Hey. I tried MPC w/ EAC this morning compressing Nas - Made You Look. I'm not certain this is the future format yet. As far as compression size i believe that this is excellent for the bitrates it uses. But, i think that the format will never cross over to become popular, maybe just to audiophiles. It took 5 seconds to compress a mpc so i'm not convinced it took deep compression methods. and the bitrates arent as sparatic as lame aps they just kinda jump up and down one frame. I believe that for the mb it uses its excellent but, LAME still outnumbers it because it is further in development and is a better less-bulky format (you need a plugin to play mpc's for WINAMP! WINAMP is beast if you have to spoonfeed it the format is whack).

oh sorry and i forgot to mention i tried it with quality 7 xlevel

This is wrong. MPC is heavily optimized. It even uses some Assembler routines.



This is arguable. But yes, the support for MPC is not that widespread yet.

Hi,

This might be difficult to realize, but MPC is not an "accelerated" codec with compromises... it's a very advanced codec which also happens to be very fast

So, don't get fooled by the speed: the calculations are done very carefully - and give in general, more accurate results than any other codec. The psymodel of MPC is among the very best.

In other words, in this case, the speed is not a sign of careless sound processing.. it's a sign of clean design and very good optimization.

Hi. See how well Winamp can play MP3s after you delete that little in_mp3.dll file. Sure, you can argue that it's bundled with the player, but installing a plug-in is painless, provided the author bundles it in an installer.

Dump WinAmp. Use FooBar2k. Be happy.

BTW: I thought I saw a little trolly!

Are you using the 0.98 plugin for winamp2 ?
(Some old versions didn't update bitrate display every frame.)

mpc bitrate can be anything from 3.5 to over 1000 kpbs, changing from frame to frame. You can use the mpcbits program from Case's page:


It was mentioned recently.

that is correct on april the 1st.

Your not sure if it takes deep compression methods, but the size is excellent for the bitrates it uses? It is far from bulky, it encodes CD audio at higher bitrates, that's it and that's all. It also doesn't use bs compression methods such as cbr and abr, which further reduces it's bulkiness. Winamp really has nothing to do with the codec itself, so to say you have to install a seperate plugin (omfg) in that program to analyze mpc's operations really isn't fair. I am quite happy you didn't post non ABX listening tests this would have topped it.

for made you look it was 5 mb aps but almost 3 with mpc. whats up?

which is the best encoder? lame3902, mpc, or ogg? thanks

Thats because mpc is more efficient at acheving transparancy than mp3. mp3 needs lots more bits to compress (almost) transparently as its an inferiour compressor, and by design it will never be as good as mpc quality, even at its highest setting!

Better use of VBR. MPC doesn't use preset bitrate sizes. It's simply being more efficient. Really, if you want to see whether you want to use MPC as your main format or not, do some ABX tests and see if you can tell CD from MPC from MP3, then go with what seems most efficient and the best quality to you then. Bitrate and filesize really don't mean all that much--in fact, wouldn't you think that a smaller filesize is better if it still sounds the same (or in this case, most would agree better)?

As for Winamp, just to make it as full-functioned as Foobar, you need a myriad of plugins, so the options that I would suggest to you are:

1) Adopt Foobar as your main player--no skins, awesome feature set
2) Just add a few more plugins to Winamp. I've never noticed much of a performance hit adding an input plugin or two....

depends on what you want... quality at mid to hi bitrates then mpc... low bitrates then its ogg... maximum compatibility then its lame mp3...

i thought ogg and mpc was high and lame is considered low now? i still like LAME more than mpc. is ogg worth trying?

If you have decent hearing and want the best sound, I'd say go with MPC.. if you want a more usable format and smaller sizes, go with ogg. To me ogg sounds fine well below 128kbs, so it's a great improvement on lame. Yes you should!! I regretted waiting so long to try it out properly

Yes, Vorbis is great at low bitrates, but not on every music...
Produces 'pleasing' (hard to detect without original) metallic artifacts on some other.
(eg. Dire Straits - Money for Nothing track with bitrate up to q6 [~210kbps]
LAME has much more problems on that of course (not with aps [~250 kbps]),
MPPENC has none with q5 and produces smaller files than oggenc [~170 kbps])

I see. But, how can MPC be better than LAME mp3 if the bitrates are almost the same as mine? I thought the reason of MPC was to hit higher bitrates but, it didn't seem that way.

and i reripped with extreme xlevel. i must say the quality is worse to lame's aps. i dont know why but, to my ears it sounds as if lame hits all the right notes and the quality of mpc is warped. i dunno maybe im just not very accepting of new formats and am sceptical as an audiophile wannabe but, i prefer LAME.

