HydrogenAudio

Lossy Audio Compression => MP3 => MP3 - General => Topic started by: Blanka on 2005-02-26 07:56:36

Title: How to convince people to stop MP3 myths
Post by: Blanka on 2005-02-26 07:56:36
 Has anyone ever had a hard job trying to dispell certain mp3 myths? Like:

1) Joint Stereo messes up the sound.
2) Joint Stereo ruins high frequencies.
3) You can always tell the difference.

etc etc.

I feel like I'm bashing my head against a brick wall trying to talk to these people... 
Title: How to convince people to stop MP3 myths
Post by: evereux on 2005-02-26 08:11:01
Oh, I think we're all pretty familiar with your problem. Your best bet is to have a read through the FAQs here, they cover some of the issue's you mention. Also, read up on ABX testing (also covered in a FAQ) and point the naysayers towards that FAQ and ask for ABX results.
Title: How to convince people to stop MP3 myths
Post by: DreamTactix291 on 2005-02-26 08:12:46
Don't worry.  I can assure you you're not the only one who has come across the problem of others' ignorance and general not caring about the subjects.  Both this and VBR are touchy subjects people are stubborn about.

You know how I combat this one as well as people saying VBR isn't of high quality?  It's simple.  I ask them "If it destroyed quality then why are both VBR and forms of joint stereo used in lossless codecs?"

About the best you can do really.
Title: How to convince people to stop MP3 myths
Post by: Blanka on 2005-02-26 08:16:34
Quote
Oh, I think we're all pretty familiar with your problem. Your best bet is to have a read through the FAQs here, they cover some of the issue's you mention. Also, read up on ABX testing (also covered in a FAQ) and point the naysayers towards that FAQ and ask for ABX results.
[a href="index.php?act=findpost&pid=277196"][{POST_SNAPBACK}][/a]


I've seen people say that they don't need to ABX as they can hear the problem! 
Title: How to convince people to stop MP3 myths
Post by: westgroveg on 2005-02-26 08:16:58
Quote
Has anyone ever had a hard job trying to dispell certain mp3 myths? Like:

1) Joint Stereo messes up the sound.
2) Joint Stereo ruins high frequencies.
3) You can always tell the difference.

etc etc.

I feel like I'm bashing my head against a brick wall trying to talk to these people... 
[a href="index.php?act=findpost&pid=277194"][{POST_SNAPBACK}][/a]

Easiest way: perform public listening tests & prove your claims. Simple.
Title: How to convince people to stop MP3 myths
Post by: AtaqueEG on 2005-02-26 08:24:13
Still, there's people that will never be convinced.

I have seen people unable to ABX, still think that their files have "something" that they cannot put their finger on.
Title: How to convince people to stop MP3 myths
Post by: PatchWorKs on 2005-02-26 09:01:46
Quote
etc etc.


MP3 are crap

Simplest way: switch to Vorbis
Title: How to convince people to stop MP3 myths
Post by: jormartr on 2005-02-26 12:03:08
The general situation I have found is "Why encode at higher bitrates when 128kbps is just the same as the original".
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-02-26 12:46:40
I don't even care about trying to dispel these myths. People are happy believing them and encoding their music with them, so let them be.

The only reason I believe there would be for trying to dispel those myths is if you guys were into heavy file sharing...
Title: How to convince people to stop MP3 myths
Post by: Lyx on 2005-02-26 14:23:43
IMHO, it is now impossible to dispel the JS-myth on a large scale.

The only solution is disabling the switch completely in LAME.

- Lyx
Title: How to convince people to stop MP3 myths
Post by: Lyx on 2005-02-26 14:25:55
Quote
The only reason I believe there would be for trying to dispel those myths is if you guys were into heavy file sharing... :rolleyes:
[a href="index.php?act=findpost&pid=277243"][{POST_SNAPBACK}][/a]


So, do you think that all non-selfencoded MP3s are illegal? May i remind you about netlabels, artists who give their music or some tracks away for free and taped live-shows?

- Lyx
Title: How to convince people to stop MP3 myths
Post by: Benjamin Lebsanft on 2005-02-26 14:26:13
Quote
The only solution is disabling the switch completely in LAME.[a href="index.php?act=findpost&pid=277262"][{POST_SNAPBACK}][/a]

I'll second that
Title: How to convince people to stop MP3 myths
Post by: Gabriel on 2005-02-26 14:43:23
Quote
IMHO, it is now impossible to dispel the JS-myth on a large scale.

The only solution is disabling the switch completely in LAME.

Or to have user interfaces similar to this one:
http://lame.sourceforge.net/lame_ui_example.html (http://lame.sourceforge.net/lame_ui_example.html)
Title: How to convince people to stop MP3 myths
Post by: Lyx on 2005-02-26 15:06:56
Quote
Quote
IMHO, it is now impossible to dispel the JS-myth on a large scale.

The only solution is disabling the switch completely in LAME.

Or to have user interfaces similar to this one:
http://lame.sourceforge.net/lame_ui_example.html (http://lame.sourceforge.net/lame_ui_example.html)
[a href="index.php?act=findpost&pid=277268"][{POST_SNAPBACK}][/a]

Hmm, isn't this essentially the same? The stereo-option is just hidden in the frontend, making it appear as if lame doesn't have this option - the net result is removing the stereo-switch - but with the difference that the app needs to "support" it. I have reasonable doubt that the majority of the authors of the existing apps will understand that removing an already existing option is an improvement.

- Lyx
Title: How to convince people to stop MP3 myths
Post by: Acid Orange Juice on 2005-02-26 15:52:58
particularly, the myths that I more hate are these:

1) 128 CBR kbps mp3 is CD quality 

2) Joint stereo is bad, because destroys all the stereo information. 

3) Simple stereo is the best choice, because preserve all the stereo information. 

4) mp3 never is or will be transparent. 

5) VBR thing is bad, CBR is a lot better. 

6) 192 CBR kbps mp3 in simple stereo mode is the best solution for archival quality. 

7) Fhg is the best mp3 encoder. 

All these myths come from the ignorance. I try to convince to my friends that they are wrong, but; it's impossible 
Title: How to convince people to stop MP3 myths
Post by: Jens Rex on 2005-02-26 16:49:54
I just don't bother educating people anymore. I just ignore them. It's too frustrating, and you can't educate everyone. Educating one clueless moron, is like a drop in the sea of morons.

What does bother me however, is that the danish music paysites are using retarded LAME settings to encode their music, and that's really annoying. If they sold --aps MP3 or even just properly encoded 192 CBR MP3, I'd consider using them.

I even tried sending them a mail, and explaining the situation to them, giving helpful links etc., but all I got in return was some standard bullshit reply.
Title: How to convince people to stop MP3 myths
Post by: HotshotGG on 2005-02-26 17:32:03
Quote
All these myths come from the ignorance. I try to convince to my friends that they are wrong, but; it's impossible crying.gif


Personally if my friends don't listen to what I want I just to tell them to go F off. People are generally dumb, ignorant, and foolish are not going to listen to you anyway no matter what reasonable explanation you have for them they generally don't care.  That's why we have our uber-pretentious community here of course ;-D.  The only thing you can really do is "educate" the uninformed public and hope they listen.
Title: How to convince people to stop MP3 myths
Post by: Jojo on 2005-02-26 18:02:08
I'd just get rid of the stereo switch and re-name joint-stereo to 'stereo' - or map the stereo switch to joint-stereo

Sure, why would we care? Well, since all the other 'modern' codecs already use Joint-Stereo as default and call it 'Stereo' some newbie will actually believe that mondern codecs are far better than mp3...simply because they don't how to encode in mp3...
Title: How to convince people to stop MP3 myths
Post by: Brink on 2005-02-27 01:48:58
You cant convince people that something is terrible if they cant notice. So I dont try to dispell myths. If they really want to know information(or truth about audio coding), they probably will search for themselves.
Title: How to convince people to stop MP3 myths
Post by: Dologan on 2005-02-27 02:59:26
Or maybe we should map Joint Stereo as Good Modern Stereo and "normal" stereo as Bad Old Stereo. This way, maybe placebo will help us for once.
Title: How to convince people to stop MP3 myths
Post by: Cerbie on 2005-02-27 03:04:06
It is not your obligation to educate anyone until they notice skips from using sync or burst mode, clips and chirps from other corruption, garbly artifacts, etc., and want to get it done right.

Note: this is coming from a previous ToS #8 violator , who originally believed it all...until hundreds (maybe thousands, really) of hours of listeing began to train his brain into having good ears.
Title: How to convince people to stop MP3 myths
Post by: Bennyp on 2005-02-27 03:50:00
My opinion is that MP3 encoding washes out the high range. I prefer to encode my signals with ogg vorbis of quality 7 or higher, just to be safe.

As well, I generally avoid playing MP3 files that are less than 256 kbps CBR or 192kbps VBR at high volumes or in a performance situation. I once read that exposure to low quality lossy encodings can cause cumulative hearing loss and tinnitus. While I can not say if that is true, I prefer to err or the side of caution.
Title: How to convince people to stop MP3 myths
Post by: znode on 2005-02-27 04:27:49
Quote
My opinion is that MP3 encoding washes out the high range. I prefer to encode my signals with ogg vorbis of quality 7 or higher, just to be safe.

As well, I generally avoid playing MP3 files that are less than 256 kbps CBR or 192kbps VBR at high volumes or in a performance situation. I once read that exposure to low quality lossy encodings can cause cumulative hearing loss and tinnitus. While I can not say if that is true, I prefer to err or the side of caution.
[a href="index.php?act=findpost&pid=277443"][{POST_SNAPBACK}][/a]


Troll?
Title: How to convince people to stop MP3 myths
Post by: Acid Orange Juice on 2005-02-27 05:52:07
Quote
I once read that exposure to low quality lossy encodings can cause cumulative hearing loss and tinnitus. While I can not say if that is true, I prefer to err or the side of caution.
[a href="index.php?act=findpost&pid=277443"][{POST_SNAPBACK}][/a]


Don't exist scientific evidence that supports this.. (or at least credible).

This is another new myth that we can add to the list. 
Title: How to convince people to stop MP3 myths
Post by: AtaqueEG on 2005-02-27 05:56:56
Quote
5) VBR thing is bad, CBR is a lot better. 
[a href="index.php?act=findpost&pid=277281"][{POST_SNAPBACK}][/a]


This one is actually kinda true.

CBR (320K/-api) is the VERY best LAME has to offer. It is overkill though, and it is not like I can hear a difference (my transparency thresold is about 160k in LAME)

Although you will not catch me dead telling that to my friends or anyone else.
Title: How to convince people to stop MP3 myths
Post by: AtaqueEG on 2005-02-27 06:04:43
Quote
Quote
I once read that exposure to low quality lossy encodings can cause cumulative hearing loss and tinnitus. While I can not say if that is true, I prefer to err or the side of caution.
[a href="index.php?act=findpost&pid=277443"][{POST_SNAPBACK}][/a]


Don't exist scientific evidence that supports this.. (or at least credible).

This is another new myth that we can add to the list. 
[a href="index.php?act=findpost&pid=277463"][{POST_SNAPBACK}][/a]


Long exposure to any kind of loud sound will degrade your hearing. But this has nothing to do with lossy encoding. But this guy is obviously a troll.
Title: How to convince people to stop MP3 myths
Post by: znode on 2005-02-27 08:16:36
Quote
Long exposure to any kind of loud sound will degrade your hearing. But this has nothing to do with lossy encoding. But this guy is obviously a troll.
[a href="index.php?act=findpost&pid=277470"][{POST_SNAPBACK}][/a]


Actually, I reread the thread title and the post again just now, and it looks like it was a joke that did a Mach 5 over my head. I feel embarassed now for such a lack of sense of humour.

Sorry Bennyp!
Title: How to convince people to stop MP3 myths
Post by: AgentMil on 2005-02-27 09:31:53
/ignore

Thats what I do these days no point arguing or stressing over it IMHO. Also I ain't fussy anymore when it comes to quality as long as it inaudibile when I listen to it I don't really care, music for me is to keep the silence away not to critic it. Sometimes I wonder why I purchased my Sennheiser headphones .

About that post about lossy encoding damaging your ears I think that a load of crock there are plenty of other stimuli that comes from other sound sources that aren't lossy so you still get the full spectrum of sound. I would imagine you would have to listen to lossy encoding over a very long time to get some sort of symptoms, but then again if you do everything in moderation nothing bad really comes from it.


Regards
Title: How to convince people to stop MP3 myths
Post by: Mr_Rabid_Teddybear on 2005-02-27 11:00:07
Decoding mp3 to wav gives the music better quality, because wav is better than mp3.

I've met this quite a few times, this is a level of knowledge that's quite common among "normal" people, lot's of them highly educated and long time computer users, only very very few have even heard of "joint stereo" so they can worry about it. Actually most people find 64 kbps CBR mp3 to be of acceptable quality.

Well, just encode your own music with something useful....
(mppenc --quality 6.00 here)
Title: How to convince people to stop MP3 myths
Post by: guada 2 on 2005-02-27 13:22:02
Hello everyone,

I would not give a favorable or unfavorable opinion on the question of the format mp3. 
Remind yourselves when it came out of the boxs, one was not unhappy of its arrival.It marked a lot of mind, and is made itself know all. 
Nowadays it stays" inescapable", contrary to the Divx that disappears little by little. 
 
Nevertheless other formats saw the day with more and more appreciated other resonant colors. what is the present case: AAC, Plop, Ogg, Mpc, Ape and well of others. 
 
Every format finds its place in the daily, the experts, the beginners and the hopeless. 
So are understanding towards others. 
It is only my opinion, but I share yours. 
 
Bye.
Title: How to convince people to stop MP3 myths
Post by: bond on 2005-02-27 14:00:24
just remembered the saying

"100 lemmings cant be wrong!" 
Title: How to convince people to stop MP3 myths
Post by: phong on 2005-02-28 16:52:32
See, the problem you're having is that you are a rational person.  As such, you expect people to respond to rational reasoning.  These people have superior brains that are capable of many kinds of thinking that you can't even comprehend.  Their minds are actually capable of constructing arguments that affirm their preconceptions completely circumventing any inconvienient facts.  They can produce a thought arc that goes around and "short-circuts" any argument you could produce.  You are unable to percieve this arc because it lies outside the realm of the rational.  Indeed, it may well reside in some sort of alternate universe that these lucky individuals have tapped into.

