I just noticed in the polls that more people use mpc than mp3!

Bearing in mind that I know nothing about mpc, could anyone explain the benefits over mp3?

I use alt-preset extreme... does anyone recommend I upgrade to mpc?

Please search. Some great threads have already covered this.
So MPC at it Greatest Sound Quality..., ...sounds better than .OGG at it's Best?
320 kps MP3 or --quality 7 --xlevel MPC ?, Chosing betw. MP3 & MPC
Why is MPC perceived to be the best?

That poll only applies to Hydrogenaudio. It makes no sense outside this forum (where most people, by far, use MP3, and in second place comes WMA)

The summary is: MPC is better than MP3 in quality, but worse in support.

Thanks, guys.

It sounds amazing and what really surprises me is the speed of compression, really fast. Not supported very well though but hey I only use Winamp and there is a plugin for that, so it's fine for me.

And you had to bump a two years old topic just to say that... OK...

I guess that creating a new thread was not enough
EDIT: remove last sentence.

EDIT2: for memory, this test may be useful...

Damned if you do, damned if you don't. It's hard being a newbie to the whole "forum" thing.

I've seen newbies crucified for making a new thread when a two year old thread already covered the topic.…

1) I don't recommend it too much work, very tiny benefit
2) yes, at the given bitrate mpc reach better quality than mp3
3) if you need better quality than mp3, stick with ogg instead, I can't tell if it's better than mpc, but at least much more supported and there is develompent still going into it

Not even in this forum anymore, I would say.

Too bad for MPC

I still have everything in mpc and continue to encode to it.

it could be argued that diferent samples were used ... even diferent people probably submited results ... anyway ....

Maybe it's time for a new lossy encoder poll. Can this be arranged?

EDIT: Scratch that. The one from Feb 2005 is still at the top of the polls list and apparently being added to.

Yea, but it's over a year old...with much more development across several formats since that time. I don't know how one gets permission to create new polls.

Re: hybridfan being a newb, check his registration date: Joined: 11-July 03


Why are you still encoding to it? What perceptual benefit can there possibly still be? I guess some people are just set in their ways. I had some MPC backups from 2003/4 but reencoded what I still have lossless/original copies of to Vorbis this year.

Finally, has anyone compared compression times with the latest Lancer release versus the latest MPC? Which is faster...it'd be nice to put this Hybridfan hype to rest.

What a load of BS Do you actually know what you are talking about? If MPC was transparent back then, it is still now. There's not so much quality improvement in other lossy encoders really. MPC has already been very well tuned and there is not much room for improvement. Most of what SV8 was planned to bring is not related to quality.

There're many benefits to MPC, one of them is that it is a subband codec therefore transcoding to MP3/Vorbis results in less audible loss.

IMO only Vorbis now could come close to the level of excellence MPC has been. And if Vorbis is doing so well recently, it's because of Aoyumi. I don't remember xiph really contributing to the development much otherwise for a while. Similarly if MPC was doing very well in the past, it was because of Klemm. Now that there's no bigshot, MPC development seems to be stalled. But it already does an amazing job and I personally do not care & heavily use MPC. I trust MPC at high bitrates more and I have plenty of space (though not as much as lossless) on my dedicated music system. When I need portability I simply transcode and transfer, my DAP does not play Vorbis either. It all depends on what your personal requirements are. Otherwise the future is lossless anyway.

You know, I get the impression you've really got a chip on your shoulder. Here's some more BS for you in re: to MPC:

1.) You might as well go lossless if you're going to be doing transcoding, tho' I'll give you this.
2.) Sequencing is a sluggish pain.
3.) Hardware support has never been so hot for MPC, tho' Rockbox has gone a long ways in rectifying this.

Vorbis, in spite of whose done the tuning, is currently a smarter choice for lossy, especially if you want a combination of smaller file sizes and greater hardware support (personal DAPs being the key here).

PS. What you quoted was in regards to the POLL being a year old...maybe that's why you mistakenly called me on BS. I think that's long enough with enough development commited to have another lossy poll....and I think lossy is here to stay for some time, at least 9 years if we go by your year-old prediction.

Or maybe you just can't bear to see MPC lose the Hydrogenaudio popularity contest.

How do you know that? Do you know that I have 400GB hard drive full 70% of which is MPC? Is it worth to invest in >1TB of space (>2TB with backups) for no audible difference? I'd rather invest in speakers.

What do you mean?

And for Vorbis it's too hot? Only Samsung and iAudio really supports it. iRiver has half baked support. It's very likely that you'd end up transcoding if your archive is in Vorbis unless you don't mind seriously limiting your DAP choices. I think if portability is important I'd rather use MP3, since it's supported everywhere and (guess what) LAME is doing a great job too.

I give you that Vorbis is also a good choice. But one can't just dismiss MPC so easily.