MPC is not intended to hit high bitrates. It is made to hit whatever bitrate necessary to achieve transparency to the original audio sample. Whether it be 64 kbps or 320 kbps. High bitrate does not equal high quality.

ABX testing is the standard method of comparing audio encoders. Can you tell the difference between the original & the MPC encode & the LAME APS encode?

The purpose of MPC is to achieve better quality than Lame at the same bitrates, or, looking at it in a different way, to achieve the same quality, but without needing to take so much space.

Bigger files, higher bitrates don't always mean better quality. It can easily mean sub-optimal encoding algorithm.

Try a Xing MP3 Encoder at 320 kbps, and see what I mean

Once more, to make it clearer:

It is quite common to get an MPC file that uses a bitrate of around 170, but sounds better than an MP3 file of around 210.

And once more, for completeness sake: if you REALLY want to know if the lower quality you think you hear actually exists and it's not just a product of your imagination, use ABX blind testing.

Go here:
http://www.hydrogenaudio.org/forums/index....t=ST&f=1&t=4917
And check the tools for Blind Testing.

nice description feces... mpc being warped and not hitting the notes...

I use MPC, i've got ALL my cds encoded in it.. sadly i'm never at my computer and then very rarely do i listen to music at the moment.. I also have a jukebox3 and i listen to this lots.. especially at work... as such I love Lame too... I love it for how good it makes mp3s sound... I dont think i've yet to here one of my encodes on my jukebox sound bad, and I have £300 of headphones plugged into it.. sometimes nearly £500.. (work cans, and home etys both with fixup mini headphone amp)... I encode with --alt-preset extreme... standard i'm sure would of been good enough but i was susceptible to the placebo urge to go for higher bitrates.. plus on my 20gig jukebox i can get 200 albums at alt-preset extreme which is more than enough...

I assume you are just making these comments to wind people up... if not and you truly dont like MPC.. you could have problems with your plugins.. I put up recordings in MPC format for people to download... I have had a few people complain of really poor audio quality... it turned out to be some plugin problems.. I think he was using winamp3 too at the time... trying foobar to see if its a night and day difference with mpcs... or using mppdec to decode them back into waves to play them may help eliminate this... other than that you either have the biggest hit of inverse placebo where you just can't believe anything is better than lame alt-preset standard and so you are unable to even hear the differences.. or perhaps you have really poor hearing or equipment that makes it difficult to here problems with poor quality 96k mp3s say.... or you are just trying to make an april fool joke or something...

Whatever it is its quite entertaining to read but in short.. if you dont like MPC, dont use it... stick to your mp3s... at least you are using lame and the alt presets... the quality of that shouldn't be too lacking on most things... if you want higher bitrates go for --alt-preset extreme or even insane any higher than insane tho and you will have to try mpc for higher bitrates... or if you want to really push the boat out... give lossless a try... try monkeys audio or some of the other ones... they sound perfect and make nice and big files 1/2 the original waves size... it'd be interesting to hear your opinions on those formats and whether lame --aps still sounds superior... perhaps you're getting what some minidisk fans get.. where they talk about the compression actually improving on the original source...

I think you like LAME that much that it is your anticipation that -aps must sound better.

Anticipation can have a big influence on one's perception. Last week, i saw a test on TV. They were in a TV studio and there was a bottle with a yellow fluid. They told the audience that this fluid has a distinct smell to it, and they handed out yellow cards, saying that everyone should hold up his card as soon as he smells something. After they opened the bottle, within some seconds the first people next to it raised their card. A few minutes later, about half of the 100 or so people had their cards up. When asked, they said it smelled "egg-like", one woman said she had problems to breathe. The show's host then asked one woman to drink from the fluid. She said the fluid has no taste. The host then disclosed to the audience that the bottle contains nothing else than colored drinking water.

i can tell the difference between LAME and mpc and i believe that LAME is better qualtiy. and LAME between the actual cd sounds the same to an audiophile like myself so... i believe that LAME is better than mpc. if any one has comments please feel free to add onto this thread its good to hear others opinions.

