IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
a classical music and mp3 issue?
twlkwind
post Dec 2 2002, 02:07
Post #1





Group: Members
Posts: 24
Joined: 19-May 02
Member No.: 2076



Is it true that classical music lose more details after encoded with lame or other encoder like mppenc?
Go to the top of the page
+Quote Post
fewtch
post Dec 2 2002, 03:24
Post #2





Group: Members
Posts: 1460
Joined: 5-February 02
From: Seattle WA. USA
Member No.: 1261



QUOTE (twlkwind @ Dec 1 2002 - 06:07 PM)
Is it true that classical music lose more details after encoded with lame or other encoder like mppenc?

Classical music should actually encode better than most other kinds of music, although the audiophile definition of "detail" doesn't appear to be something that can be ABX'ed... tongue.gif


--------------------
Bring back dynamic range... www.loudnessrace.net
Go to the top of the page
+Quote Post
twlkwind
post Dec 2 2002, 04:15
Post #3





Group: Members
Posts: 24
Joined: 19-May 02
Member No.: 2076



Can someone explain the relation between music style and compression format (e.g. mp3)? I mean which style can be easily encoded and why?
Go to the top of the page
+Quote Post
twlkwind
post Dec 2 2002, 04:35
Post #4





Group: Members
Posts: 24
Joined: 19-May 02
Member No.: 2076



The situation is i am managing a music related forum. Lately there are many discussion about which audio format we should use -- ape or mp3?
The ape supporter consider we should use ape when sharing classical music. I know well-tuned mp3 is enough but i don't know how to explain the reason. So pls help me.
Go to the top of the page
+Quote Post
Garf
post Dec 2 2002, 07:50
Post #5


Server Admin


Group: Admin
Posts: 4883
Joined: 24-September 01
Member No.: 13



MP3 is enough because it's impossible to hear a difference from the CD for the great majority of music. If they claim otherwhise, they should prove they can ABX alt-preset-standard on most songs.

You could also consider MPC or Vorbis which are technically superior to MP3.
Go to the top of the page
+Quote Post
fewtch
post Dec 2 2002, 08:05
Post #6





Group: Members
Posts: 1460
Joined: 5-February 02
From: Seattle WA. USA
Member No.: 1261



QUOTE (twlkwind @ Dec 1 2002 - 08:35 PM)
The situation is i am managing a music related forum. Lately there are many discussion about which audio format we should use -- ape or mp3?
The ape supporter consider we should use ape when sharing classical music. I know well-tuned mp3 is enough but i don't know how to explain the reason. So pls help me.

I believe it's because classical music tends to be quieter (with less going on all at once) than something like heavy metal, but don't quote me on that... I'm no expert on lossy encoding.


--------------------
Bring back dynamic range... www.loudnessrace.net
Go to the top of the page
+Quote Post
twlkwind
post Dec 2 2002, 08:05
Post #7





Group: Members
Posts: 24
Joined: 19-May 02
Member No.: 2076



I've been pushing the mpc format on my forum since i know it's better audio quality and faster encoding speed.
There is only one thing i want to know:
Is there any relation between music style and compression format? You know different music has different behavior. So is it your opinion that we can encode POP and Classical and Jazz and all the other music using Lame's standard switch(alt-presets) in order to get the transparent audio quality ? The well-known critical
songs shouldn't be included of cource.

This post has been edited by twlkwind: Dec 2 2002, 08:11
Go to the top of the page
+Quote Post
Garf
post Dec 2 2002, 08:08
Post #8


Server Admin


Group: Admin
Posts: 4883
Joined: 24-September 01
Member No.: 13



QUOTE (twlkwind @ Dec 2 2002 - 09:05 AM)
Is there any relation between music style and compression format? You know different music has different behavior. So is it your opinion that we can encode POP and Classical and Jazz and all the other music using Lame's standard switch(alt-presets) in order to get the transparent audio quality?

Heavy metal and the likes are known to produce higer bitrates, but the resulting files are still transparent.
Go to the top of the page
+Quote Post
Dibrom
post Dec 2 2002, 08:13
Post #9


Founder


Group: Admin
Posts: 2958
Joined: 26-August 02
From: Nottingham, UK
Member No.: 1



QUOTE (twlkwind @ Dec 2 2002 - 12:05 AM)
I've been pushing the mpc format on my forum since i know it's better audio quality and faster encoding speed.
There is only one thing i want to know:
Is there any relation between music style and compression format? You know different music has different behavior. So is it your opinion that we can encode POP and Classical and Jazz and all the other music using Lame's standard switch(alt-presets) in order to get the transparent audio quality?