Anyway let's focus on what the original poster is asking. If you are interested in playback on a home audio system then MPC would probably be a better choice than MP3, as it'll be higher in quality per bitrate. Also Ape2 tags of MPC are more flexible. However for all purposes of portability you're better off with MP3.

Apologies, I meant "seeking" instead of "sequencing." That's what I get for posting in haste.

Is there any evidence to backup this idea?
I tried once on a Short re-encoding blind listening test and MPC wasn't superior to Vorbis on the decoding side. Of course, I didn't tested enough samples to draw reliable conclusion, but I didn't discover any elaborated test proving that subband encoders are better for transcoding purpose. Do you have a link?


That's your opinion, but what about real experience - a listening test?
Again, the last I did in the specific area of Musepack's excellency ended on aoTuV superiority. And again, the results are personal and limited to non-killer-and-classical music - but again for a last time I still have problems to find a more elaborate listening test showing MPC benefits over a modern encoder. Do you have a link to backup your claims?

You know what's funny, people keep talking lossless while operating 128k tets.

I think Doom9 boards has nice a feature, that's probably worth adopting: sending threads with 'better than' and 'best' in the title straight to hell

Yes so what if I did, I have only started using it and impressed with the results, maybe because it isn't supported that well it should be bumped.

IMHO that's rather a limitation of doom9 not offering a proper valid method to backup such claims. What's the point of discussing different codecs in the view of an end user if no ranking in regard of previously defined aspects can be made?

In my opinion, developers should focus on improving codecs like AAC, Ogg Vorbis or even MP3 (although MP3 is slowly reaching its maximum abilities).
Also, what's the point in receiving more support when its development almost ceased?

With such nice feature, MPC wouldn't be adopted by more than 3 guy on the world. It's precisely because MPC was said to be so better to all lossy formats that people started to use something with few software support (in 2001-2002), no hardware one, obscure patent situation, scrubby seeking... I'm one of these people who followed all mpc's gurus in the early days.

Moreover, HA was founded during the alt-preset maturation. It's a task which supposes that reliable opinions about audio quality is possible. As a consequence, it would be illogical to deny to members the right to ask about quality comparison. Don't forget that sticky were made to inform people what precisely the... best settings are!

Well, ok. Although recommended and best aren't the same.

There's helluva difference between 'I really wanna try this codec and wanna know what the good settings for this particular purpose would be' and 'What's the best codec of them all' soon falling down to 'I heard yer prefered codec is better/bigger/longer than mine, well, that's not true...', the latter is like invitation to the bashing/flamewars/fanboyism/hurt feelings/namescalling and what not.

So if you're into streaming over ISDN or encoding audio for DVD backup, you probably exclude certain codecs from the consideration, and vice versa.

In another words I can't quite remember something starting like this and leading to something constructive.

Well, if you would let me, i will add some comments...
Usually i use Wavpack to make lossless copies of discs...but when it comes to my PC audio or Rockboxed Ipod i stick to lossy codecs...
Here i have two options...

1. MPC, -8.0, appr. 280-300kbps
yeah, i really like it for jazz, classics and some other kinds of deep and sensitive music (don't ask me how i choose, it is intuition, i guess)...the panaroma is vivid and reach and i simply it adore...
I even turn a blind eye to the fact it has no seeking...i simply love the way it sounds...

2. Ogg, -q6 (aoyumi), appr. 192kbps
I use this codec based on some conclusions kindly made by guruboolez who admitted that this preset of this codec has shown tremendous results in his tests...so, if it comes to pop or dance music, i usually choose this codec...

I didn't check Ogg at higher bitrates and I don't feel i need it since i am pretty much content with MPC...however, MPC is not really good at low bitrates, so i choose Ogg -q6...

That is basically my point...

True, but recommendation (at least here) are based on the quality criterion (i.e. we're recommending x over y because x is clearly the best). --alt-preset were recommended over --r3mix (ultra-popular) only because it was better. LAME was recommended over Fhg, Blade, Xing for the same reason. The entire forum is haunted by the same question: what's the best setting? what's the best encoder? what's the best format? The purpose of TOS#8 is precisely a safeguard against trolling or complete anarchy due to the subjective argument. It's because ABX tests are possible and reliable that people could perfectly:
- claim that x is better than y
- help to improve any encoder by a proper and rational feedback
- ask to other people for trustable advices.

ABX tests are apparently much harder to set on video (lack of software?) and without such proper methodology calls for advice could quickly turn into massive trolling and bashfest. Hence Doom's TOS#12 I suppose.


True, but again this board has a protection against this risk. And it worked pretty well.


That's obvious. But in the present case, someone was very precise: LAME --preset extreme or Musepack? He didn't asked for a general position about MP3 vs MPC.


