Musepack Forums  

Go Back   Musepack Forums > Main > General

Reply
 
Thread Tools Search this Thread Display Modes
Old 03 June 2006, 11:06 pm   #1
Gow
Guest
 
Gow's Avatar
 
Join Date: Jun 2006
Location: IN, USA
Posts: 1
Default My return to using MPC...

I used to use mpc as my preferred lossy audio format a few years back (1.15 alpha / introduction of 1.14...also used 1.12 and 1.13), than got swayed by the ogg aoTuV tests that boasted superior lossy, than went through the "lossless" phase followed by the iTunes AAC phase. *shudder* :lol:

Now I make a return to using MPC (which is made easier due to the EAC trick), because it is a good audio format for transperancy to my ears and the cons of the formats are unimportant to me. After all, the two possible choices for DAPs are going to have or do have Rockbox developed for it.

By the way, I encode at the setting "insane" / --quality 7.

What I don't get is that because of the trouble of finding an experience developer for the format who can handle an advance code that "people at a certain forum" just writes MPC as a dead format. Not taking in the fact that 1.15v has a good encoder plus no bugs that I have noticed while listening (maybe bugs...not an expert or claim to be). They have also attacked people for using the format...when I thought the case was always, as with anything in life, a personal computer is personal...the user should decide.

I hope an experienced developer comes to this format...for to further develop the code could be quite a challenge that anyone wanting to prove their worth could benefit from doing. Than with the shortcomings of the format addressed, the reasons not to use this format according to "people at a certain forum" would be invalidated.

Personally, it is of my opinion that constant development is not all too great as it makes past encodes look shabby and the user has to re-encode the audio files (ogg, mp3, iTunes AAC anyone?). This is what ultimately started to turn me off from ogg aoTuV...constant bugfixing or audio encoder adjustments. Sort of ticks me off when I have to re-encode something to fix a bug that I may have missed with due to my audio library's size and is fixed in a new release. It is why from ogg I switched to lossless, first FLAC and due to mostly laziness on my part (didn't know at the time how to set up *.exe encoders in EAC or didn't care about) I tried iTunes, which ripped and encoded CDs as I inserted them.

Now granted, maybe the future of audio formats is lossless but that future is still years off from competeing with the file size of lossy, but until that time I figured I would go back to a lossy format my ears loved and to break away from iTunes AAC for the superior ripping of EAC and an audio format I enjoy listening to.

Of course, keep in mind this is all opinion, conjecture and the like. As a medical drug ad states:

**Results May Vary**

That is the point with everything. What one person likes can differ from what another likes.

My only advice for developers of this format is this:
- Don't get high and mighty...overconfidence is a bad thing.
- Fix those bugs...cause I love it when they are fixed 8)
- Don't give up because something is hard...best things in life are best achieved by putting your all into it.

Probably sounds like advice from a advice book in the supermarket line...heh.

That about wraps it up for me.

P.S. Any iTunes users (or former ones), one question I need answered (I will probably google it after I type this): I set iTunes 6 to encode AAC at 256kbps VBR...the end product for the most part looks and plays like CBR in Winamp. Is it really VBR or pseudo VBR (ABR)...or just CBR with a VBR option that gets ignored. Just something to type further about.

Out!
Gow is offline   Reply With Quote
Old 04 June 2006, 07:35 am   #2
forlorn
Member
 
Join Date: Jun 2006
Posts: 3
Default

I think complaining about improvements to an encoder is one of the most ridiculous things ever. You know, the oft-used cliche about Rome not being built in a day and all. MPC has been around since 1997 while Vorbis first hit in 2000, thereby giving MPC a 3-year headstart in the tuning and tweaking deparment so let's be fair with our comparisons here...

Opinions here are fine only if they're pro-MPC. You should be fine.

Veer from that mode of thinking, however, and consider your voice silenced. Even engaging a discussion about the bitrate where perceptual transparency is reached by the majority (making a suggestion that it might be below 200 now) will get your account disabled and the only valid listening tests are the ones that show MPC's superiority. Reference another and you will be labeled as an idol-worshipper, no matter how well detailed the testing methodology is laid out. Those results are simply brushed off as an anomaly and are not to be taken seriously.