So, feel not frustration or pity, and instead fell shame and envy; your brain just isn't as good.
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-02-28 17:18:41
Quote
Quote
IMHO, it is now impossible to dispel the JS-myth on a large scale.

The only solution is disabling the switch completely in LAME.

Or to have user interfaces similar to this one:
http://lame.sourceforge.net/lame_ui_example.html (http://lame.sourceforge.net/lame_ui_example.html)
[{POST_SNAPBACK}][/a]
(http://index.php?act=findpost&pid=277268")

Rather like this: [a href="http://homepage.ntlworld.com/jfe1205/LAME/lamedropXPd2-3.97a7.zip]http://homepage.ntlworld.com/jfe1205/LAME/...XPd2-3.97a7.zip[/url]? A proposed version of lamedropXPd for the next releases of LAME.
Title: How to convince people to stop MP3 myths
Post by: kwanbis on 2005-02-28 17:22:36
Quote
Quote
5) VBR thing is bad, CBR is a lot better. 
[a href="index.php?act=findpost&pid=277281"][{POST_SNAPBACK}][/a]


This one is actually kinda true.

CBR (320K/-api) is the VERY best LAME has to offer. It is overkill though, and it is not like I can hear a difference (my transparency thresold is about 160k in LAME)

he probably means something on the lines CBR192 is better than VBR preset standard
Title: How to convince people to stop MP3 myths
Post by: Lyx on 2005-02-28 17:23:34
Quote
See, the problem you're having is that you are a rational person.  As such, you expect people to respond to rational reasoning.  These people have superior brains that are capable of many kinds of thinking that you can't even comprehend.  Their minds are actually capable of constructing arguments that affirm their preconceptions completely circumventing any inconvienient facts.  They can produce a thought arc that goes around and "short-circuts" any argument you could produce.  You are unable to percieve this arc because it lies outside the realm of the rational.  Indeed, it may well reside in some sort of alternate universe that these lucky individuals have tapped into.

So, feel not frustration or pity, and instead fell shame and envy; your brain just isn't as good.
[a href="index.php?act=findpost&pid=277881"][{POST_SNAPBACK}][/a]


LOL!
Title: How to convince people to stop MP3 myths
Post by: Acid Orange Juice on 2005-02-28 17:45:40
Quote
Quote
5) VBR thing is bad, CBR is a lot better. 
[a href="index.php?act=findpost&pid=277281"][{POST_SNAPBACK}][/a]


This one is actually kinda true.

CBR (320K/-api) is the VERY best LAME has to offer. It is overkill though, and it is not like I can hear a difference (my transparency thresold is about 160k in LAME)

Although you will not catch me dead telling that to my friends or anyone else.
[a href="index.php?act=findpost&pid=277466"][{POST_SNAPBACK}][/a]


Could you please read very carefully my post?

I am not talking in specify of the case of -api; I am talking that in GENERAL many people believe that VBR is bad, and as consequence of this bad MYTH they encoded in CBR all the time,  because they really believe that the VBR routines destroys the music.

Many threads here in HA that confirm the superiority of the VBR routines in LAME in comparison with CBR. Of course; -api setting is a exception.
Title: How to convince people to stop MP3 myths
Post by: Gabriel on 2005-02-28 22:11:50
Quote
Rather like this: http://homepage.ntlworld.com/jfe1205/LAME/...XPd2-3.97a7.zip (http://homepage.ntlworld.com/jfe1205/LAME/...XPd2-3.97a7.zip)? A proposed version of lamedropXPd for the next releases of LAME.

Don't know, I do not see anything. I launch it, the process is here, but I do not see any visual interface...
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-02-28 23:05:58
Quote
Quote
Rather like this: http://homepage.ntlworld.com/jfe1205/LAME/...XPd2-3.97a7.zip (http://homepage.ntlworld.com/jfe1205/LAME/...XPd2-3.97a7.zip)? A proposed version of lamedropXPd for the next releases of LAME.

Don't know, I do not see anything. I launch it, the process is here, but I do not see any visual interface...
[a href="index.php?act=findpost&pid=277968"][{POST_SNAPBACK}][/a]

Right mouse click over the box for the menu.
Title: How to convince people to stop MP3 myths
Post by: earphiler on 2005-02-28 23:06:34
I remember when I was first introduced to VBR and I heavily used EncSpot, I discriminated against it because it bothered me that the bitrate would jump around; it felt very buggy  looking back at it is quite comical
Title: How to convince people to stop MP3 myths
Post by: qualityequalizer on 2005-02-28 23:27:44
Quote
particularly, the myths that I more hate are these:

1) 128 CBR kbps mp3 is CD quality  :sick:

2) Joint stereo is bad, because destroys all the stereo information. :ermm:

3) Simple stereo is the best choice, because preserve all the stereo information. :sick:

4) mp3 never is or will be transparent. :angry:

5) VBR thing is bad, CBR is a lot better. :angry:

6) 192 CBR kbps mp3 in simple stereo mode is the best solution for archival quality. :angry:

7) Fhg is the best mp3 encoder. :angry:

All these myths come from the ignorance. I try to convince to my friends that they are wrong, but; it's impossible  :cry:
[a href="index.php?act=findpost&pid=277281"][{POST_SNAPBACK}][/a]


mp3 is not transparent. How can it be when certain people can detect differences between 320 mp3 and the original?

I can in some cds detect differences between 320 and original, and i used to believe 128 was cd quality. 
So it is all subjective..
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-02-28 23:37:01
Quote
Right mouse click over the box for the menu.
[a href="index.php?act=findpost&pid=277988"][{POST_SNAPBACK}][/a]


What box, good sire?

I ran it here at my PC. There is no indication of it anywhere other than at the task manager. No drag-n-drop window, no icon at the system tray or the taskbar... nothing.

Win2k / AMD Sempron 2500 / 768M RAM
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-02-28 23:39:39
Quote
mp3 is not transparent.
[a href="index.php?act=findpost&pid=277993"][{POST_SNAPBACK}][/a]


MP3 CAN BE transparent. Claiming a lossy format is or is not transparent makes no sense. It depends on the sample you're testing, your hearing, your equipment, the encoding parameters...
Title: How to convince people to stop MP3 myths
Post by: schonenberg on 2005-03-01 01:29:44
Quote
IMHO, it is now impossible to dispel the JS-myth on a large scale.

The only solution is disabling the switch completely in LAME.

- Lyx
[a href="index.php?act=findpost&pid=277262"][{POST_SNAPBACK}][/a]


Is there any reason we would ever really need simple stereo since Lame JS doesn't destroy stereo?

If no one can think of a good reason, maybe it should be taken out of Lame, or at least taken out of the --longhelp and removed from the docs. What do you think?
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-01 01:42:42
Quote
Is there any reason we would ever really need simple stereo since Lame JS doesn't destroy stereo?[a href="index.php?act=findpost&pid=278022"][{POST_SNAPBACK}][/a]


Hrm... multilanguage content?
Title: How to convince people to stop MP3 myths
Post by: odious malefactor on 2005-03-01 04:15:25
Quote
Quote
Right mouse click over the box for the menu.
[a href="index.php?act=findpost&pid=277988"][{POST_SNAPBACK}][/a]


What box, good sire?

I ran it here at my PC. There is no indication of it anywhere other than at the task manager. No drag-n-drop window, no icon at the system tray or the taskbar... nothing.

Win2k / AMD Sempron 2500 / 768M RAM
[a href="index.php?act=findpost&pid=277996"][{POST_SNAPBACK}][/a]

Perhaps it's only for XP, hence its moniker.

It worked fine on my XP box at the office, but no GUI appears on my Win2k box here at home.
Title: How to convince people to stop MP3 myths
Post by: VCSkier on 2005-03-01 05:50:58
heh, kinda funny....  i actually last week i had to do an "informative speech" for my speech class, and i did it on "music encoding," trying to fix some of this ignorance.  it was fun... i dont know if anyone followed it, but at least i tried... 
Title: How to convince people to stop MP3 myths
Post by: schonenberg on 2005-03-01 05:58:22
Quote
Quote
Is there any reason we would ever really need simple stereo since Lame JS doesn't destroy stereo?[a href="index.php?act=findpost&pid=278022"][{POST_SNAPBACK}][/a]


Hrm... multilanguage content?
[a href="index.php?act=findpost&pid=278025"][{POST_SNAPBACK}][/a]


Sorry, I don't understand.
Title: How to convince people to stop MP3 myths
Post by: Mono on 2005-03-01 06:07:08
Quote
Quote
Quote
Is there any reason we would ever really need simple stereo since Lame JS doesn't destroy stereo?[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=278022")


Hrm... multilanguage content?
[a href="index.php?act=findpost&pid=278025"][{POST_SNAPBACK}][/a]


Sorry, I don't understand.
[a href="index.php?act=findpost&pid=278065"][{POST_SNAPBACK}][/a]

Sometimes another language of a program is put in the other channel.

[span style='font-size:8pt;line-height:100%']Discussion about Joint Stereo quality split [a href="http://www.hydrogenaudio.org/forums/index.php?showtopic=31963]here[/url].
- the moderation[/span]
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-01 08:10:08
Quote
Quote
Right mouse click over the box for the menu.
[a href="index.php?act=findpost&pid=277988"][{POST_SNAPBACK}][/a]


What box, good sire?

I ran it here at my PC. There is no indication of it anywhere other than at the task manager. No drag-n-drop window, no icon at the system tray or the taskbar... nothing.

Win2k / AMD Sempron 2500 / 768M RAM
[a href="index.php?act=findpost&pid=277996"][{POST_SNAPBACK}][/a]

Really!!!!  How utterly bizarre. It works fine here, otherwise I wouldn't have linked to it. 

I'll check the compile options, etc., but I didn't change any of those from the previous version!!
Title: How to convince people to stop MP3 myths
Post by: Gabriel on 2005-03-01 08:23:47
Quote
Right mouse click over the box for the menu.

Which box?
Title: How to convince people to stop MP3 myths
Post by: evereux on 2005-03-01 08:38:38
Quote
See, the problem you're having is that you are a rational person.  As such, you expect people to respond to rational reasoning.  These people have superior brains that are capable of many kinds of thinking that you can't even comprehend.  Their minds are actually capable of constructing arguments that affirm their preconceptions completely circumventing any inconvienient facts.  They can produce a thought arc that goes around and "short-circuts" any argument you could produce.  You are unable to percieve this arc because it lies outside the realm of the rational.  Indeed, it may well reside in some sort of alternate universe that these lucky individuals have tapped into.

So, feel not frustration or pity, and instead fell shame and envy; your brain just isn't as good.
[a href="index.php?act=findpost&pid=277881"][{POST_SNAPBACK}][/a]


That's good. 
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-01 09:07:26
Quote
Rather like this: http://homepage.ntlworld.com/jfe1205/LAME/...XPd2-3.97a7.zip (http://homepage.ntlworld.com/jfe1205/LAME/lamedropXPd2-3.97a7.zip)? A proposed version of lamedropXPd for the next releases of LAME.
[a href="index.php?act=findpost&pid=277887"][{POST_SNAPBACK}][/a]

OK, I've changed a few options, recompiled and uploaded again (on the same link). Can someone who had a problem before tell me if this is any different, please? TIA.
Title: How to convince people to stop MP3 myths
Post by: Synthetic Soul on 2005-03-01 09:31:36
I've just tried on Win2K and can't see anything I'm afraid.

The Ini has been created and the process lamedropXPd.exe is running.
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-01 09:52:48
Quote
I've just tried on Win2K and can't see anything I'm afraid.

The Ini has been created and the process lamedropXPd.exe is running.
[a href="index.php?act=findpost&pid=278138"][{POST_SNAPBACK}][/a]

Thanks for trying. How very odd!! Back to the drawing board, I guess.
Title: How to convince people to stop MP3 myths
Post by: Gabriel on 2005-03-01 11:00:27
Quote
Can someone who had a problem before tell me if this is any different, please?

I can see the interface, but only under XP.
Regarding the configuration UI, I like it.
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-01 11:05:05
Quote
Quote
Can someone who had a problem before tell me if this is any different, please?

I can see the interface, but only under XP.
Regarding the configuration UI, I like it.
[a href="index.php?act=findpost&pid=278163"][{POST_SNAPBACK}][/a]

Thanks, Gabriel. I'll try to figure out why it's invisible other than in XP!!!
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-01 11:08:09
Yep, still nothing on my win2k.
Title: How to convince people to stop MP3 myths
Post by: Lyx on 2005-03-01 11:15:55
It's kinda comforting to see that so many people here are prefering 2k over XP. Not everything is lost yet, it seems ;-)

- Lyx
Title: How to convince people to stop MP3 myths
Post by: Lyx on 2005-03-01 11:35:58
Quote
Quote
Quote
Quote
Is there any reason we would ever really need simple stereo since Lame JS doesn't destroy stereo?[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=278022")


Hrm... multilanguage content?
[a href="index.php?act=findpost&pid=278025"][{POST_SNAPBACK}][/a]


Sorry, I don't understand.
[a href="index.php?act=findpost&pid=278065"][{POST_SNAPBACK}][/a]

Sometimes another language of a program is put in the other channel.

[span style='font-size:8pt;line-height:100%']Discussion about Joint Stereo quality split [a href="http://www.hydrogenaudio.org/forums/index.php?showtopic=31963]here[/url].
- the moderation[/span]
[a href="index.php?act=findpost&pid=278068"][{POST_SNAPBACK}][/a]


Possible solution:
- rename stereo to something like "dual-mono"
- rename joint-stereo to stereo (the switches as well)
- alias the dupe JS switch to stereo(which is now the same)

This would mean that instead of choosing the mp3-stereo-mode, the user chooses the signal-type of the input/output. So, its less a technical choice but more an intuitive choice. People still get their switch with the nice name "stereo" but in reality its now all JS and automatically aliased also when apps use the LAME-dll. People who want to encode two mono-signals (the multi-language example) in one stream, can still access the simple stereo mode, but via a new switch called "dual-mono"(which sounds like bad-quality for the myth-people yet still is meant exactly for what the name implies).