Another thing to try is to turn off the EQ in Winamp before comparing files. The EQ can make it hard at times to tell the difference between two formats.

also if u guys are concerned with the transparency of LAME use the best command line (i believe insane or extreme) and it will stay at a constant 320 for a stable bitrate of absurd unneeded bitrate numbers.

Samples? ABX results? I know you didn't do ABX tests. You shouldn't post these things when you have no idea what you're talking about, sorry.

A few humorous remarks from the chat:



You can't tell the difference, yet you can?

Try ABC/HR and then post the results.
Many forum members (MPC developers and I included) would be very interested in that sample.
It doesn't need to be a whole piece, only the fragment you think is worst.

OMG, another "audiophile" that will not listen to reason, and will keep refusing to do proper blind testing... I'm outta here.

Live happily with your fuzzy warm feeling. Good luck to all the rest, if you are willing to fight with the same thing again.

feces1223, you look at encoding speed and filesize/bitrate and draw conclusions about sound quality. This is the wrong approach; you should go the other way around. You could ask: how can mpc sound so good while being so efficient? Bitrate doesn't have to say much (try mpeg2 vs mpeg4 video) when you are comparing two different systems like mpc and mp3.

MPC is highly tuned and was allways geared for high quality. Mp3 and Ogg were not, they have been aimed at 128ish bitrates with acceptable quality but not with transparency in mind. That is the main reason actually. If mp3 had been designed for transparency, they would have left out some tricks that work well for lower bitrates but make transparency very difficult (for example: lousy compression for sounds >16kHz). Since Ogg is targetting the market currently occupied by mp3, the developers neglected the high end a little (but Ogg isn't crippled like mp3, so the potential is there).

I think lame alt-preset standard still beats Ogg q6 in terms of quality; Ogg messes up quite often (but this is not the end of the story as Garf is currently working on the higher bitrate modes). Ogg is superior to simple lame defaults though (lame -b 192 -h bla.wav). For medium-low bitrates (80-140 maybe) Ogg is fantastic. (ymmv)

Also, I don't think installing a Winamp input plugin is too much to ask. People install fancy visualisation and magic DSP plugins everyday.

edit: damn, I'm slow.. while I wrote my post, 11 others popped up.

I would simply recommend spending less time looking at the bit-rate indicator on your Winamp, and more time listening to the music. Unless your ears and stereo are very good there's not much chance that there will be anything to complain about using MPC at quality 7. Encoding time, bit-rate fluctuations, etc. is all irrelevant. How does the music sound?

As good as LAME is, and it seems to be the best for MP3, it's simply not as good as MPC for high quality, high bit-rate music. As you said, perhaps you're just skeptical, slow to accept a format you're new to. Give it some time.

If you're absolutely convinced that MPC isn't good enough (though I don't see how that's possible for all but a very picky golden-eared few) you might try lossless encoders instead.

No question that MP3 is the more supported format though. Spread the word about the better codecs and hopefully that'll change in time.

The rules of this forum state that when you claim something that goes against the common knowledge (after monthes of use, tens of samples have been found to be blindly distinguishable from the original with Lame, and only very few with MPC, and at inferior bitrates), you must provide the results of a blind ABX test, otherwise, your opinion can't be taken into account.

You must understand that in this forum, among 10 people posting tricks to improve the sound of MP3, in average, 9 of them use switches that don't have any binary effect (disabled by other opposite switches, redundant, etc...) and 1 actually improves the sound, but at the cost of an big filesize increasing.

Therefore we can't discuss claims made without blind testing, because if the 10 above people would have all performed ABX tests, 9 of them would have got no result, and avoid wasting our time explaining them that their modifications were actually not taken into account by the encoder.

If Lame sounds better, you must be able to succeed in a decoded MPC versus original wave ABX test. The programs linked in the FAQ are easy to use, you just have to decode one of your bad MPCs and listen to it.