As I said, MPC "success" is based on similar threads... A lot of people are (or were) convinced that MPC --insane or --braindead is something necessary. No need to be a genius to understand that most of these people could safely use --standard as well or even MP3 at 130 kbps to get the same perceptual quality. Just take a look on the few results sent on various listening tests or read Seed's complaints about the total lack of feedback of the MPC audiophile community The existence of a small MPC community is certainly not based on personal and double-blind experience of its superiority at 180 kbps, but rather on blind-trust and self-persuasion of most users. If I'm wrong, I wonder where are located all these testers. Not on HA for sure

The Doom9 ban isn't on discussion, it's on people who say "this is better than that" without backup or where it's better (support? speed? top quality? low-bitrate quality? interface?), or similarly, people who come in and ask "what's the best x" without specifying exactly what they're looking for. It leads to a lot of flamewars and bullshit, as you might imagine, but the rule is no different than calling posters here on bullshit claims. *shrug* (Backing up what ak said.)

Personally, I love mpc, because it's a great format to transcode into my beloved vorbis with minimal degredation.

We are still waiting for proofs on that.

lol, great conciliatory answer.

You're a new member according to your joining date (but perhaps you're reading HA for a long time). HA.org is four years old, and several (if not most) debates on the what is best? subject were nicely conduced. There's a recycle bin and vigilant moderators to limit excesses such as trolling or blatant zealotry. Most people are debating on a rational basis, basing most of their claims on ABX results. There's no need to forbid such debates as long as a valid method for valid statements exists.

In the early day, MPC was always showed by their supporter as the ultimate audiophile solution for lossy encodings, beating AAC, LAME and Vorbis. You can still see some persistence of this fading audiophile aura on small and less informed boards on the web. Now that MPC is quite dead, having no advantages at all [it doesn't mean that it suddenly becomes bad], I find quite amusing that a MPC developer is suggesting on this board that opinions or calls for opinions about audio quality should be avoided. It reminds me another MPC developer who suddently became allergic to ABX results and TOS#8 when it started to play against his format.

I've never had any interest in MPC, personally... I'm sure it's better than MP3 in bitrate/quality and transcoding quality, but the total lack of support across the board and lack of development is a little more than a big turnoff.

The only use I've ever had for MPC is that some metal/punk on P2P networks is/was encoded in it, which actually rather surprised me. I've seen more MPC files than ogg or even WMA when looking on the e-donkey network (via eMule).

haha, I've noticed this same phenomenon. It seems some metal heads are closet audiophiles.

I've seen the same phenomenon. I remember that in some in discussions people claimed that mpc can have the same quality that mp3, but using lesser bitrates. My metal encoded songs usually go up to 230kbps in standard, so i think they started using mpc for that.

...or maybe as they like to show themselves as underground ones, why not use an "underground" format?

Probably the folks from MPC-Donkey. A lot of people there are metal heads.

weird...
April jokes ?

I knew (and posted) during the prediscussion of the latest 128k multiformat test, that cross-test-references are necessary and interesting in general,
and that without this, sooner or later somebody will come and connect the old tests (eg. with mpc versions) and the new test (eg. without mpc), to show "somewhat", haha, a comparison between mpc measured/ranked in old tests to compare the old "value" with formats in new tests and their new "values".
You made my day !


MPC,
that is great usage in Windows & Linux,
no (serious) flaws in higher qualities like 7 or 8,
wonderful small-sized high quality backup solution (on DVD+R) for my Lossless music.
Other formats are still struggling to reach the quality.

For portable factory made hardware: MP3 Lame -V5 -vbr-new the way I'm going. That's universal playable on devices.



well, the story of software & their developers (not especially mpc related) is weirdo sometimes, but if the result is fine..., they deserve respect, even if it isn't 100% perfect, that's life, to have tried and have made something, that's important

Lack of testers in the higher bitrate range:
- headphone testing does not reveal the complete truth about formats
- no fun to participate due to avoid being involved later into personal conflicts, as formats' artefacts could be taken personally by the dev..., HA is too small...
- eg. the latest 130k multiformat test with Lame -V5 participating, where the modern codecs succeeded well, all... -> Despite the easy spotting of lack of highs/attacks with higher frequencies with -V5 -vbr-new via good speakers in bigger living-room, and me is meanwhile over 30 with imo worse listening abilities than in my 17-28 years..., Lame -V5 is good, if you don't make A/B-X listening at good HiFi.
As I don't invest much of my freetime in virtual internet fora, but follow eg. HA regularly, I rate some core HA group (with some nice exceptions) as relative small group of eager technicians, some people who know what they are talking about, and a group, who simply rewrite again and again, what others have said before, who want to belong to the core, to be felt accepted by the group. Some sort of group/sect behaviour with own dynamics.
Due to group dynamics and relations, the groups' opinions tend sometimes in 1 direction, sometimes to other directions with the goal to make "policy", to diss formats/sub-groups of people, sometimes arised by personal conflicts.