Advantage: the issue which applies to 99,99% of cases (misuse of simple stereo) gets solved while still keeping the simple-stereo switch available for "dual-mono"-encodings.

Disadvantage: the new "dual-mono"-switch breaks backwards-compatibility because it is a new switch. In other words: apps which want to make the simple-stereo/dual-mono switch available need to update their apps. However, this can as well be seen as a good thing: Because of the changed name authors will probably take a closer look at it and possibly re-add the swittch NOT with the label "stereo" but instead "dual-mono".

just my 2 euro-cents,
- Lyx
Title: How to convince people to stop MP3 myths
Post by: kwanbis on 2005-03-01 12:52:51
Quote
I've just tried on Win2K and can't see anything I'm afraid.

The Ini has been created and the process lamedropXPd.exe is running.
[a href="index.php?act=findpost&pid=278138"][{POST_SNAPBACK}][/a]

same here (2K SP4)
Title: How to convince people to stop MP3 myths
Post by: robert on 2005-03-01 13:51:22
@Lyx

if you have to encode bi-lingual content, Lame already has a switch for dual channel encoding: -md. This is a 2 mono chanel at once mode, but unfortunately many players play it as if it were stereo.
Title: How to convince people to stop MP3 myths
Post by: Jojo on 2005-03-01 20:15:53
Quote
I can in some cds detect differences between 320 and original, and i used to believe 128 was cd quality.
[a href="index.php?act=findpost&pid=277993"][{POST_SNAPBACK}][/a]

sure    what encoder and setting did you use?
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-01 20:21:19
Quote
Quote
I can in some cds detect differences between 320 and original, and i used to believe 128 was cd quality.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=277993")

sure 
[a href="index.php?act=findpost&pid=278334"][{POST_SNAPBACK}][/a]


It's actually possible. What if he has a Fatboy Slim CD with Kalifornia?

But then again, qualityequalizer has already been exposed as a [a href="http://www.hydrogenaudio.org/forums/index.php?showtopic=31981&hl=]potential troll[/url], so I would indeed take his claims with lots of salt.
Title: How to convince people to stop MP3 myths
Post by: schonenberg on 2005-03-01 20:39:34
Quote
Quote
I once read that exposure to low quality lossy encodings can cause cumulative hearing loss and tinnitus. While I can not say if that is true, I prefer to err or the side of caution.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=277443")


Don't exist scientific evidence that supports this.. (or at least credible).

This is another new myth that we can add to the list. 
[a href="index.php?act=findpost&pid=277463"][{POST_SNAPBACK}][/a]


That was on slashdot, the great purveyor of myths.
[a href="http://science.slashdot.org/article.pl?sid=02/12/20/2029212]http://science.slashdot.org/article.pl?sid=02/12/20/2029212[/url]

All I could be on google was this:
http://users.informatik.haw-hamburg.de/~wi...r/MP3-risk.html (http://users.informatik.haw-hamburg.de/~windle_c/Logologie/MP3-Gefahr/MP3-risk.html)
Ear damage by MP3, DVD and digital television?
    -risks of neuroacoustically datareduced music
Title: How to convince people to stop MP3 myths
Post by: krabapple on 2005-03-01 22:18:53
Quote
Quote
Quote
I once read that exposure to low quality lossy encodings can cause cumulative hearing loss and tinnitus. While I can not say if that is true, I prefer to err or the side of caution.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=277443")


Don't exist scientific evidence that supports this.. (or at least credible).

This is another new myth that we can add to the list. 
[a href="index.php?act=findpost&pid=277463"][{POST_SNAPBACK}][/a]


That was on slashdot, the great purveyor of myths.
[a href="http://science.slashdot.org/article.pl?sid=02/12/20/2029212]http://science.slashdot.org/article.pl?sid=02/12/20/2029212[/url]

All I could be on google was this:
http://users.informatik.haw-hamburg.de/~wi...r/MP3-risk.html (http://users.informatik.haw-hamburg.de/~windle_c/Logologie/MP3-Gefahr/MP3-risk.html)
Ear damage by MP3, DVD and digital television?
    -risks of neuroacoustically datareduced music
[a href="index.php?act=findpost&pid=278345"][{POST_SNAPBACK}][/a]


I hope everyone reads the replies on Slashdot too...a lot of them point out the , um 'peculiarities' of that
'neuronomy' page and its author.  Looks like the hard experimental data supporting his particular speculations and inferences re mp3, hasn't yet materialized.

Over in audiophile-cuckoo-land, there's a rather famous amp designer (Mark Levinson -- trivia buffs might be interested to know he is the recent ex-husband of  Kim 'Samantha' Cattrall) who has been claiming for some years now that listening to 16/44 digital audio will damage your neuromuscular system.  Hasn't done much for his credibility.

Anyway, hardly a week goes by that I'm not trying to correct some nonsense spouted about mp3s on Usenet or the Web, usually along the lines of 'mp3s can NEVER sound as good as CDs'.  I always end up pointing them to HA.org. 
Title: How to convince people to stop MP3 myths
Post by: Gambit on 2005-03-02 01:08:09
john33, some remarks:

When you switch between bitrate/quality there is no "encoding engine quality"/"variable bitrate mode" selected (it is blank).

What is the difference between "fast" "high" and "standard" in the bitrate mode? And are all those three necessary?

What does the "no gap" option do?
Title: How to convince people to stop MP3 myths
Post by: Acid Orange Juice on 2005-03-02 03:56:51
Quote
mp3 is not transparent.
I can in some cds detect differences between 320 and original.
[a href="index.php?act=findpost&pid=277993"][{POST_SNAPBACK}][/a]


..........

What encoder did you use?

mp3 can be transparent (with exception of some few samples).

This has been confirmed many many times here in several listening tests..

I recommend you to use the SEARCH function.

You have not given any proove (ABX tests) that confirm his arguments.

This makes my not take it to you seriously.
Title: How to convince people to stop MP3 myths
Post by: Acid Orange Juice on 2005-03-02 04:53:30
Quote
That was on slashdot, the great purveyor of myths.
http://science.slashdot.org/article.pl?sid=02/12/20/2029212 (http://science.slashdot.org/article.pl?sid=02/12/20/2029212)
[a href="index.php?act=findpost&pid=278345"][{POST_SNAPBACK}][/a]


This article is incredibly ridiculous.   

I knew that had read it before, but did not remember where.
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-02 08:47:02
Quote
john33, some remarks:

When you switch between bitrate/quality there is no "encoding engine quality"/"variable bitrate mode" selected (it is blank).

When using the 'quality' (VBR) mode, the 'encoding engine quality' is set within the quality presets.
Quote
What is the difference between "fast" "high" and "standard" in the bitrate mode? And are all those three necessary?

These equate to the 'q' settings: fast = 9, high = 2 and standard = 5. (Although I just discovered an error in the code so that this is not currently set correctly!! I'll change that.)
Quote
What does the "no gap" option do?

It's the invocation of the 'nogap' switch that's a standard option.
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-02 09:57:27
(lamedropXPd2) Made a number of changes and fixed a couple of bugs, so can someone kindly tell me if it's still invisible under win2k?, TIA. It's on the same link as before: http://homepage.ntlworld.com/jfe1205/LAME/...XPd2-3.97a7.zip (http://homepage.ntlworld.com/jfe1205/LAME/lamedropXPd2-3.97a7.zip)?
Title: How to convince people to stop MP3 myths
Post by: Gambit on 2005-03-02 10:00:45
Quote
When using the 'quality' (VBR) mode, the 'encoding engine quality' is set within the quality presets.


No, that's not what I mean. Argh, I don't know how to describe it better, I'll post a screenshot when I get home.


Quote
These equate to the 'q' settings: fast = 9, high = 2 and standard = 5. (Although I just discovered an error in the code so that this is not currently set correctly!! I'll change that.)


I'm not really sure it's a good idea to give the user those options. And at least put them in the right order (fast, standard, high). And I'm also not sure if calling q5 "standard" is a good idea, might be a bit confusing, because I get the feeling most people consider q3 the LAME quality standard. But that's more a question of the subjective perception I guess...
Oh, and you mean q3, right? Since the old q2 is now q3...


Quote
It's the invocation of the 'nogap' switch that's a standard option.


You know that that option is pretty much useless and providing it to the user will most probably cause more harm than good.
Title: How to convince people to stop MP3 myths
Post by: vitos on 2005-03-02 10:01:22
Quote
(lamedropXPd2) Made a number of changes and fixed a couple of bugs, so can someone kindly tell me if it's still invisible under win2k?
[a href="index.php?act=findpost&pid=278507"][{POST_SNAPBACK}][/a]

Now it's working under W2k SP4 PL. However I don't know if it was or not visible earlier on my system.
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-02 10:19:00
Quote
Quote
(lamedropXPd2) Made a number of changes and fixed a couple of bugs, so can someone kindly tell me if it's still invisible under win2k?
[a href="index.php?act=findpost&pid=278507"][{POST_SNAPBACK}][/a]

Now it's working under W2k SP4 PL. However I don't know if it was or not visible earlier on my system.
[a href="index.php?act=findpost&pid=278509"][{POST_SNAPBACK}][/a]

Thanks!!!  Hopefully it is for others too.
Title: How to convince people to stop MP3 myths
Post by: Pri3st on 2005-03-02 10:31:55
Visible here, win2k sp2.
Title: How to convince people to stop MP3 myths
Post by: Gabriel on 2005-03-02 10:37:35
Quote
These equate to the 'q' settings: fast = 9, high = 2 and standard = 5. (Although I just discovered an error in the code so that this is not currently set correctly!! I'll change that.)


The standard quality is not 5 with 3.97.
For "standard" quality, I would advise you to simply not pass any quality level to Lame. This way, it will set up itself the default value.

Fast: I think that results should be more acceptables at 7 instead of 9.
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-02 11:05:27
Quote
(lamedropXPd2) Made a number of changes and fixed a couple of bugs, so can someone kindly tell me if it's still invisible under win2k?, TIA. It's on the same link as before: http://homepage.ntlworld.com/jfe1205/LAME/...XPd2-3.97a7.zip (http://homepage.ntlworld.com/jfe1205/LAME/lamedropXPd2-3.97a7.zip)?
[a href="index.php?act=findpost&pid=278507"][{POST_SNAPBACK}][/a]


Excellent. Works on my Win2k now
Title: How to convince people to stop MP3 myths
Post by: Gabriel on 2005-03-02 11:14:54
Perhaps we could split this into a LameDrop topic ?
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-02 11:15:42
@Gambit: I think I follow what you're saying. I've modified it accordingly and uploaded again.

@Gabriel: I've modified it so that 'Standard' passes no value, as you suggest, and 9 was a typo (  ), it was actually 7.

@Roberto: Excellent!! Although I'm not sure what was going on in the previous versions!! 

Thanks to the above, and to others who have responded. Any further comments regarding the newly uploaded version?
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-02 11:23:52
Just for the record, if CBR is selected, the bitrate used will be the standard bitrate nearest to what has been entered in the box.
Title: How to convince people to stop MP3 myths
Post by: Synthetic Soul on 2005-03-02 11:44:50
Quote
Thanks to the above, and to others who have responded. Any further comments regarding the newly uploaded version?

I don't know how "alpha/beta" this is, so please forgive me if I am being pedantic.

Generally, very minor suggestions(/nitpicks), and probably already in hand.  I do think the panel disabling is a reasonable issue though.

Quote
Just for the record, if CBR is selected, the bitrate used will be the standard bitrate nearest to what has been entered in the box.

When the checkbox is selected could the value be amended as soon as it has been changed, to signify the actual bitrate they will attain, in this case?  I.e.: if someone has the checkbox set and enters "127", as soon as they exit the text box the value is amended to "128".
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-02 11:59:58
Quote
You're English sir!  Think what The Queen would say.
[a href="index.php?act=findpost&pid=278531"][{POST_SNAPBACK}][/a]


HRM Queen Elisabeth told me she uses LameDropXPd all the time.
Title: How to convince people to stop MP3 myths
Post by: Synthetic Soul on 2005-03-02 12:25:44
Quote
When the checkbox is selected could the value be amended as soon as it has been changed... yada, yada, yada, yawn, yawn
Sorry, I've just realised that you can't know what the actual bitrate will be until you know what frequency the input file has (e.g.: a file at 24kHz can be encoded at  CBR 144kbps, but a file at 44.1kHz cannot).

Quote
HRM Queen Elisabeth told me she uses LameDropXPd all the time.
Well, she'd be a fool not to.
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-02 18:14:03
Another version available for critique, on the same link, if anyone is interested.
Title: How to convince people to stop MP3 myths
Post by: kwanbis on 2005-03-02 18:24:58
Quote
Thanks!!!   Hopefully it is for others too.

works here now ... just for curiosity, what was it?
Title: How to convince people to stop MP3 myths
Post by: Never_Again on 2005-03-02 18:39:42
The current tagging options are not very useful. There should be an option to derive tags from the filename using custom schemes, e.g. %Artist-%Album-%Track-%Title.
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-02 18:52:47
Quote
works here now ... just for curiosity, what was it?

I wish I new!!  I took it as an opportunity to revamp the code and it was cured as part of that process! 
Quote
     The current tagging options are not very useful. There should be an option to derive tags from the filename using custom schemes, e.g. %Artist-%Album-%Track-%Title.

I agree.  I'll look at using a similar scheme as that used in oggdropXPd before it gets released.
Title: How to convince people to stop MP3 myths
Post by: Synthetic Soul on 2005-03-02 19:57:25
Quote
Another version available for critique, on the same link, if anyone is interested.
If you insist...

I think the disabled group/panel is a massive improvement.  It's now immediately obvious that only one applies.  The proper case labels are also far more aesthetic (I'm sure Mr. Amorim can pass on The Queen's opinion in due course).  The slider in its full glory is a thing of beauty.

The "Encoding Engine Quality" still has no default.  What will it be if left as it stands?  I assume "Standard", but we all know assume made an ass out of u and me...

Nice touch on the CBR bitrate, but I think you may need to consider my rantings below.  Maybe you should just warn them it will be amended to the nearest valid value at the time of encoding.  Or maybe you only accept files at 32kHz or above...
Quote
Quote
When the checkbox is selected could the value be amended as soon as it has been changed... yada, yada, yada, yawn, yawn
Sorry, I've just realised that you can't know what the actual bitrate will be until you know what frequency the input file has (e.g.: a file at 24kHz can be encoded at  CBR 144kbps, but a file at 44.1kHz cannot).

