bickering, and get back on topic, which is providing a hardware musepack decoder! pronto!

What i write here is my *opinion*, i don't need rule 8 thrown at me.

AAC and Ogg Vorbis have both implemented for reasons, for AAC that would be apple who put it in their ipod. Ogg Vorbis is geared towards <128 kbps bitrates(it's not bad at >128, just good at <128) and since the majority people like small files, then thats likely to get popular. I remember someone saying(don't know the name anymore) that ogg q0 wasn't low enough, people should use q-1.

Musepack is focussed towards >128 Kbps, to provide extremely good quality at high bit rates. The majority of people isn't interrested in transparancy, they're interested in files that are as small as possible. Only small group of people want a format like mpc and unfortunately thats not enough for companies to build support into their players.

I don't mind, because i don't need a portable, so i'll stick with mpc.

Madman2003.

Fair points. However, there are users like me who have their entire music collection encoded in the Best lossy codec ever (MPC of cause ) and the last thing we want to do is transcode all our music to a lower quality format just for compatibility. I'd rather convert them back to wav and get it playing them!

I've compared many, many encodes in all three codecs at the bitrates their authors claim achieve transparency (mpc at --standard, lame -aps, vorbis -q6 and nero aac at the transparent setting). So I've not only got the collected testing experience of HA backing up this statement but also the opinions of the codec authors themselves. How is this a violation of the ABX rule?

Sorry to jump up and ruin your "MPC vs. Other codecs" fight.
At the cost of being labeled as the annoying lawyer of the day ;), i want to point you gentlemen to the Terms of Service, article 6:

"6. You must stay on topic when posting a new thread or posting to a pre-existing thread"

So please take this off topic fight to one of the other countless similar threads or get it on via private message.
Thank you for the attention.

well, Microsoft claims WMA (non-pro) is CD-quality (read: transparent) at 128kbps...

Can you point me to a thread on HA that supports your statement (transparency of MPC at 20% lower bitrate than other codecs)?

You cannot use the claims of the authors as a proof of your statement. Lots of those claims are just marketing BS. Only listening test results can be accepted.

Please note that I'm not trying to bash MPC, I only want to warn about the dangers of making unsubstantiated claims on HA.

And of course nerochiaro and xmixahlx are right: these posts are off-topic and should be split into another thread. I'll notify the moderators.

This claim is hardly unsubstantiated. I'm basing it on the results of encoding my ~ 1200 cd collection with these codecs using the HA recommended transparent settings.

I like the Immortal pic, by the way.

Its not just the sound quality. Fast encoding/decoding speed , low cpu usage, gapless, replaygain, transparent at the default setting (q5), smaller files (170k), mid bitrates are still very competitive with other codecs (q4). Vorbis and itunes newbies often don't know what quality setting / bitrates to choose. Many mp3 users end up with voodoo switches that can ruin quality. Worse thing for mpc users is going for q7-10 or not using --xlevel.

Type mppenc, oggenc, lame and you see the difference. To me it is obvious that mpc was built for quality and nothing more. It seems that apart from political reasons, mpc should be the perfect codec for portables.