Open discussion is not allowed in this forum and it's really no wonder why it's dead. The TOS ought to be amended to make this clear so others won't waste their time upping the thread count here.

Finally, I was using MPC for encoding a large portion of my collection but witnessed the trend over the last year. I wasn't going to cling on to my favorite format when evidence of its superiority to everything else was waning. Not to mention, the seeking problem inherent in the format has always been a bane to its existence.

MPC is a great format for high fidelity but it no longer towers above the others from great heights and there is no harm done in pointing out this fact contrary to how Shy might follow up my post, then subsequently lock the thread and finally disable my account. I want the public eye to see just how this forum operates and expose where the true idolatry lies.
forlorn is offline   Reply With Quote
Old 04 June 2006, 04:27 pm   #3
Lefungus
Procrastinator
 
Lefungus's Avatar
 
Join Date: Jul 2004
Posts: 131
Default

First, let's just say I also disagree about how things are handled here. Closing topics after topics is, in my opinion, not the proper solution.

However, there is one annoying trend, well reflected in your post.

"Reference another and you will be labeled as an idol-worshipper, no matter how well detailed the testing methodology is laid out"

I assume you're talking about all those HA tests made by a single person. Whether those tests are valid or not is irrelevant, so let's assume they are. Just keep in mind those tests aren't made by YOU.

"I wasn't going to cling on to my favorite format when evidence of its superiority to everything else was waning."

Hype around mpc has never been so low. No new releases, no new code, no new interesting tests. No community, no nothing. Evidence is definitely here, using mpc is not *cool* anymore.
It has never been about transparency. You're not doing those tests yourself but reading evidence on the net, so by all means, switch to something else.
For unsecure people, I recommend using lossless, you won't get that nasty stomach feeling anymore about that codec your never choosed yourself. And if ever that codec's hype drop dangerously, you can switch ! You can't beat that.
Lefungus is offline   Reply With Quote
Old 04 June 2006, 06:34 pm   #4
Shy
Admin
 
Shy's Avatar
 
Join Date: Jul 2004
Posts: 372
Default

forlorn, first of all, no one here disabled your account or "silenced" you. If you feel you're facing some sort of dictatorship (I know some other forums you frequent are much more "liberal"), you're welcomed to leave, as I don't appreciate your conspiracy theories the least. I assume you got locked out by phpBB after submitting a wrong password 5 times.
This forum, unlike other forums, is not your playground, personal opinion blog or nerve treatment course, it's a support forum. The fact that your post still stands shows that this is no forum like Doom9's or some others. We don't close topics because we try to silence MPC haters, we do when they clearly violate the TOS or when the topic has fulfilled its purpose, and we don't delete posts.

As for your "head start" claim, Musepack has been in development since 1997, and Vorbis has been in development since 1994. Both were released at around the same time, 1999-2000.

As Lefungus said, the issue was with your sweeping commentary based on nothing but one guy's test and not your own test or verification of that test, or anyone else's. I don't say anyone worships idols for no reason. You decide what is considering a person the almighty golden eared guru whose words are nothing but facts without verifying a thing and then making sweeping claims.

Of course open discussion is allowed here, as long as it's not dragged to useless topics and a vast amount of controversial "opinions" that have nothing to do with the forum's topic, an audio format called Musepack, not other formats like Vorbis (which you've focused on completely in the first half of your other post).

"Evidence" that the format's superiority is waning? Yet again, you have nothing to base your claims on except non existent "evidence."

If you have a problem with our TOS which you think should be amended, again, you don't have to post here, but as long as you do, you'll have to follow those terms.

Finally, let me remind yet again that we don't even care if you choose to use any other format regularly. We don't make it our goal to drive people away from using other formats. We know what's good for us and offer it to anyone interested, and don't gain popularity by trolling, unlike people who make it their goal to tarnish the names of their competitors. We greatly appreciate talented people such as Monty, the main Vorbis developer, who we have a good relationship with. Simply put. if you like trolling, you won't find it here, you're free to use anything you want, and you're not free to troll.
Shy is offline   Reply With Quote
Old 05 June 2006, 12:47 am   #5
forlorn
Member
 
Join Date: Jun 2006
Posts: 3
Default

I'm impressed and surprised that this thread is still open. You've proven me wrong, Shy. My other account simply wouldn't allow me to log in after the other topic was closed, even after the first attempt...it simply sent me to a blank page. When trying to retrieve my password thru email, my account was deemed inactive for some reason.