I feel I should point out that I know from experience that it is far easier to make suggestions regarding/criticise an existing object, than to create a perfect object from scratch immediately.  I hope you see my nitpickings simply as constructive criticism for you to consider at your whim.

Who am I kidding?  Do as I tell you John!
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-02 20:18:09
Quote
(I'm sure Mr. Amorim can pass on The Queen's opinion in due course).
[a href="index.php?act=findpost&pid=278630"][{POST_SNAPBACK}][/a]


Her Majesty is very pleased and would like to thank sir John XXXIII for his efforts.

She also said that LameDropXPd can be labelled "By Royal Appointment of HRM Queen Elisabeth II" from now on.

No more.
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-02 21:41:49
Quote
The "Encoding Engine Quality" still has no default.  What will it be if left as it stands?  I assume "Standard", but we all know assume made an ass out of u and me...

Ooops, sorry. It was meant to have a default of 'Standard'. I just forgot to change it. It's done ready for the next version!!
Quote
Nice touch on the CBR bitrate, but I think you may need to consider my rantings below.  Maybe you should just warn them it will be amended to the nearest valid value at the time of encoding.  Or maybe you only accept files at 32kHz or above...

I'll add something to that effect regarding the nearest value.
Quote
I feel I should point out that I know from experience that it is far easier to make suggestions regarding/criticise an existing object, than to create a perfect object from scratch immediately.  I hope you see my nitpickings simply as constructive criticism for you to consider at your whim.

If I hadn't wanted opinions, I wouldn't have asked for them!  Thanks for taking the trouble to respond. 
Quote
Who am I kidding?  Do as I tell you John!

Whatever you say, sir.
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-02 21:42:31
Quote
Quote
(I'm sure Mr. Amorim can pass on The Queen's opinion in due course).
[a href="index.php?act=findpost&pid=278630"][{POST_SNAPBACK}][/a]


Her Majesty is very pleased and would like to thank sir John XXXIII for his efforts.

She also said that LameDropXPd can be labelled "By Royal Appointment of HRM Queen Elisabeth II" from now on.

No more.
[a href="index.php?act=findpost&pid=278636"][{POST_SNAPBACK}][/a]

Actually, she was so pleased she asked me to represent her at the upcoming wedding!!!!!!!
Title: How to convince people to stop MP3 myths
Post by: Gambit on 2005-03-02 23:25:28
The Bitrate "Encoding Engine Quality" ComboBox is still blank by default.

The Bitrate GroupBox is unnecessarily big now.

And, I would vote for only two "Encoding Engine Quality" options for CBR: Fast, High.
Actually, to be absolutely honest, I would only use High with CBR and only let choose Fast/Standard for VBR.
Title: How to convince people to stop MP3 myths
Post by: vitos on 2005-03-02 23:47:55
Quote
And, I would vote for only two "Encoding Engine Quality" options for CBR: Fast, High.
Actually, to be absolutely honest, I would only use High with CBR and only let choose Fast/Standard for VBR.
[a href="index.php?act=findpost&pid=278717"][{POST_SNAPBACK}][/a]

And that would be understood as: fast - low quality, standard - high. So far we all know, that "fast" aka "vbr-new" doesn't mean lower quailty, it's sometimes even better.
Title: How to convince people to stop MP3 myths
Post by: Gambit on 2005-03-03 00:07:36
Quote
Quote
And, I would vote for only two "Encoding Engine Quality" options for CBR: Fast, High.
Actually, to be absolutely honest, I would only use High with CBR and only let choose Fast/Standard for VBR.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=278717")

And that would be understood as: fast - low quality, standard - high. So far we all know, that "fast" aka "vbr-new" doesn't mean lower quailty, it's sometimes even better.
[a href="index.php?act=findpost&pid=278724"][{POST_SNAPBACK}][/a]


Well, you are selecting the "Variable Bitrate Mode" and not "quality" for VBR, but I agree that newbies might be confused by it. A little note explaining the difference would be prolly appropriate.

Oh, and a little cosmetic bug:
[a href="http://www.burrrn.net/TehBukkake/ScreenshotForSirJohn.png]http://www.burrrn.net/TehBukkake/ScreenshotForSirJohn.png[/url]
Title: How to convince people to stop MP3 myths
Post by: Jojo on 2005-03-03 01:38:30
I don't like that LameDrop interface at all...if I was a newbie I'd come to the conclusion that there is a huge difference between 'standard' (70) and 'insanse' (100)...however, we all know that this is not true and that in most cases --preset standard is already transparent and that many people cannot even hear the difference between --preset standard and the original CD on the so called 'problem samples'....I mean it's not like some JPEG where you can see the difference between quality 100 and 70...
Title: How to convince people to stop MP3 myths
Post by: Madrigal on 2005-03-03 02:45:57
Quote
Quote
   The current tagging options are not very useful. There should be an option to derive tags from the filename using custom schemes, e.g. %Artist-%Album-%Track-%Title.

I agree.  I'll look at using a similar scheme as that used in oggdropXPd before it gets released.
[a href="index.php?act=findpost&pid=278610"][{POST_SNAPBACK}][/a]
Please be sure to include info from foldernames as part of the scheme, e.g.:

%Artist\%Year - %Album\%Track - %Title

Regards,
Madrigal
Title: How to convince people to stop MP3 myths
Post by: smz on 2005-03-03 03:00:48
Quote
Perhaps we could split this into a LameDrop topic ?
[a href="index.php?act=findpost&pid=278520"][{POST_SNAPBACK}][/a]

It seems highly appropriate...
Title: How to convince people to stop MP3 myths
Post by: Gabriel on 2005-03-03 09:29:16
Quote
I don't like that LameDrop interface at all...if I was a newbie I'd come to the conclusion that there is a huge difference between 'standard' (70) and 'insanse' (100)...however, we all know that this is not true

Standard is 80 not 70.
But anyway there is a huge difference between 80 and 100. The difference is about 2 steps on a 10 points scale. The fact that 80 is already providing transparent encoding most of the time for most people does not means that higer settins should be hidden.


Regarding the interface, I am not sure if the nogap option is really usefull. If you already call the correct function at the end of encoding, the Lame header will already include the delay and padding values that can be used for a true gapless playback.
Title: How to convince people to stop MP3 myths
Post by: PatchWorKs on 2005-03-03 11:17:27
Te be serious: in my thesis i'll try to demonstrate that we don't need stereo for evaluation purposes that, from my point of view, is a very similar problem.
Here are some interesting documents:

Sound Systems: Mono versus Stereo (http://www.mcsquared.com/mono-stereo.htm)
Stereo vs  Mono (http://www.bcae1.com/stermono.htm)
"You Can't Do That": The Great Stereo vs. Mono vs. Re-mixing Debate. (http://www.geocities.com/joelcrowservo/stereomono.html)
FAQ on recording a live show in Mono vs Stereo (http://www.soundprofessionals.com/cgi-bin/gold/category.cgi?category=liveshowinmono)
FAQ on Binaural vs. Stereo recordings (http://www.soundprofessionals.com/cgi-bin/gold/category.cgi?category=binauralcardioid)
Stereo vs. Mono: Compare Values (http://www.pbs.org/wgbh/pages/roadshow/series/highlights/2004/chicago/follow1_2.html)
Mono, Stereo, Digital: The Three Ages of Man (http://www.stereophile.com/asweseeit/706/index.html)

and finally: Two ears are better than one (http://www.phonak.com/com_028-0414-02_two_ears_are_better_than_one.pdf) (pdf from Phonak, one of the world’s leading manufacturers of hearing systems  )
Title: How to convince people to stop MP3 myths
Post by: Synthetic Soul on 2005-03-03 11:24:51
I have mocked up a "suggested alternative" to the Encoding Options dialogue.

Basic Interface (http://www.neilpopham.pwp.blueyonder.co.uk/temp/encoding-full.gif) | Quality radio button selected (http://www.neilpopham.pwp.blueyonder.co.uk/temp/encoding-quality.gif) | Bitrate radio button selected (http://www.neilpopham.pwp.blueyonder.co.uk/temp/encoding-bitrate.gif)

The Quality radio button would be selected by default, as this is the preferred target method.

I think this layout makes more sense as there is a tangible relationship between the Quality radio button and the quality settings, and the Bitrate radio button and the bitrate settings.

[span style='font-size:8pt;line-height:100%']NB: I know the slider tick marks don't add up, but I've never built a dialogue in VC6 before and can only assume that you need to set the max value programatically.  I think you get the idea.  I used VC6 so I could pass any files on if required... and if someone tells me how.[/span]

Edit:

Added Tagging Options (Custom) (http://www.neilpopham.pwp.blueyonder.co.uk/temp/tagging-custom.gif) | Tagging Options (Default) (http://www.neilpopham.pwp.blueyonder.co.uk/temp/tagging-default.gif) | Output Directory (http://www.neilpopham.pwp.blueyonder.co.uk/temp/output-directory.gif)

The Filename Mask is a suggested solution to Never Again's request "to derive tags from the filename using custom schemes".  The "+" button would be used to add new masks, which would then be selectable from the drop down.  The "Create Mask" dialogue would explain what tags can be used.  An example mask would be:

%artist%\%album%\%tracknumber% - %title%
Title: How to convince people to stop MP3 myths
Post by: PatchWorKs on 2005-03-03 11:40:48
Blanka, which is your site ?

Joint Stereo: The Myths ... and The Realities (http://harmsy.freeuk.com/mostync/)
sonic:compression - myths (http://xmixahlx.com/oldsites/soniccompression/myths.html)
HQscene Standards Encoding Guide v3.01 (http://hqscene.fatty.dk/)

Anyway, listen to me: drop MP3, switch to Vorbis !!!

(http://www.rjamorim.com/test/multiformat128/plot18z.png)
Title: How to convince people to stop MP3 myths
Post by: Gambit on 2005-03-03 13:06:19
Quote
Regarding the interface, I am not sure if the nogap option is really usefull. If you already call the correct function at the end of encoding, the Lame header will already include the delay and padding values that can be used for a true gapless playback.


john33 already removed that checkbox.
Title: How to convince people to stop MP3 myths
Post by: guruboolez on 2005-03-03 13:24:33
Quote
I don't like that LameDrop interface at all...if I was a newbie I'd come to the conclusion that there is a huge difference between 'standard' (70) and 'insanse' (100)...however, we all know that this is not true and that in most cases --preset standard is already transparent and that many people cannot even hear the difference between --preset standard and the original CD on the so called 'problem samples'....I mean it's not like some JPEG where you can see the difference between quality 100 and 70...
[a href="index.php?act=findpost&pid=278740"][{POST_SNAPBACK}][/a]

And? Must any software match all newbies misconceptions? What about vorbis (and OggDrop), which offers -q10 mode, as transparent as -q4 for a lot of newbies as well?
LAME offers -V0, V1 & V2 mode. A complete GUI should offers the full VBR range, even if most people are happy with lower VBR settings.
Moreover, there are audible difference between -V2 and -V0. Not always, but on some critical parts. -V2 is maybe transparent to you, but not to everybody.

BTW, nice GUI John33. Is it possible to add lossless format as input?
Title: How to convince people to stop MP3 myths
Post by: Busemann on 2005-03-03 13:32:11
Quote
Anyway, listen to me: drop MP3, switch to Vorbis !!!
[a href="index.php?act=findpost&pid=278843"][{POST_SNAPBACK}][/a]


..if you encode at 128kbps 
Title: How to convince people to stop MP3 myths
Post by: john33 on 2005-03-03 14:44:17
Thanks to everybody for their input. I'll take it all into consideration when producing the final pre-release.

Part of the reason for doing this now was that, dare I say it  , LAME 4.0 looks like it is not too far away and I wanted to have something in place for its arrival. Having said that, though, it has fewer encoding options so will necessarily look a little different.
Title: How to convince people to stop MP3 myths
Post by: Jojo on 2005-03-03 15:57:50
Quote
Quote
I don't like that LameDrop interface at all...if I was a newbie I'd come to the conclusion that there is a huge difference between 'standard' (70) and 'insanse' (100)...however, we all know that this is not true and that in most cases --preset standard is already transparent and that many people cannot even hear the difference between --preset standard and the original CD on the so called 'problem samples'....I mean it's not like some JPEG where you can see the difference between quality 100 and 70...
[a href="index.php?act=findpost&pid=278740"][{POST_SNAPBACK}][/a]

And? Must any software match all newbies misconceptions? What about vorbis (and OggDrop), which offers -q10 mode, as transparent as -q4 for a lot of newbies as well?
LAME offers -V0, V1 & V2 mode. A complete GUI should offers the full VBR range, even if most people are happy with lower VBR settings.
Moreover, there are audible difference between -V2 and -V0. Not always, but on some critical parts. -V2 is maybe transparent to you, but not to everybody.

BTW, nice GUI John33. Is it possible to add lossless format as input?
[a href="index.php?act=findpost&pid=278864"][{POST_SNAPBACK}][/a]

I didn't say that it should be hidden...but maybe it could be reworked...how about setting --preset standard as default or write VBR 80 in bold letters...
Title: How to convince people to stop MP3 myths
Post by: guruboolez on 2005-03-03 16:02:16
Quote
how about setting --preset standard as default
[a href="index.php?act=findpost&pid=278899"][{POST_SNAPBACK}][/a]

Why by default? It implies that most lame users would use this encoder for transparent and ~200kbps encodings. Is it true? I don't know... But as far as I know, lame encode at 128 kbps by default, which is probably something more popular (widely used by most people).
Title: How to convince people to stop MP3 myths
Post by: Gabriel on 2005-03-03 16:31:40
To me the default (if there is one) should be 128kbps for bitrate based encoding (and "restrict to cbr encoding" disabled), and quality 50 for quality based encoding.

Most people DO NOT WANT 200kbps encoding, they want 128kbps encodings. Those who want more are already knowledgeable enough to change the values themselve.
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-03 16:47:19
If the default mode creates files near 200kbps, the average user will say "eww, LAME sucks, it compresses badly" and will go back to his MusicMatch at 128kbps.