The solution is simple,
go back to one of HA principles, listen yourself, judge yourself.

I have to say I'm a little guilty of that, but that's not saying I started encoding to MPC blindly, I did the required research, I understood the shortcomings about the nonexistent HW compatibility (didn't have an mp3 player at the time), and I seldom seek through my songs so I was the perfect candidate. I did manage to convert some freinds over to the MusePack camp before I switched completeley to mp3 when I got a DAP and an mp3 capable stereo for my car.

Ahh the memories... fast encoding, simple and easy tagging, and the warm fuzzy feeling I got when I felt like part of an (underground) audio-minority!

Prove it.



Don't generalize over anything should be another principle.

The only fellows with golden ears (most notably, Guru) who have done such tests have proven otherwise. And what would be the point for MPC inclusion in the last 128 kbps test? If I'm not mistaken, there had been no tuning for it in that range since the last public test. I don't think anyone is still arguing against the superiority of MPC over MP3 (tho' it would be interesting to know how well the latest LAME release measured up against it at 128 kbps). That just seems to be the claim of someone who is paranoid about the fading glory of their favorite format...

really?

Keep in mind ~128kbps was/is not MPC's strong point.

if you are that close to a perfect 5, i wonder how much can you improve ...

LOL, got me there

No, serious? Does it imply that MPC is a great format because it doesn't have serious troubles at ~250 kbps? AFAIK, several encoders are in the same position.


Could be said for other formats/encoders, with additional feature as:
• no curruption issue (one error within MPC stream, and all following parts are lost; other formats as Vorbis, MP3, MP4/AAC, the resync is still possible).
• better seeking


Wake up: we're not in 2001 anymore.



True: headphones are like microscop. They make subtle distortions audible whereas speakers erase them. Without headphone, there's no need anymore for ultra-robust encoders. Even inferior implemetation of MP3 are transparent at this bitrate.


I've submitted problem samples and ABX tests to several developers:
- LAME
- iTunes AAC
- Nero AAc
- FAAC
- MPC
- WavPack Lossy
- Vorbis (Garf, Aouymi, QuantumKnot)

and the conflict only occurs with one "development" team.
This problem only exist with MPC aka people who can't improve the psymodel by themselves but who can deny all validity to any annoying listening test - and they're champion for this kind of trolling.

A while back i tried lame -V 2 and at on the few tracks i tried i didn't hear any obvious flaws. I've also tested high bitrate ogg vorbis (an older aotuv iirc) in the past and it was also fine. The only problem i see is trackrecord, musepack has been at this quality for level for quite some time, so a lot of files are encoded with these "high level" encoders. This leads to the problem with something like mp3, it's the people who use it, if 95% of the world used eac(secure mode, test&copy) or similar and encoded to lame -V 2 (-vbrnew) or better than there would not be a problem. The people who use mpc are generally a little more carefull about what they do. I still wouldn't trust mp3 for backups, might trust mpc(because i have more experience with that), but i prefer flac, but mp3 would not be bad for portable transcoding.

There are also speaker systems that are like microscopes. Not mine, yet.
One different thing:
I had some low frequency clicks i wasn´t able to hear on my HD-590s that jumped onto me on my system in the living room. These were artifacts you get on read errors and produce clicks. I tried to clean them and on recordings that were perfectly clean on my headphones the lower frequency artifacts were not subtile on my main speakers.
After knowing where they are i barely heard them on my headphones but they were there. So some things even come easier on speakers for me.

Edit: I once had such a sample, beneeth others, in Massive Attacks - Angel. I don´t have it around so i can´t give the exact position.

Sometimes I wonder...

... if a codec (name your favorite here, I have mine) already performs transparently at a lower bitrate...

... then why encode at a higher bitrate?

That provides an extra layer of psychologic protection for those who worry excessively about transparency and problem samples wreaking havoc on said transparency.

The vast majority of my 45000+ track collection in foobar is mpc. In the last 3 years or so it's proven *to me* to be excellent - I can positively ABX maybe 2 tunes at xtreme (the aerosol noise in 'Housewife' by Dr Dre comes to mind) but i've never had to worry about the quality - it just 'works'.

My desktop at home is Windows, but the media pc in the lounge runs Linux. When installing FC5 onto the latter with mythtv (which can rip and encode to anything but with a nice TV style gui) I was dithering over whether to switch codecs for my new rips. I was looking at AAC and Vorbis but the different compiles and variants put me off.

Give me a recent Musepack encoder and my encodes sound good. That's why I still use it. I'm not saying it's the best (though it's certainly up there for my needs) but it's the best and easiest option I have.

I would love to see a SV8, followed by a proper listening test of all the modern codecs.