I will say this about the test I referenced and why I did. I had a cursory listen of the samples used in the open 128 kbps test earlier this year and my ears (I've got hyperacusis in the left and both have been subjected to loud volume sources in the recent past) and equipment and artifact training were not sufficient enough to pick up the differences, so I've got the sinking feeling that any test I try to conduct at such bitrates (~180) will be especially meaningless, so that is why I referenced another individual's test (and someone whose methodology has been proven reliable). Additionally, I haven't seen any similarly conducted public tests at these higher bitrates. Perhaps that is what is needed once and for all to settle the score.

And at the very least with regards to 128 kbps, I did just follow up on the soundexpert.com public listening test and it does indeed show that MPC isn't at the top of the pack (nor is Vorbis): http://www.soundexpert.info/coders128.jsp - Take that perceptibility scale with a grain of salt. I personally don't find it at fault.
forlorn is offline   Reply With Quote
Old 05 June 2006, 03:04 pm   #6
Shy
Admin
 
Shy's Avatar
 
Join Date: Jul 2004
Posts: 372
Default

I don't remember anyone proving guruboolez's methodology is reliable, or that any of his tests is reliable either.
I don't have to take that SoundExpert test with a grain of salt. According to it, MPC, Vorbis and AAC are tied (higher than "imperceptible from original").
The reason you haven't seen any group tests at above 128kbps is that no one bothers to make them, even though the interest is high and people keep talking about it and have been making suggestions for years.
Shy is offline   Reply With Quote
Old 07 June 2006, 09:41 am   #7
shadowking
Member
 
Join Date: Aug 2004
Posts: 4
Default

To leave a format for another because of active developement and bugfixes is ridicuclous. Its not hard to find out if a bug was stuffing your encodings - and when was the last time that the major formats had such a defective release that people had to re encode ?

Secondly, I keep hearing that MPC is a great codec at quality 7 and that is enough reason to use despite its now apparent shortcomings. Today mp3 / aac / ogg and mpc are strong at 250k.

I used MPC for a while because at the time (2002-2003) it offered some neat advantages over the competition:

- gapless playback
- Transparency on many signals at 170k
- Fast encoding
- replaygain support

At the time, I couldn't get mp3 to play gapless which is a huge issue, no replaygain, slow LAME 3.90 vbr encoder, fatter files on loud music.
Vorbis around 1.0 had HF noise issues and was still much slower to encode and AAC was slow, not gapless and it was hard to find a free decent encoder.

I thought that the seeking issue issue would eventualy be solved and maybe MPC would get gapless HW support like today's rockbox. Maybe if it was supported like mp3 I would have stayed with it forever.

Now at 2006, things are different. Its not like MPC quality has declined but I cannot see too many reasons anymore to use it like I did then. Now all the competition have tuned VBR modes that are performing well at 180k, LAME and vorbis encoders have had significant speed increase without affecting quality - LAME vbr-new speed is competitive with MPC, while vorbis LANCER builds blows them both away. Gapless playback isn't such a big issue on the PC since foobar2000 handles this fine for some time now. Ditto with replaygain. MPC is still behind the rest in hardware support (although its improving). MPC still doesn't seek (this might also change). MPC developement is dormant. MPC is useless at bitrates <100k, while vorbis and aac are stronger in these lowish bitrates and are good at higher bitrates. A hybrid format like Wavpack and Dualstream be both a lossy and lossless encoder, they encode quite fast, seek well and quality is decent at 300k which isn't tragic considering so many believe MPC needs 250k.

I am not trashing MPC, but to me the real advantage was this fast gapless encoder that needed 170k for good quality. It still is a good encoder at 170k, but the other formats are now all this and more.