So, here's another vote to keep the defaults at 128CBR/5VBR. I'm tired of this popular attitude around here of "let's try to force the users to use what WE think is best for them".
Title: How to convince people to stop MP3 myths
Post by: kwanbis on 2005-03-03 17:45:32
Quote
Thanks to everybody for their input. I'll take it all into consideration when producing the final pre-release.

Part of the reason for doing this now was that, dare I say it   , LAME 4.0 looks like it is not too far away and I wanted to have something in place for its arrival. Having said that, though, it has fewer encoding options so will necessarily look a little different.
[a href="index.php?act=findpost&pid=278885"][{POST_SNAPBACK}][/a]

john ,the "Encoding Engine Quality" and "Quality" are confusing ... they seem to overlap
Title: How to convince people to stop MP3 myths
Post by: Jojo on 2005-03-03 18:38:31
Quote
So, here's another vote to keep the defaults at 128CBR/5VBR. I'm tired of this popular attitude around here of "let's try to force the users to use what WE think is best for them".
[a href="index.php?act=findpost&pid=278916"][{POST_SNAPBACK}][/a]

in that case there shouldn't be any default at all...
Title: How to convince people to stop MP3 myths
Post by: Busemann on 2005-03-03 18:43:13
Quote
If the default mode creates files near 200kbps, the average user will say "eww, LAME sucks, it compresses badly" and will go back to his MusicMatch at 128kbps.


I don't think that's true for everyone. I have lots of friends who want "cd quality", but have used 128kbps which is the default, and therefore think mp3 is bad.

Why not have two "defaults"; one for small size & one for transparency?
Title: How to convince people to stop MP3 myths
Post by: Gambit on 2005-03-03 18:55:22
Quote
If the default mode creates files near 200kbps, the average user will say "eww, LAME sucks, it compresses badly" and will go back to his MusicMatch at 128kbps.

So, here's another vote to keep the defaults at 128CBR/5VBR. I'm tired of this popular attitude around here of "let's try to force the users to use what WE think is best for them".
[a href="index.php?act=findpost&pid=278916"][{POST_SNAPBACK}][/a]


I don't agree. Why not 96 or 160? Because 128 is the "standard"? Who set that standard? And why should we follow it? I don't know ONE knowledgeable person that would use 128 for their encodings. Only (or most) people that use 128 do so because they don't know better.
Title: How to convince people to stop MP3 myths
Post by: Benjamin Lebsanft on 2005-03-03 19:00:21
I would vote for --preset standard as, well, standard ^^
Title: How to convince people to stop MP3 myths
Post by: Gambit on 2005-03-03 19:01:55
To make it short: 128kbps CBR mp3 is OBSOLETE.
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-03 19:17:51
Quote
Who set that standard?
[a href="index.php?act=findpost&pid=278963"][{POST_SNAPBACK}][/a]


years of marketing and brainwashing.

Quote
And why should we follow it?


Nobody is telling you to follow it.

Quote
I don't know ONE knowledgable person that would use 128 for their encodings.


I know several that are very happy with it.

Quote
To make it short: 128kbps CBR mp3 is OBSOLETE.

[a href="index.php?act=findpost&pid=278965"][{POST_SNAPBACK}][/a]


--alt-preset standard as default is ARROGANT.
Title: How to convince people to stop MP3 myths
Post by: Lyx on 2005-03-03 19:31:35
*votes for v4/preset medium as the default, because thats the "transparency"-setting meant for mortals*

Using 128kbit CBR as the default is like - well, if we do the same crap as everyone else just because everyone else does it, then why invent another frontend anyways. Yes, 128kbit CBR(if anything then at least make it ABR) is crap and making it the default is de-evolution.

- Lyx
Title: How to convince people to stop MP3 myths
Post by: Brink on 2005-03-03 19:32:19
People want 1mb-1min. I dont think setting preset standard as default is a good idea.
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-03 19:34:44
Quote
*votes for v4/preset medium as the default, because thats the "transparency"-setting meant for mortals*
[a href="index.php?act=findpost&pid=278971"][{POST_SNAPBACK}][/a]


I completely agree with starting the program in VBR mode at V4 or V5. But once the user hits the CBR radio button, for whatever reason he deems appropriate, I believe it should default to 128.
Title: How to convince people to stop MP3 myths
Post by: sTisTi on 2005-03-03 19:47:41
another vote for V4/preset medium; best quality/file size ratio IMO. Transparent for most (normal ) people, but no complete overkill for portable use.
Title: How to convince people to stop MP3 myths
Post by: Synthetic Soul on 2005-03-03 20:04:40
Another vote here for a default Quality setting of 50, and a default Bitrate setting of 128.

I think if a new user sees "128" on the screen they may feel more comfortable that they are in the right place, as "128" and "MP3" seem synonymous in many places.

With that in mind I can see many unknowing users switching to the Bitrate option, because of the "128", and not understanding what a scale of 10-100 means.  However, you can't hold their hand all the way.  At least an ABR 128kbps MP3 will match their expectations (filesize/quality), and perhaps surpass them.
Title: How to convince people to stop MP3 myths
Post by: Busemann on 2005-03-03 20:12:56
Quote
I think if a new user sees "128" on the screen they may feel more comfortable that they are in the right place, as "128" and "MP3" seem synonymous in many places.


I hardly think anyone knowledgeable enough to download LAME thinks like that...
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-03 20:18:05
Quote
I hardly think anyone knowledgeable enough to download LAME thinks like that...
[a href="index.php?act=findpost&pid=278991"][{POST_SNAPBACK}][/a]


What lead you to believe only 31337 audiophiles are after LAME?

There are countless guides around the net targeting newbies and pointing them at LAME.
Title: How to convince people to stop MP3 myths
Post by: Busemann on 2005-03-03 20:20:17
Quote
There are countless guides around the net targeting newbies and pointing them at LAME.
[a href="index.php?act=findpost&pid=278994"][{POST_SNAPBACK}][/a]


..guides for users seeking quality, then?
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-03 20:26:56
Quote
..guides for users seeking quality, then?
[a href="index.php?act=findpost&pid=278998"][{POST_SNAPBACK}][/a]


Quality can come at 128kbps ABR/VBR for lots of people.
Title: How to convince people to stop MP3 myths
Post by: Synthetic Soul on 2005-03-03 21:42:56
Quote
Quote
I think if a new user sees "128" on the screen they may feel more comfortable that they are in the right place, as "128" and "MP3" seem synonymous in many places.
I hardly think anyone knowledgeable enough to download LAME thinks like that...

Sorry, I thought we were discussing LamedropXPd. 

How knowledgeable do you have to be to click on a hyperlink and open a zip file?
Title: How to convince people to stop MP3 myths
Post by: vitos on 2005-03-03 21:47:12
I second what rjamorim says. Really, stop fooling yourself that everyone needs APS. For most people today MP3 equals 128k or 192k (unfortunately maybe because of many illegal releases coded at this bitrate).
When someone buys MP3 player today. the first thing that matters is size - it's written on it that it should keep a number of hours or tracks, based on default 128kbps bitrate. So unexperienced people believe and expect that. Many of them don't now it can sound better (maybe even not because of the compression artifacts, but hardware is a crap). What's more sad - even when you show them what MP3 at that bitrate does to music, and they can hear it (many actually can) - they don't care! At least that's from my own observations... 

For those, who want achieve more - they would be interested enough to get known with --preset standard.

What I am opposed to is CBR. When user wants to enable it, frontend should inform that this mode is obsolete an not recommended, present practically only for compatibility with some prehistorical hardware players. I think 99% of hardware can play VBR without any problem (am I wrong?). Then I would vote for 192kbps as default bitrate for CBR.
Title: How to convince people to stop MP3 myths
Post by: ChiGung on 2005-03-03 23:52:51
Quote
I think 99% of hardware can play VBR without any problem (am I wrong?). Then I would vote for 192kbps as default bitrate for CBR.
[a href="index.php?act=findpost&pid=279029"][{POST_SNAPBACK}][/a]

My new one doesnt like vbr - playtime flickers, its a matchbox sized 256MB flash player with radio and record function and cheap (for a first worlder). I prefer cbr really, its neat and can be easier on batteries. I think abr looks silly, something dishonest about abr 128 or 160 etc - those are the CBR rates, ABR should be honest and aim for 100, 125, 150 instead of trying and failing to decieve (they always come out +/- a few kbs either way.
I think 128 CBR is more than fine for most people, being commercial musicmatch etc wouldnt use it if it didnt satisfy the majority of users, if you need more quality, use the audiophile preset standards or lossless.

Its enough to avoid tape hiss, deterioration and get tiny gadgets. General users need not encumber themselves with the same requirements and filesizes as developers and enthusiasts.
Title: How to convince people to stop MP3 myths
Post by: Jojo on 2005-03-04 00:01:14
Quote
I prefer cbr really, its neat and can be easier on batteries.
[a href="index.php?act=findpost&pid=279049"][{POST_SNAPBACK}][/a]

troll? what mp3 player to you have? Besides that, it doesn't matter if a mp3 is encoded in CBR or VBR, the decoding power needed is the same and therefore it won't affect battery...
Title: How to convince people to stop MP3 myths
Post by: ChiGung on 2005-03-04 00:15:38
Quote
troll? what mp3 player to you have? Besides that, it doesn't matter if a mp3 is encoded in CBR or VBR, the decoding power needed is the same and therefore it won't affect battery...
[a href="index.php?act=findpost&pid=279052"][{POST_SNAPBACK}][/a]

Its a panram music yoyo. I use rechargeables which last pretty long, but it runs down quicker when recording in high bitrates. I havent tested playback, but I think its quite possible it will use a higher power cpu mode to playback higher bitrate mp3s and therefore abr and vbr as they contain unpredictable stretches of high bitrate frames. Its just a hunch.

You can observe that low bitrate mp3 can take less processing power to playback by looking in taskmanager 
Title: How to convince people to stop MP3 myths
Post by: music_man_mpc on 2005-03-04 00:39:06
Quote
I havent tested playback, but I think its quite possible it will use a higher power cpu mode to playback higher bitrate mp3s and therefore abr and vbr as they contain unpredictable stretches of high bitrate frames. Its just a hunch.
[a href="index.php?act=findpost&pid=279055"][{POST_SNAPBACK}][/a]

Thats a pretty strange hunch, especially when considering that CBR files can use frames as low as 32kb/s and as high as 320kb/s due to the bit resevoir.

edit:  Perhaps you could test your hypothesis before posting next time.
Title: How to convince people to stop MP3 myths
Post by: ChiGung on 2005-03-04 01:03:47
Quote
Thats a pretty strange hunch, especially when considering that CBR files can use frames as low as 32kb/s and as high as 320kb/s due to the bit resevoir.

edit:  Perhaps you could test your hypothesis before posting next time.
[a href="index.php?act=findpost&pid=279059"][{POST_SNAPBACK}][/a]

However the bit reservoir is limited in size (predictable) whereas the length of a stretch of high bitrate frames are not.

Quote
You can observe that low bitrate mp3 can take less processing power to playback by looking in taskmanager 

You can do this as well, test your own hypothesis 
Title: How to convince people to stop MP3 myths
Post by: Never_Again on 2005-03-04 01:16:58
So you come up with "hunches" but are too lazy to test them yourself?
Title: How to convince people to stop MP3 myths
Post by: ChiGung on 2005-03-04 02:18:35
Quote
So you come up with "hunches" but are too lazy to test them yourself?
[a href="index.php?act=findpost&pid=279070"][{POST_SNAPBACK}][/a]

Too lazy is a rather perjorative term, but considering id have to refill my player and run it for days to be sure, its not worth it for me. In this area my hunch is modestly informed. Its very easy to observe that low bitrates *do* use less processing power (winamp+taskmaster) and it follows that VBR *could* use more than CBR than VBR with the understanding that cheap firmware which messes up playtime of VBR files is also liable to stay in the powermode required to decode runs of higher bitrates, which it would not be presented with cbrs limited size and limited bit reservoir.
Not so far from the subject ive read of other players which last longer playing mp3s than aac.
In the end, its personal preference. I suggest folks should use the settings they prefer and am interested to hear individual outlooks.
Title: How to convince people to stop MP3 myths
Post by: music_man_mpc on 2005-03-04 06:32:24
Quote
Quote
You can observe that low bitrate mp3 can take less processing power to playback by looking in taskmanager 

You can do this as well, test your own hypothesis 
[a href="index.php?act=findpost&pid=279065"][{POST_SNAPBACK}][/a]

You made the hypothesis, not me.  I just said that it sounded strange and asked that you test such ideas before posting here.  As far as testing this in taskmanager goes, my computer is not slow enough for that or do you expect me to time the intervals in which CPU usage is 1% and 0%? 
Title: How to convince people to stop MP3 myths
Post by: ChiGung on 2005-03-04 08:27:54
Your hypothesis was that vbr mp3s couldnt induce higher power usage because of the existance of the bit reservoir.
One of mine (further explained to you and co) was that it could.
The testing of one mp3 player could not disprove my hypothesis, but with explaination I have informed it -tho evidently not your good self.
I will be dissapointed if your objection to my not backing up such rudimentary insights into mp3 decoding power with quoted test results is born out by the forum rules and in the meantime will proceed with my current understanding of their application.
Title: How to convince people to stop MP3 myths
Post by: uart on 2005-03-04 14:43:36
Quote
Quote
I prefer cbr really, its neat and can be easier on batteries.
[a href="index.php?act=findpost&pid=279049"][{POST_SNAPBACK}][/a]

troll?
[a href="index.php?act=findpost&pid=279052"][{POST_SNAPBACK}][/a]


It didn't look like trolling to me.
Title: How to convince people to stop MP3 myths
Post by: Jojo on 2005-03-04 16:04:57
Quote
Quote
troll? what mp3 player to you have? Besides that, it doesn't matter if a mp3 is encoded in CBR or VBR, the decoding power needed is the same and therefore it won't affect battery...
[a href="index.php?act=findpost&pid=279052"][{POST_SNAPBACK}][/a]

Its a panram music yoyo. I use rechargeables which last pretty long, but it runs down quicker when recording in high bitrates. I havent tested playback, but I think its quite possible it will use a higher power cpu mode to playback higher bitrate mp3s and therefore abr and vbr as they contain unpredictable stretches of high bitrate frames. Its just a hunch.