I am not a moderator, but imo threads with this kind of speculation/pseudoscience should be locked.

No, we need converts to fill this forum with useless/useful questions

<edit>
Forgot to add: and new insight.

Don't listen to me, look at the calendar
</edit>

Not, this is no april fool. I already talked to him yesterday in the chatroom, that's when he started to ask about MP4, MPC and so on.

I explicitly said to me.

Where was the joke then..?

Your opinions are worthless until you conduct some blind testing.

I myself am not concerned about transparency with LAME at all because I achieve it better with mpc.

you guys thanks for some nice replies. Cut me a break i just learned that such formats existed! i know im an idiot but dont rub it in . Any ways ill search up ABX tests and try to run some. i will make a new post with the hardfacts soon (im gonna get it straight)!!!.

ok i got the two wavs and WinABX. now wat? do i play both wavs straight through and it will show results?

@feces1223
You have unwittingly stumbled into an ABX or your opinion is shit forum. Your ears cannot be trusted unless you ABX, and, after ABXing myself (well, not actually myself, I used audio files ), I am a firm believer in this philosophy. So, please do what everyone is telling you: ABX MPC with Mp3, or whatever, and then give us your opinion.

Later.

Ok i listened to both back to back with very nice Boston Speakers. I can NOT tell the difference between the two. They clearly sound exactly the same. I'm sorry maybe i'm just not listening to hard but, i've tested several times and they sound like there is no difference (--alt-preset standard vs. --insane --xlevel)

@feces1223
Did you play the entire song in your tests? Short clips, like 30 seconds or so, are more reliable for ABX purposes. And, also, if you could use headphones during the test, your results would be more accurate. Speakers are very forgiving, and, furthermore, the ambient noises (internal and external) in the room colour the sound. I hope this helps.

Later.

My first response, after reading the beginning of this thread, was that it was an April 1st joke that *really* got the guys on the site riled up (good one). After I realized that it was not, in fact, a joke, I decided not to post, wasting my time on another "troll." However, feces1223's last post has given me great hope Thanks, man. You always do yourself a favor when you ABX. Many people "hear" things that aren't real, due to personal expectation. Kind of like the saying "hunger makes the best seasoning." I used to do this quite a bit when I first got into audio compression (back then I was still compressing VBR MP3s with Xing's codec, lol), and I don't doubt that I still do it some. Anyway, to keep this from becoming a rant, thanks for not being a troll, and listening to reason

would make more sense to abx --alt-preset standard vs mpc -standard

eh well thats just the two i compared. Best imo against the best other setting. And i tried with headphones. Couldn't tell the difference between to the two. And i did do the first 25 seconds before u even recommended. So, ill stick with my LAME but, thanks a lot guys for encouraging me to test it blindly. It really did prove that i'm not just getting a gut feeling and to have some raw evidence. Well, if new versions come out of MPC or future codecs, i'd be glad to try them (it IS fun to test them out). and i most likely will adapt to mpc over time if it becomes a more popular format. I definitely recomend it though although i still trust the signature Dibrom complile of 3.90.2

No no, okay if you ABX between a MPC --standard and LAME --alt-preset standard, and find no difference again (as i expect will be the case) why not switch to the MPC with its lower bitrate? or try even MPC --quality 4 and see if you can tell the difference. What you are seeing now is that both --alt-preset standard and --insane sound perfect, which means they are BOTH TOO HIGH FOR YOU. Don't waste bits!

--alt-preset standard was INTENDED to be the minimum transparent bitrate for MP3.
--standard was INTENDED to be the minimum transparent bitrate for MPC.

if you can't tell the difference, that's GOOD. Compare them both to an original WAVE. Still no difference? GOOD. Use the MPC method then because it's smaller! Understand?

Great. I only brought up these two points because you mentioned that you used Boston speakers, and, since you are new to ABX, I didn't know if you used the whole song or not. Oh yes, like you, I too prefer Lame 3.90.2.

Later.

no more comparisons. sorry i could go on forever testing different settings with different encoders but, i've made my mind to stick with dibroms compile of 3.90.2 until mpc is further developed and a more signature format. thats my final verdict, it might change down the road but, it stands where it is as of now.