One can argue against me, but its not going to make things any better for this format.
shadowking is offline   Reply With Quote
Old 07 June 2006, 11:33 pm   #8
xmixahlx
Musepack developer
 
xmixahlx's Avatar
 
Join Date: Nov 2004
Location: seattle, washington usa
Posts: 111
Send a message via ICQ to xmixahlx
Default

* mpc is great at quality >4
* seeking has been worked on and partially solved
* lancer is not worth using and is inferior to the reference and aotuv
* mpc isn't dormant, but no one is actively working on the psymodel
* the hybrid idea is stupid (NOTE: yet wavpack is fantastic)
* yes, formats are catching up to where musepack was 4 years ago...

none of those scream *jump ship* people will use what fits their personal criteria.

my criteria is:
* OSS or at least linux support
* >100 GB of lossy - musepack (and ~5GB of various formats)
* <50 GB of lossless - FLAC/SHN bootlegs
* AUDIO CD copies and Backups of all of my music
* Don't have a DAP

for me i've used musepack for lossy and flac/wavpack for lossless.
both are no-brainers for me and i haven't had a situation where i needed something different, so i am still chugging along.


later
__________________
-xmixahlx, the one they call "mike"
http://xmixahlx.com -|- http://rarewares.org
xmixahlx is offline   Reply With Quote
Old 08 June 2006, 12:17 am   #9
shadowking
Member
 
Join Date: Aug 2004
Posts: 4
Default

Quote:
Originally Posted by xmixahlx
* mpc is great at quality >4
* seeking has been worked on and partially solved
* lancer is not worth using and is inferior to the reference and aotuv
* mpc isn't dormant, but no one is actively working on the psymodel
* the hybrid idea is stupid (NOTE: yet wavpack is fantastic)
* yes, formats are catching up to where musepack was 4 years ago...

later
- Wow, So how is Lancer so inferior - care to provide creadible evidence ?
- Why is hybrid encoding stupid ? - It saves time and Wavpack has some HW support.

- The formats are not catching up but exceeding mpc. Above 128k quality isn't much of an issue with any of them and AAC / Vorbis are quite capable below 128k. Four years ago there was some aac/vorbis HW support, they were still better than mpc in low bitrates. Four years ago other formats were seeking and transparent at 250 k.

If anything, At least MPC did achieve something in that it forced other format devs to take vbr tuning seriously some years ago.
shadowking is offline   Reply With Quote
Old 08 June 2006, 05:37 am   #10
forlorn
Member
 
Join Date: Jun 2006
Posts: 3
Default

Quote:
The formats are not catching up but exceeding mpc.
Watch out, you'll need proof and individual ABX tests aren't sufficient (unless you've done them yourself). Still, I don't think there are any current (2005-2006) public listening tests available that show the opposite (MPC still being superior in quality above 128 kbps) to be true either. That leaves us in quite the conundrum.

I'd like to know what's wrong with the Lancer builds myself actually. And apparently the other formats still have 4 years to go before reaching the quality of MPC. I'm not trying to troll....I would just like to know where these claims stem from.
forlorn is offline   Reply With Quote
Old 30 June 2006, 12:27 pm   #11
user
Member
 
Join Date: Feb 2005
Location: in the great wide open
Posts: 11
Default

soundsexpert test:

PERCEIVED AUDIO QUALITY

results dated from 2006-06-29


--------------------------------------------------------------------------------


aac he CBR@131.6 (Winamp 5.21) 6.19 (≈1%)


aac CBR@132.3 (Winamp 5.21) 5.71 (≈3%)


aac VBR@128.0 (NeroRef 0506) 5.38 (≈5%)


mpc VBR@128.4 (1.15v) 5.37 (≈1%)


ogg VBR@129.9 (OggEnc 2.8) 5.27 (≈1%)



aac VBR@134.5 (iTunes 6.0) 5.15 (≈2%)


mp3 VBR@112.7 (Lame 3.97b2) 4.55 (≈3%)


ATRAC3 CBR@135.4 (LP2) 3.73 (≈0%)


mp3 CBR@127.9 (Shine) 2.07 (≈1%)


Though the test-method of soundsexperts can be questioned, the results show a tendency.
mp3 shine cbr the worst, atrac bad, mp3-lame vbr a step better, then the modern formats in several variations at the top, including MPC and its competitors aac/ogg.