You can observe that low bitrate mp3 can take less processing power to playback by looking in taskmanager 
[a href="index.php?act=findpost&pid=279055"][{POST_SNAPBACK}][/a]

completely nonsense...I can have a CBR file with a bitrate of 320kbps and a VBR file that averages at 50kbps...also, VBR files might have 320kbps frames, but can also have 128kbps ones...

Anyway, I've tested that with an iPod, which shows the CPU rate in real time...and it doesn't make any difference what bitrate that files has or if it is ABR, VBR or CBR. Therefore, if you have a CBR 192kbps file or a VBR file that with the same average bitrate, there will not be any difference...

Quote
Its very easy to observe that low bitrates *do* use less processing power (winamp+taskmaster) and it follows that VBR *could* use more than CBR than VBR with the understanding that cheap firmware which messes up playtime of VBR files is also liable to stay in the powermode required to decode runs of higher bitrates, which it would not be presented with cbrs limited size and limited bit reservoir.
Not so far from the subject ive read of other players which last longer playing mp3s than aac.
In the end, its personal preference. I suggest folks should use the settings they prefer and am interested to hear individual outlooks.
[a href="index.php?act=findpost&pid=279080"][{POST_SNAPBACK}][/a]

wrong again...bitrates have nothing to with CPU power needed to decode the file...it is true that AAC needs more CPU power than mp3. But again, that has nothing to do with bitrates. Even a 32kbps AAC file needs more CPU power for decoding than mp3 CBR 320kbps or any other possible mp3 setting...
Title: How to convince people to stop MP3 myths
Post by: Jojo on 2005-03-04 16:24:42
Quote
Quote
*votes for v4/preset medium as the default, because thats the "transparency"-setting meant for mortals*
[a href="index.php?act=findpost&pid=278971"][{POST_SNAPBACK}][/a]


I completely agree with starting the program in VBR mode at V4 or V5. But once the user hits the CBR radio button, for whatever reason he deems appropriate, I believe it should default to 128.
[a href="index.php?act=findpost&pid=278974"][{POST_SNAPBACK}][/a]

ok, -V4 seems like a good tradeoff...so it would use that as a default setting. I also like the idea to jump to 128kbps if a user selects CBR.

However, I think there should be a little area in LAMEdrop that gives some details about bitrate and stuff about the current selected mode to not leave the user in the dark:

-V5: ~128kbps, for portable use
-V4: ~165kbps, transparent on portables
-V3: ~175kbps, transparent for most people
-V2: ~193kbps, transparent except for a few samples (recommended)
-V1: ~210kbps, transparent except for a few samples, slightly bloated
-V0: ~240kbps, transparent except for a few samples, bloated

and maybe there should be another box on the very top, that gives some details about the selected mode (ABR, VBR, CBR). 1 or 2 sentences would be enough to explain the advantages / disadvantages of the selected method...
Title: How to convince people to stop MP3 myths
Post by: Gabriel on 2005-03-04 16:26:15
note: without internal knowledge, you can never assume that a player/decoder is optimally programmed/designed.
Title: How to convince people to stop MP3 myths
Post by: Gabriel on 2005-03-04 16:28:34
Quote
and maybe there should be another box on the very top, that gives some details about the selected mode (ABR, VBR, CBR). 1 or 2 sentences would be enough to explain the advantages / disadvantages of the selected method...

I think that the current interface is clear enough.
The average user doesn't care what is the difference between CBR and ABR, and does not even care about the meaning of ABR and the likes.
Title: How to convince people to stop MP3 myths
Post by: uart on 2005-03-04 16:31:38
I can clearly remember when I first switched from using CBR-128 to variable bit rate mp3s (quite a few years back on my old k6-2) that I noticed a definite increase in winamp CPU usage. I cant remember the exact figures but I think it jumped from something like 3% up to about 5%. The problem with this observation is that the VBR settings I was using at the time were averaging about 160kbps, so I'm not sure how much was contributed by the VBR versus cbr and how much was simply due to the higher bit rate.

The truth is that it probably depends totally on the implementation of the decoder as to whether there are any penalties in decoding VBR, it may well be the case with some implemantations but not others. Perhaps Gabriel or one of the other developer could comment on whether there is likely to be a generally applicable decding penatly for VBR over CBR.
Title: How to convince people to stop MP3 myths
Post by: Busemann on 2005-03-04 16:55:08
Quote
completely nonsense...I can have a CBR file with a bitrate of 320kbps and a VBR file that averages at 50kbps...also, VBR files might have 320kbps frames, but can also have 128kbps ones...

Anyway, I've tested that with an iPod, which shows the CPU rate in real time...and it doesn't make any difference what bitrate that files has or if it is ABR, VBR or CBR. Therefore, if you have a CBR 192kbps file or a VBR file that with the same average bitrate, there will not be any difference...


I would take that cpu meter hack for for what it is.. It's not the holy grail for decoding complexity or something like that, so his vbr files could very well take more power than low-bitrate cbr files on his device.

Also remember that the iPod has two cpu's that works independently, so what that single number means is a bit unclear I think.
Title: How to convince people to stop MP3 myths
Post by: Jojo on 2005-03-04 17:22:59
Quote
Quote
completely nonsense...I can have a CBR file with a bitrate of 320kbps and a VBR file that averages at 50kbps...also, VBR files might have 320kbps frames, but can also have 128kbps ones...

Anyway, I've tested that with an iPod, which shows the CPU rate in real time...and it doesn't make any difference what bitrate that files has or if it is ABR, VBR or CBR. Therefore, if you have a CBR 192kbps file or a VBR file that with the same average bitrate, there will not be any difference...


I would take that cpu meter hack for for what it is.. It's not the holy grail for decoding complexity or something like that, so his vbr files could very well take more power than low-bitrate cbr files on his device.

Also remember that the iPod has two cpu's that works independently, so what that single number means is a bit unclear I think.
[a href="index.php?act=findpost&pid=279221"][{POST_SNAPBACK}][/a]

well, I wouldn't call it a hack, but rather a tool the developers use as well (otherwise it wouldn't be so easy to access). It seems pretty accurate though...when loading new files or when having it connected to the USB plug it shows 80*. If the iPod is idle it shows 20*. If a song is playing the number varies a bit and switches every sencond or so...the range is about 3. AAC takes a higher number than mp3 and as a result to that the battery is drained faster.

After a firmware update a while ago I wondered why my battery was dead after 4 hours. I checked the _show_speed info and it revealed that the processor was running up high at 80* all the time. I restored the firmware completly which solved the problem and the battery life was back to normal...so I think _show_speed is pretty accurate. It is true that other iPods show different numbers, but that's because they use a different CPU. The numbers is not what's interesting, but rather their ranges...

*All numbers are based on iPod mini.
Title: How to convince people to stop MP3 myths
Post by: Busemann on 2005-03-04 17:25:59
Quote
Quote
Quote
completely nonsense...I can have a CBR file with a bitrate of 320kbps and a VBR file that averages at 50kbps...also, VBR files might have 320kbps frames, but can also have 128kbps ones...

Anyway, I've tested that with an iPod, which shows the CPU rate in real time...and it doesn't make any difference what bitrate that files has or if it is ABR, VBR or CBR. Therefore, if you have a CBR 192kbps file or a VBR file that with the same average bitrate, there will not be any difference...


I would take that cpu meter hack for for what it is.. It's not the holy grail for decoding complexity or something like that, so his vbr files could very well take more power than low-bitrate cbr files on his device.

Also remember that the iPod has two cpu's that works independently, so what that single number means is a bit unclear I think.
[a href="index.php?act=findpost&pid=279221"][{POST_SNAPBACK}][/a]

well, I wouldn't call it a hack, but rather a tool the developers use as well (otherwise it wouldn't be so easy to access). It seems pretty accurate though...when loading new files or when having it connected to the USB plug it shows 80*. If the iPod is idle it shows 20*. If a song is playing the number varies a bit and switches every sencond or so...the range is about 3. AAC takes a higher number than mp3 and as a result to that the battery is drained faster.

After a firmware update a while ago I wondered why my battery was dead after 4 hours. I checked the _show_speed info and it revealed that the processor was running up high at 80* all the time. I restored the firmware completly which solved the problem and the battery life was back to normal...so I think _show_speed is pretty accurate. It is true that other iPods show different numbers, but that's because they use a different CPU. The numbers is not what's interesting, but rather their ranges...

*All numbers are based on iPod mini.
[a href="index.php?act=findpost&pid=279228"][{POST_SNAPBACK}][/a]


Sounds like the _show_speed number is mW usage then.
Title: How to convince people to stop MP3 myths
Post by: guruboolez on 2005-03-04 18:12:55
Quote
Quote
Its very easy to observe that low bitrates *do* use less processing power (winamp+taskmaster) and it follows that VBR *could* use more than CBR than VBR with the understanding that cheap firmware which messes up playtime of VBR files is also liable to stay in the powermode required to decode runs of higher bitrates, which it would not be presented with cbrs limited size and limited bit reservoir.
Not so far from the subject ive read of other players which last longer playing mp3s than aac.
In the end, its personal preference. I suggest folks should use the settings they prefer and am interested to hear individual outlooks.
[a href="index.php?act=findpost&pid=279080"][{POST_SNAPBACK}][/a]

wrong again...bitrates have nothing to with CPU power needed to decode the file
[a href="index.php?act=findpost&pid=279202"][{POST_SNAPBACK}][/a]


Try with foobar2000 and its diskwriter (with buffering): 320 kbps MP3 encoding needs more CPU power to be decoded than 128 kbps MP3 encoding.
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-04 18:17:46
Quote
Quote

wrong again...bitrates have nothing to with CPU power needed to decode the file
[a href="index.php?act=findpost&pid=279202"][{POST_SNAPBACK}][/a]


Try with foobar2000 and its diskwriter (with buffering): 320 kbps MP3 encoding needs more CPU power to be decoded than 128 kbps MP3 encoding.
[a href="index.php?act=findpost&pid=279245"][{POST_SNAPBACK}][/a]


Jojo is too fast judging stuff he obviously doesn't know anything about.
Title: How to convince people to stop MP3 myths
Post by: Never_Again on 2005-03-04 19:39:27
Well, if you know about this stuff why don't explain it then, instead of judging Jojo?
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-04 19:49:47
Quote
Well, if you know about this stuff why don't explain it then, instead of judging Jojo?
[a href="index.php?act=findpost&pid=279271"][{POST_SNAPBACK}][/a]


I'd rather shut up than make a fool out of myself like Jojo just did, accusing ChiGung of being a troll out of his own ignorance.

And, in case you are having a very hard time figuring it out: I'm not judging Jojo. I just made a remark about his behaviour in this thread.
Title: How to convince people to stop MP3 myths
Post by: ChiGung on 2005-03-04 20:59:52
No probs - My comments combined, with a contrary liking of cbr and poking fun at abr usage a bit. i like to thrash it out, and need corrected sometimes. I strung out my take on the overuse of the 'test or dont disagree' mantra. Disagreements can really stimulating especialy for occasional forum junkies like myself

Now Ive just made some samples to test a tricker case -that transcoding can work, no idea what they sound like yet - this could be a sobering one
Title: How to convince people to stop MP3 myths
Post by: Jojo on 2005-03-05 20:33:19
Quote
Quote
Quote
Its very easy to observe that low bitrates *do* use less processing power (winamp+taskmaster) and it follows that VBR *could* use more than CBR than VBR with the understanding that cheap firmware which messes up playtime of VBR files is also liable to stay in the powermode required to decode runs of higher bitrates, which it would not be presented with cbrs limited size and limited bit reservoir.
Not so far from the subject ive read of other players which last longer playing mp3s than aac.
In the end, its personal preference. I suggest folks should use the settings they prefer and am interested to hear individual outlooks.
[a href="index.php?act=findpost&pid=279080"][{POST_SNAPBACK}][/a]

wrong again...bitrates have nothing to with CPU power needed to decode the file
[a href="index.php?act=findpost&pid=279202"][{POST_SNAPBACK}][/a]


Try with foobar2000 and its diskwriter (with buffering): 320 kbps MP3 encoding needs more CPU power to be decoded than 128 kbps MP3 encoding.
[a href="index.php?act=findpost&pid=279245"][{POST_SNAPBACK}][/a]


interesting...what's the difference between 128kbps and 320kbps?
Title: How to convince people to stop MP3 myths
Post by: Jojo on 2005-03-05 20:47:24
Quote
Quote
Quote

wrong again...bitrates have nothing to with CPU power needed to decode the file
[a href="index.php?act=findpost&pid=279202"][{POST_SNAPBACK}][/a]


Try with foobar2000 and its diskwriter (with buffering): 320 kbps MP3 encoding needs more CPU power to be decoded than 128 kbps MP3 encoding.
[a href="index.php?act=findpost&pid=279245"][{POST_SNAPBACK}][/a]


Jojo is too fast judging stuff he obviously doesn't know anything about.
[a href="index.php?act=findpost&pid=279248"][{POST_SNAPBACK}][/a]

well, since we were talking about portable players in the first place and the author was just assuming stuff without testing it I provided an example. So even if there's a difference in terms of CPU power needed in that matter, the original claim is still the same since we were talking about portable players. Therefore, the question is if there are mp3 players out there that use optimal CPU power for every bitrate and if they do how big is the difference. If the difference is just 0.00001 Mhz or something it can be neglected...
Title: How to convince people to stop MP3 myths
Post by: guruboolez on 2005-03-05 21:45:24
CBR 128 needs less CPU cycles to be decoded (e.g. it's faster than CBR 320).
Title: How to convince people to stop MP3 myths
Post by: SirGrey on 2005-03-05 21:47:18
Jojo, to make it understandable:
[joke on]
Why when you need to go for buisness trip from Europe to USA you use plain ?
The boat have similiar speed - comparing to the speed of ray of light 
[joke off]

The more is bitrate, the more CPU power is needed to decode it.(if algorhytm is the same) It is obvious, I think.
There are 2 main questions:
1. Can the processing unit handle it.
2. Can the power source handle it for X time units.
Portables often has this two pushed to limits - bright example is iRiver player, which can decode ogg up to the fixed bitrate, and battery life is less than for mp3 (if I remeber correctly).
EDIT:
>>CBR 128 needs less CPU cycles to be decoded (e.g. it's faster than CBR 320).
It should, because algorhytm is the same, but huffman vocabulary is bigger, the more data needs to be processed, the more data needs to be taken from storage and get to the output and etc.
Title: How to convince people to stop MP3 myths
Post by: ChiGung on 2005-03-05 23:29:58
Thanks SirGrey, I couldnt put it so well.

Jojo i know its pure speculation whether these issues effect a significant number of portables without testing a range of them. In the balance of things I dont think it should effect a large amount of users, but there is plenty of flickering playback time happening with vbr -on the cheaper devices, for some users that will be a reason to use cbr, especially if the difference isnt obvious through cheap headphones.

[edit: removed possible flamebait, and said goodnight]
Goodnight
Title: How to convince people to stop MP3 myths
Post by: smz on 2005-03-06 00:06:39
hmmmm..  this thread started about mp3 myths, passed through a "lamedropXPd2 phase", was split to a thread about stereo (don't-know-how-to-call-it-don't-want-to-be flamed-calling-it-true) and joint stereo, is now all about CPU usage vs CBR/VBR and bitrate... well... "Where do you want to go today?"

Sergio
Title: How to convince people to stop MP3 myths
Post by: ChiGung on 2005-03-06 00:13:32
hehe, try again. 1 myth at a time.
Title: How to convince people to stop MP3 myths
Post by: disgustipated on 2005-03-06 07:58:05
Quote
Has anyone ever had a hard job trying to dispell certain mp3 myths? Like:

1) Joint Stereo messes up the sound.
2) Joint Stereo ruins high frequencies.
3) You can always tell the difference.

etc etc.

I feel like I'm bashing my head against a brick wall trying to talk to these people... 
[a href="index.php?act=findpost&pid=277194"][{POST_SNAPBACK}][/a]


Dude - just let it go. Don't worry about what other people think - its really not worth wasting your energy on. It's not your job to dispell myths about mp3 any more than it is your job to re-educate people who think Einstein was wrong or that the moon landing was a hoax. Just let them keep truckin' along and hopefully they will grow and uncover their own truths as they go.

As pointed out elsewhere in this thread, people believe what they want and it doesn't matter how rational and clear your explanation is, they will stubbornly keep believing what they want to and not listen until they are ready.

Just focus on enjoying and appreciating the things you have learned and don't worry about preaching to others - its a waste of time. I have been guilty of it myself, and yes it really IS like banging your head against a brick wall - so don't do it !!

Peace,
Title: How to convince people to stop MP3 myths
Post by: Jojo on 2005-03-06 19:29:28
Quote
CBR 128 needs less CPU cycles to be decoded (e.g. it's faster than CBR 320).
[a href="index.php?act=findpost&pid=279580"][{POST_SNAPBACK}][/a]

ok, how much of a difference does that make?
Title: How to convince people to stop MP3 myths
Post by: ChiGung on 2005-03-06 23:43:41
It simply illustrates that higher bitrates can take more processing power to decode.
There are other issues about power management and vbr decoding described, its not suggested that any of them are absolute. I never said as has been suggested elsewhere that 'vbr sucks' only that cbr has some merits still. Id recommend vbr for most requirements.
Title: How to convince people to stop MP3 myths
Post by: schonenberg on 2005-03-09 15:27:57
Quote
Quote
When using the 'quality' (VBR) mode, the 'encoding engine quality' is set within the quality presets.


No, that's not what I mean. Argh, I don't know how to describe it better, I'll post a screenshot when I get home.


Quote
These equate to the 'q' settings: fast = 9, high = 2 and standard = 5. (Although I just discovered an error in the code so that this is not currently set correctly!! I'll change that.)


I'm not really sure it's a good idea to give the user those options. And at least put them in the right order (fast, standard, high). And I'm also not sure if calling q5 "standard" is a good idea, might be a bit confusing, because I get the feeling most people consider q3 the LAME quality standard. But that's more a question of the subjective perception I guess...
Oh, and you mean q3, right? Since the old q2 is now q3...


Quote
It's the invocation of the 'nogap' switch that's a standard option.


You know that that option is pretty much useless and providing it to the user will most probably cause more harm than good.
[a href="index.php?act=findpost&pid=278508"][{POST_SNAPBACK}][/a]


the -q option just affects encoding speed only, right? Has nothing to do with bitrate?
Title: How to convince people to stop MP3 myths
Post by: vitos on 2005-03-09 23:40:28
Quote
the -q option just affects encoding speed only, right? Has nothing to do with bitrate?
[a href="index.php?act=findpost&pid=280629"][{POST_SNAPBACK}][/a]

Right. It switches different algoritms - faster encoding but not so effective (worse quality with the same chosen bitrate) or slower but better. You could have read this in Lame docs...
-q2 or -h is recommended.
Title: How to convince people to stop MP3 myths
Post by: schonenberg on 2005-03-10 15:00:12
Quote
Quote
So you come up with "hunches" but are too lazy to test them yourself?
[a href="index.php?act=findpost&pid=279070"][{POST_SNAPBACK}][/a]

Too lazy is a rather perjorative term, but considering id have to refill my player and run it for days to be sure, its not worth it for me. In this area my hunch is modestly informed. Its very easy to observe that low bitrates *do* use less processing power (winamp+taskmaster) and it follows that VBR *could* use more than CBR than VBR with the understanding that cheap firmware which messes up playtime of VBR files is also liable to stay in the powermode required to decode runs of higher bitrates, which it would not be presented with cbrs limited size and [a href="index.php?act=findpost&pid=279080"][{POST_SNAPBACK}][/a]

It sounds like that is what you _want_ to believe, as opposed to backing it up.

So more data goes in, but the same amount of data comes out with higher bitrate mp3, in my humble opinion.

Whether the displayed length and bitrate is jumping around or not, i.e. whether it's bitrate is vbr or cbr, the chips decoding time should always be the same in my opinion. In my GUESS decoding an mp3 would be less cpu time than refreshing the display.

Isn't the display recalculated in realtime anyway, whether the length or bitrate is changing or not? Each second the 'time left' changes, and is displayed every second, correct?
Title: How to convince people to stop MP3 myths
Post by: schonenberg on 2005-03-10 15:12:02
Quote
Quote
and maybe there should be another box on the very top, that gives some details about the selected mode (ABR, VBR, CBR). 1 or 2 sentences would be enough to explain the advantages / disadvantages of the selected method...

I think that the current interface is clear enough.
The average user doesn't care what is the difference between CBR and ABR, and does not even care about the meaning of ABR and the likes.
[a href="index.php?act=findpost&pid=279212"][{POST_SNAPBACK}][/a]


However many would care given the information right in the program as a tooltip, as otherwise the masses who would never read any documentation or this website never would and are fooled by musicmatch/real/microsoft etc.... Maybe someone could put a poll somewhere else or ask people in real life that if they know about Lame and use it, is it for its vbr quality?
Or Lame could sold to musicmatch and Real?

Personally I never would have heard about ABR or VBR if those options didn't exist in CDex. And if that program have some popup about what that actually is I would have used VBR much earlier.
Title: How to convince people to stop MP3 myths
Post by: schonenberg on 2005-03-10 16:33:05
Quote
Quote
the -q option just affects encoding speed only, right? Has nothing to do with bitrate?
[a href="index.php?act=findpost&pid=280629"][{POST_SNAPBACK}][/a]

Right. It switches different algoritms - faster encoding but not so effective (worse quality with the same chosen bitrate) or slower but better. You could have read this in Lame docs...
-q2 or -h is recommended.
[a href="index.php?act=findpost&pid=280805"][{POST_SNAPBACK}][/a]


heh, I install lame by putting lame.exe in my \dos path, I don't pollute my \dos dir with docs. That's a shortcoming of Windows 9x pathetic shell.

Why not call it speed and not quality? Or speed/quality? It affects speed more than anything else, imho.
I tried encoding at -q 0 trying to get the maximum quality possible, but I didn't want to take five hours to encode one file!  Yikes!

I associate speed with quality, it's common sense that slower speed is better quality. Common sense can be incorrect, though. Like the case of Nero Fast/High.

Definitions of quality on the Web:

    * an essential and distinguishing attribute of something or someone; "the quality of mercy is not strained"--Shakespeare
    * a degree or grade of excellence or worth; "the quality of students has risen"; "an executive of low caliber"
    * a characteristic property that defines the apparent individual nature of something; "each town has a quality all its own"; "the radical character of our demands"
    * timbre: (music) the distinctive property of a complex sound (a voice or noise or musical sound); "the timbre of her soprano was rich and lovely"; "the muffled tones of the broken bell summoned them to meet"
    * high social status; "a man of quality"
    * choice: of superior grade; "choice wines"; "prime beef"; "prize carnations"; "quality paper"; "select peaches"
    * of high social status; "people of quality"; "a quality family"

Definitions of speed on the Web:

    * distance travelled per unit time
    * a rate (usually rapid) at which something happens; "the project advanced with gratifying speed"
    * changing location rapidly
    * focal ratio: the ratio of the focal length to the diameter of a (camera) lens system
    * amphetamine: a central nervous system stimulant that increases energy and decreases appetite; used to treat narcolepsy and some forms of depression
    * rush: step on it; "He rushed down the hall to receive his guests"; "The cars raced down the street"
    * accelerate: move faster; "The car accelerated"
    * travel at an excessive or illegal velocity; "I got a ticket for speeding"
    * travel rapidly: move very fast; "The runner zipped past us at breakneck speed"
    * accelerate: cause to move faster; "He accelerated the car"
Title: How to convince people to stop MP3 myths
Post by: ChiGung on 2005-03-10 16:51:41
Quote
Why not call it speed and not quality? Or speed/quality? It affects speed more than anything else, imho.

Where the q settings are configured in the source file 'lame.c' they are introduced with this comment:

/* set internal feature flags.  USER should not access these since
* some combinations will produce strange results */

iirc , they can effect quality, speed, and/or bitrate or neither significantly or not. 
Title: How to convince people to stop MP3 myths
Post by: SirGrey on 2005-03-10 18:53:54
>>In my GUESS decoding an mp3 would be less cpu time than refreshing the
>>display.
Try to write IFFT by yourself and you will see how you are wrong
To be serious, i386 25Mhz was able to decode it flawesly, but under dos with optimized decoder... (If I remeber correctly those days  ).
As you also remeber, it easily handle Win3.11...
Title: How to convince people to stop MP3 myths
Post by: rjamorim on 2005-03-10 18:58:51
Quote
To be serious, i386 25Mhz was able to decode it flawesly, but under dos with optimized decoder... (If I remeber correctly those days  ).
[a href="index.php?act=findpost&pid=281067"][{POST_SNAPBACK}][/a]


 

To the best of my knowledge, compliant MP3 @44.1kHz decoding requires at least a 486 DX4-100 or a Pentium 90.
Title: How to convince people to stop MP3 myths
Post by: Gabriel on 2005-03-10 19:42:09
Quote
To the best of my knowledge, compliant MP3 @44.1kHz decoding requires at least a 486 DX4-100 or a Pentium 90.

I was able to play 128kbps on a 486 dx2-80 with winplay3 under win3.1 but this was the limit.
Title: How to convince people to stop MP3 myths
Post by: SirGrey on 2005-03-10 20:33:27
 
My fault. I ment i486 25Mhz.
Sorry.

Hmmm...
Guys, do you remeber that dos player: it had blue screen in text mode and at the bottom there was a white text bar displaying cpu load ?
I think P90 showed 15-25% load there on 160Kbit mp3...
I could be wrong, though. That time I have a plenty of machines to work with (from i386 33 to P 120), so results are mixed in my head now
BTW, under windows perfomance was very poor comparing to that dos player.

>>winplay3
He-he. Fhg player...
EDIT:
Well-organized archive is a good thing.
That player was Xtc-play.exe 
Title: How to convince people to stop MP3 myths
Post by: music_man_mpc on 2005-03-10 20:58:19
Quote
Quote
To the best of my knowledge, compliant MP3 @44.1kHz decoding requires at least a 486 DX4-100 or a Pentium 90.

I was able to play 128kbps on a 486 dx2-80 with winplay3 under win3.1 but this was the limit.
[a href="index.php?act=findpost&pid=281080"][{POST_SNAPBACK}][/a]

I had no problems decoding mp3 on my Pentium 75mHz either.

@ SirGrey  What decoder were you using?  I know for certain that my old Intel 486SX 25mHz wouldn't play mp3s.  Plus I remember that using taskmanager my Pentium 200 MMX showed ~20% CPU so I find it hard to believe that your P 90mHz was about the same, unless you were using some ultra-fast decoder.
Title: How to convince people to stop MP3 myths
Post by: ChiGung on 2005-03-10 23:18:42
Quote
I had no problems decoding mp3 on my Pentium 75mHz either.

The problems come when you try to do anything else at the same time 
Title: How to convince people to stop MP3 myths
Post by: Gabriel on 2005-03-11 09:21:08
486@25MHz was working, but only using a special "fast mode" reducing quality (probably by not decoding the whole spectrum)
Title: How to convince people to stop MP3 myths
Post by: SirGrey on 2005-03-11 12:40:06
>>music_man_mpc: @ SirGrey What decoder were you using?

Xtc-play.exe
dos player, used vesa graph. modes thought. It was initially designed as mod, s3m etc. player.
I do not know, what decoder was incorporated into it.
That time it was positioned as "fastest", but I do not know, if it is true, either...

>>486@25MHz was working, but only using a special "fast mode" reducing quality (probably by not decoding the whole spectrum)

It is quite possible...
Title: How to convince people to stop MP3 myths
Post by: schonenberg on 2005-03-11 14:58:43
Quote
Quote
Quote
Quote
I once read that exposure to low quality lossy encodings can cause cumulative hearing loss and tinnitus. While I can not say if that is true, I prefer to err or the side of caution.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=277443")


Don't exist scientific evidence that supports this.. (or at least credible).