So, yes MPC at quality 4 and better is still very competitable with all the other modern HA-hyped formats...



some answers and corrections:

- Why is hybrid encoding stupid ? - It saves time and Wavpack has some HW support.

well, hybrid being stupid is probbaly exaggerating from ximaxx personal point of view.
Now my personal opionion:
I need Lossless and lossy, I encode to both.
How do I solve it practical ?!
I tried Wavpack lossy,
and I find this alternative a great solution, but:
- too few hardware support, especially for the lossy wavpack ?
- I encode these days via mareo.exe to 3 formats at different bitrates/qualities:
1. Lossless Flac or Wavpack, atm preferring Flac due to hardware support
2. MPC at --quality 7 - 8 --ms 15 --xlevel for use on laptop AND cheap/small-sized archival backup of the Lossless on different media,
3. Mp3/Lame -V5 vbr-new at 130k for daily usage music, for usb-stick 1 GB in car-radio and via heardphones Koss KSC75 for sports running aoutdoors.

options 1. and 2. fulfill HiFi requiements/quality.
because of the security/backup requirement, a combined Wavpack hybrid Lossless/lossy is out of question ! 1 file away, it isn#t Lossless anymore.
And MPC reaches transparency at clear lower bitrates than wavpack lossy.

forlorn and others admit, they aren#t even able to distinct lossy at 128k vbr from eahc other or Lossless. Well, I go towards 40 and it isn#t still a prob for me.
Opposite to recent HA majority opions, hardware/HiFi equuipment plays a role, it is a difference if you listen via headphones or via good speakers etc etc. And it is a difference, how your ears are. Damaged by live concerts at high volumes ?
Please wear plugs to protect yourself in concerts...

I have been watching the lossy & lossless scene before the birth of HA, since r3mix times, and have sadly recognized, that the lossy scene is dominated nowadays by zealots, especially at HA, refusing to test the new publically untested 1.15v in their ususal 128k tests etc.
Just as modern formats get tuned to bearable results at low bitrates, the developers wars are carried out between 32k to 80k !! nowadays.
All above is out of focus and uninteresting for paid developers.
You haver to listen to your music on cell phones..
and you have to pay for your lossy 64k phone music,
if you don#t do, and listen HiFi/free mpc , you are a non-conformist! maybe a terrorist or a revolutionay or revenge/backwards old fashioned guy !
So they tell you, to get your wooden home under the green floor, to not to disturb the crowds in their 128k = transparent/overkill dream!

addon-edit:

the "portable hardware" argument:

as you see, i encode to 130k mp3 lame for protable.
Because this format is palyable in every device, that si true portability !
Ogg, aac are at same low portability scale as mpc.
MPC can play since years/months on phones & PDA/navigational mini computers with that certain player,
Rockbox, eg. MPC on ipod
But: all those special players, including aac/ogg players are to rare still at the moment, and too expensive.
as you need a good brand player.
Compare to the variety of mp3 portables:
those are sold by non-known brands also, very cheapo!
and since usb sticks, the only portable are those mini sticks (for me), all other solutions have too amyn disadvantages.
and USB stick sized:
only mp3 (and crap-wma..)

at the begin of mp3, I knew people, who built in PC to car, to play mp3 from hd !
already years ago.
Still a good solution for cars, and then MPC rulez!
__________________
users' guides: www.High-Quality.ch.vu
user is offline   Reply With Quote
Old 01 July 2006, 01:02 am   #12
xmixahlx
Musepack developer
 
xmixahlx's Avatar
 
Join Date: Nov 2004
Location: seattle, washington usa
Posts: 111
Send a message via ICQ to xmixahlx
Default

obviously my opinion that hybrid lossless/lossy being stupid is just an opinion... really, it shouldn't matter. use it if you want, but it doesn't compare with musepack at standard bitrates.

my statement about lancer was meant for the LAME MT project... i'm not sure how i mixed the two up then at the time, and only seeing shadowkings post right now.

the seeking problems with musepack will remain the greatest deficiency.


later
__________________
-xmixahlx, the one they call "mike"
http://xmixahlx.com -|- http://rarewares.org
xmixahlx is offline   Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
CDEX and MPC files Valvola General 1 24 June 2006 11:19 am
mpc and Wavelab 5 Kattara MPC for Windows 2 27 April 2006 02:52 am
All MPC files broken after file restore deus62 MPC for Windows 3 19 August 2005 05:55 pm
low level clicks and noise davidm General 20 29 June 2005 09:05 am


All times are GMT. The time now is 05:24 am.


Powered by vBulletin® Version 3.8.11 Beta 2
Copyright ©2000 - 2019, vBulletin Solutions Inc.