This is another new myth that we can add to the list. 
[a href="index.php?act=findpost&pid=277463"][{POST_SNAPBACK}][/a]


That was on slashdot, the great purveyor of myths.
[a href="http://science.slashdot.org/article.pl?sid=02/12/20/2029212]http://science.slashdot.org/article.pl?sid=02/12/20/2029212[/url]

All I could be on google was this:
http://users.informatik.haw-hamburg.de/~wi...r/MP3-risk.html (http://users.informatik.haw-hamburg.de/~windle_c/Logologie/MP3-Gefahr/MP3-risk.html)
Ear damage by MP3, DVD and digital television?
    -risks of neuroacoustically datareduced music
[a href="index.php?act=findpost&pid=278345"][{POST_SNAPBACK}][/a]


I hope everyone reads the replies on Slashdot too...a lot of them point out the , um 'peculiarities' of that
'neuronomy' page and its author.  Looks like the hard experimental data supporting his particular speculations and inferences re mp3, hasn't yet materialized.

Over in audiophile-cuckoo-land, there's a rather famous amp designer (Mark Levinson -- trivia buffs might be interested to know he is the recent ex-husband of  Kim 'Samantha' Cattrall) who has been claiming for some years now that listening to 16/44 digital audio will damage your neuromuscular system.  Hasn't done much for his credibility.

Anyway, hardly a week goes by that I'm not trying to correct some nonsense spouted about mp3s on Usenet or the Web, usually along the lines of 'mp3s can NEVER sound as good as CDs'.  I always end up pointing them to HA.org. 
[a href="index.php?act=findpost&pid=278373"][{POST_SNAPBACK}][/a]


It is an interesting hypothesis, and that's all it is.
On a similar tangent, all recorded music is ultimately an
inferior reproduction of the live sound, imho.
Title: How to convince people to stop MP3 myths
Post by: JimmyJamm on 2005-03-11 16:10:38
Quote
Has anyone ever had a hard job trying to dispell certain mp3 myths? Like:

1) Joint Stereo messes up the sound.
2) Joint Stereo ruins high frequencies.
3) You can always tell the difference.

etc etc.

I feel like I'm bashing my head against a brick wall trying to talk to these people... 
[a href="index.php?act=findpost&pid=277194"][{POST_SNAPBACK}][/a]


I am a newbie to this sight and new to all of this.  I want to start ripping my collection with the best quality.

What is the technical difference between Stereo and JS and what is mid-side?

I want to be compatible with MP3 players ... so I am hesitant to Rip in a less common format, but will if I need to.

I am also trying to decide on a Software and have been looking at EAC.  I am not getting confused at to the Encoders and the Software and the Format that I will end up with ... does EAC Rip to MP3s ?

I hope this is the place to ask this question 
Title: How to convince people to stop MP3 myths
Post by: phong on 2005-03-11 16:32:38
You probably want to look at the various FAQs.

"Stereo" means it can only use L/R encoding.  "Joint-Stereo" means it can use L/R or M/S encoding depending on which produces the best quality for a given block.  M/S (mid-side) is just a different way of representing the same information that is sometimes easier to compress.  It's computed as follows:
M = L+R
S = L-R

When it decodes, yoy get L and R back:
L = (M+S)/2
R = (M-S)/2

Some basic algebra shows that you get back what you started with.  Technically, there are some other steps in there that make it more complicated, but you don't need to worry about it.  Joint-stereo is the default, and you should use it.  All players decode it correctly.
Title: How to convince people to stop MP3 myths
Post by: vitos on 2005-03-11 23:11:02
Quote
I am a newbie to this sight and new to all of this.  I want to start ripping my collection with the best quality.

What is the technical difference between Stereo and JS and what is mid-side?

Have you read earlier posts from this thread at all?? 

Quote
I want to be compatible with MP3 players ... so I am hesitant to Rip in a less common format, but will if I need to.

I am also trying to decide on a Software and have been looking at EAC.  I am not getting confused at to the Encoders and the Software and the Format that I will end up with ... does EAC Rip to MP3s ?

http://www.hydrogenaudio.org/forums/index....showtopic=28124 (http://www.hydrogenaudio.org/forums/index.php?showtopic=28124)

Quote
I hope this is the place to ask this question 
[a href="index.php?act=findpost&pid=281267"][{POST_SNAPBACK}][/a]

This is the place to look for answers.
Be honest - have you tried using search function or looking through this forum at all?
It looks like you just selected a random thread and post without reading...
Title: How to convince people to stop MP3 myths
Post by: Pa3PyX on 2005-07-16 09:27:16
Quote
You probably want to look at the various FAQs.

"Stereo" means it can only use L/R encoding.  "Joint-Stereo" means it can use L/R or M/S encoding depending on which produces the best quality for a given block.  M/S (mid-side) is just a different way of representing the same information that is sometimes easier to compress.  It's computed as follows:
M = L+R
S = L-R

When it decodes, yoy get L and R back:
L = (M+S)/2
R = (M-S)/2

Some basic algebra shows that you get back what you started with.  Technically, there are some other steps in there that make it more complicated, but you don't need to worry about it.  Joint-stereo is the default, and you should use it.  All players decode it correctly.
[a href="index.php?act=findpost&pid=281272"][{POST_SNAPBACK}][/a]


The myths mentioned here are myths, but as all myths, they do have roots in reality. I'll try to give detail here for those who would bother to read it, since even the advanced FAQ's don't appear to give enough detail to completely explain the issues that give rise to these myths. One at a time it is, then...

1) Joint Stereo destroys sound -- for LAME, this can be true depending on the settings and the decoder being used. A good concept as it is, implementations of both encoders and decoders often ruin it. (If any LAME developers are reading this -- Gabriel, Takehiro -- please correct me if I got any wrong ideas.)
Short: --nsmsfix 1.0 (equivalently, --nssafejoint, default value for -V0 in 3.97) makes MS as good as LR in VBR. Because of a slight bug, quality can be worse in rare instances. In CBR high bitrate, it is best avoided (depending on what quantcomp mode (-X?) you are using) because of how LAME bit allocation and masking works -- which has nothing in particular to do with the concept of Mid/Side stereo at all -- merely with its implementation.

Edit: This is probably well known (I can't imagine anybody not noticing it by now), but it's best not to use stock Windows (FhG) DirectShow decoder versions prior to 1.9 to decode MS stereo files, especially lower bitrate VBR/ABR. They do not handle small frames properly. Apparently they use low pass filter (or do not decode higher sfb's correctly?) for smaller frames, which results in horrible artifacts in VBR. This has to be one of the main reasons why MS stereo has acquired such a bad reputation, as FhG is probably the most popular decoder. One should either enforce minimum bitrate (-b128) -- most LAME presets do that, or use LR stereo (-ms) if targeting this particular decoder versions below 1.9. Both Winamp (current version) and mpglib (used in Media Player Classic, I believe) are devoid of this problem -- and so is version 1.9 of FhG, but I believe it only ships with some commercial products and various codec packs (correct me if I'm wrong here).

Like it was mentioned here, MS versus LR is just a different (and algebraically equivalent) way to encode sound -- if you are using lossless decoding. (LAME does not use IS, which we will not discuss here.) MP3 uses "masking" effects for each frequency band by setting appropriate thresholds for what the psymodel thinks would be inaudible at that frequency given all the other things going on. The sound is transformed into the frequency domain (originally a lossless transformation and so is 1-1 wrt size) and the quantizer then tries to truncate some frequency coefficients while analyzing noise that results from such activity, seeing that the noise injected as a result of this operation at each frequency band does not exceed the masking thresholds. This is mainly how compression is achieved. In CBR mode, you always have some set minimum of bits you can use, and the encoder would typically use them all (frame size can still vary because of the bit reservoir). So in CBR mode, the encoder will work with the pre-allocated number of bits per frame and for this number of bits to work with, will try to minimize the injected noise. In VBR mode, on the other hand, the encoder would (ideally) choose the barebone minimum amount of bits while maintaining all noise in a frame masked. So mathematically, VBR butchers the sound a lot more than CBR at a high bitrate, it's just that the distortions are (normally) not audible because you rarely have a problem of not having enough bits to encode a frame like you often do at medium/low bitrate CBR -- i.e. you can drop to as low as 32 (or whatever your user/preset specified minimum bitrate is) if that is enough, or you can go as high as 320 if need be.

Now depending on how different your Right channel is from your Left channel, MS or LR can provide a more efficient storage. For instance, if L == R, them L-R == 0 and so the Side channel in MS does not need any bits at all, and the M == L + R needs only as many bits as either L or R would had they been encoded separately. Similarly, if you have L == -R (signals in antiphase), you don't need any bits for M == 0 and all your bits can go to S. This is simultaneously the worst case scenario for LR and the best case scenario for MS.On the other hand, if L contains some signal and R is silent (or vice versa), LR encoding is more efficient because L + R == L - R. This is the best case scenario for LR and the worst case scenario for MS.

So whether to use MS or LR for a particular frame is to be decided at runtime based on which one is more efficient. Because best/worst case scenarios are rare in practice and everything else falls into "something in between", and because the masking thresholds can be calculated differently for MS and LR modes, it is not straightforward to determine which mode is more efficient based on just the overall stereo separation. What's then the most intuitive way to decide on MS/LR? Have the encoder try it both ways, and see which one introduces less noise (or, in VBR mode, uses less bits). LAME does not do that, and goes solely by stereo separation. Why? Not just because it's faster.

It's pretty easy to modify the encoder to implement the noise-based MS/LR switching as described above. But once you've done that, you'll find that you've just opened a whole new can of worms. Why? Because MS maskings are wrong. If you encode almost any full frequency range sample using just LR (-ms) and just MS (-mf), in an unrestrained (no min/max bitrate) VBR mode, you'll find that the perceived quality is not the same at the (allegedly) the same masking thresholds. MS sounds worse -- and yes, especially at high frequencies, because of some deficiecy in the formula that calculates the MS noise thresholds -- so when you try to compare noise between MS and LR, you are comparing apples to oranges.

Workaround? --nsmsfix. Naoki introduced a switch to enable a hack that ensures that MS masking thresholds are as good as LR thresholds. So --nsmsfix 1.0 (or --nssafejoint, which is the same thing) ensures that MS thresholds are at least as good as LR thresholds. [EDIT: Ignore the previous edit, I did the math wrong.] But of course this is a better than or equal mode, not an equal mode. Using this causes MS frames to often use more bits than they need to, unnecessarily increasing the bitrate -- but since stereo separation in most typical signals is fairly low, it can still prove more efficient than encoding as LR. Thus we set --nsmsfix to higher values at lower bitrates/qualities, to avoid bloating of MS frames.

Of course, the right way to fix this would be to fix the formula that produces the incorrect results. But I don't think that's easy anymore since most of the people who developed the psymodel(s) left the project some time ago.

In addition to this, in VBR, I understand there is a bug in on_pe() that prevents the routine from allocating more than 8190 bits per MS frame, even though up to 11520 (at 44 kHz) can be allocated per LR frame -- and this can be increased to 16380 if you ignore artifical format restrictions. This is not too significant as rarely would any sample require that many bits, but can potentially lead to lower quality. This has been entered in LAME bug tracking.


2) CBR 128 kbps is CD quality

It can be, depending on the type of music you are encoding. Silence is the trivial case, of course -- all you need for layer 3 is 32 kbps, and it's perfectly transparent  Of course more often than not it won't be (and especially with the way LAME allocates bits in CBR by default), but if you actually add a counter to LAME to see how many frames the quantizer could mask all noise in (see above), you will see that while for some songs it could barely mask 15% of the frames at 128 kbps, some other songs will have around 100%. Of course this also depends on the masking thresholds (higher quality VBR modes lower them), but contrary to some other beliefs, CBR mode does allow flexibility in frame size (you can add/subtract up to 2452 (== 0.6 * 4088) bits to a granule depending on how many you have available/used in the bit reservoir -- and up to 4088 if you ignore artificial format restrictions -- making it almost as flexible as ABR with proper bit allocation.

But just as with some samples you can mask all noise at 128 kbps, with some others you'll find that even 320 kbps is not enough, not to mention that they blow all known psymodels out of the water (I'm talking about fatboy.wav).
[EDIT: Well, I should probably take this one back; a psymodel that actually takes into account the temporal domain information, i.e. adjusts the masking ratios depending on the time variation in the signal (and thus does not suffer pre-echo problems by its design) should be able to handle this sample.]
Title: How to convince people to stop MP3 myths
Post by: Artemis3 on 2005-07-16 11:11:45
Quote
Quote
To the best of my knowledge, compliant MP3 @44.1kHz decoding requires at least a 486 DX4-100 or a Pentium 90.

I was able to play 128kbps on a 486 dx2-80 with winplay3 under win3.1 but this was the limit.
[a href="index.php?act=findpost&pid=281080"][{POST_SNAPBACK}][/a]

Did the same on a cyrix 486DX-100 with mpg123 compiled with i486 optimizations using some old linux distro. Same mpg123 compiled with just i386 was unable to handle them.

Maybe mad could do better?

I remember winplay3 doing it worse unless you had those lower quality options enabled. Other players were much worse (slower).
Title: How to convince people to stop MP3 myths
Post by: Pa3PyX on 2005-07-16 11:33:23
Quote
Quote
Quote
To the best of my knowledge, compliant MP3 @44.1kHz decoding requires at least a 486 DX4-100 or a Pentium 90.

I was able to play 128kbps on a 486 dx2-80 with winplay3 under win3.1 but this was the limit.
[a href="index.php?act=findpost&pid=281080"][{POST_SNAPBACK}][/a]

Did the same on a cyrix 486DX-100 with mpg123 compiled with i486 optimizations using some old linux distro. Same mpg123 compiled with just i386 was unable to handle them.

Maybe mad could do better?

I remember winplay3 doing it worse unless you had those lower quality options enabled. Other players were much worse (slower).
[a href="index.php?act=findpost&pid=313817"][{POST_SNAPBACK}][/a]

Hehe, I actually recall some problems decompressing audio on a K6-2/300 system when watching a XviD movie, using the stock FhG filter. If I watch a movie with no audio, it plays fine, but eats some 70-90% of the CPU (that given XviD has about the fastest decoder out there). If I add in audio, I hit 100% and the video starts to lag. FhG does have low quality options, and I recall that "downsample to mono" had the greatest effect and alleviated the problem, apparently because the audio track was joint stereo and when downmixing to mono from MS, the decoder can basically forget about the side channel altogether.