There is a correlation to an extent, but not as much as one would think. The only type of music I've really encountered that really makes quite a big difference is extremely artificial and synthetic electronic music. Traditionally, psychoacoustic encoders seem to have the most problems with this type of music because many of the sounds that are used are completely unnatural, and in some cases impossible to recreate without direct manipulation of the actual sound signal via a computer of some sort. Most encoders are not really tuned to handle this type of behavior. For the rest though, including pop, jazz, metal/rock, and classical, there's usually not all that much difference. For most genres of music --alt-preset standard will work fine. For the type of electronic music I mentioned above, another format is recommended, such as MPC.
Go to the top of the page
+Quote Post
twlkwind
post Dec 2 2002, 08:18
Post #10





Group: Members
Posts: 24
Joined: 19-May 02
Member No.: 2076



Dibrom, thanks very much. You are always the authority on lossy encoding.
Go to the top of the page
+Quote Post
Dibrom
post Dec 2 2002, 08:20
Post #11


Founder


Group: Admin
Posts: 2958
Joined: 26-August 02
From: Nottingham, UK
Member No.: 1



QUOTE (twlkwind @ Dec 2 2002 - 12:18 AM)
Dibrom, thanks very much. You are always the authority on lossy encoding.

lol laugh.gif I don't know about that so much, but np anyway wink.gif
Go to the top of the page
+Quote Post
fewtch
post Dec 2 2002, 08:45
Post #12





Group: Members
Posts: 1460
Joined: 5-February 02
From: Seattle WA. USA
Member No.: 1261



A lot of the Moog synth music I "ripped" from vinyl ends up at around 200kbps (or a bit more) using --alt-preset standard. This would seem to support the statement about artificial waveforms, yet there's some alteration/distortion when played back from vinyl, which probably changes things somewhat.

Another note of interest... when using an older turntable that added more constant rumble/distortion, average bitrates were substantially higher. Now that I have a better turntable and cartridge, average bitrates dropped substantially (closer to that of comparable music on CD).

And finally... even with this Moog music from vinyl (adding all sorts of unknown/untested factors), I've encountered few songs that sound less than transparent with --alt-preset standard (I can only think of a few, and I believe they all involved white noise). Great work!!

This post has been edited by fewtch: Dec 2 2002, 08:49


--------------------
Bring back dynamic range... www.loudnessrace.net
Go to the top of the page
+Quote Post
hans-jürgen
post Dec 2 2002, 12:52
Post #13





Group: Members
Posts: 573
Joined: 2-August 02
From: Hamburg, Germany
Member No.: 2898



QUOTE (twlkwind @ Dec 2 2002 - 08:05 AM)
I've been pushing the mpc format on my forum since i know it's better audio quality and faster encoding speed.

OK, but have you heard it for yourself? wink.gif For example there is one user here who even can't find any adequat words in his own language to describe how bad MPC sounds to him with his favorite harpsichord music at low bitrates. ohmy.gif

QUOTE
There is only one thing i want to know:
Is there any relation between music style and compression format? You know different music has different behavior. So is it your opinion that we can encode POP and Classical and Jazz and all the other music using Lame's standard switch(alt-presets) in order to get the transparent audio quality? The well-known critical songs shouldn't be included of cource.


As you can see from the above example, you really can't generalize that much, although almost everyone here seems to like it that way. The sound of a harpsichord is indeed a part of "classical music", but due to its own characteristics is bound to be a "codec killer". So does this fall under your exception "well-known critical songs" or not? I don't know...

From my own recent tests at low bitrates I would rather say that MPC might be very good with complex sounds from classical music like huge orchestras in late Romantic compositions (Mahler, Bruckner and the like) or operas, because it doesn't alter the stereo image even at ~96 kbps. Also it doesn't introduce any white noise like Ogg Vorbis which is important for vocal music (female and male voices in opera or chamber or church music) and for the timbre of violins or wind instruments.

Furthermore you should not forget AAC as an alternative compression format for MP3, because it is more efficient and sounds quite similar to MPC (at least to me). But if file size is no problem at all (like it seems to be for the person who suggests the lossless format), you might also be happy with LAME's alt-preset standard or extreme. The most important thing is that you should not trust anything someone tells you (especially me), but rather test it for yourself before making generalizing comments.


--------------------
myspace.com/bluezzbastardzz
myspace.com/indigorocks
Go to the top of the page
+Quote Post
Dibrom
post Dec 2 2002, 13:38
Post #14


Founder


Group: Admin
Posts: 2958
Joined: 26-August 02
From: Nottingham, UK
Member No.: 1



QUOTE (hans-jürgen @ Dec 2 2002 - 04:52 AM)
QUOTE (twlkwind @ Dec 2 2002 - 08:05 AM)
I've been pushing the mpc format on my forum since i know it's better audio quality and faster encoding speed.

OK, but have you heard it for yourself? wink.gif For example there is one user here who even can't find any adequat words in his own language to describe how bad MPC sounds to him with his favorite harpsichord music at low bitrates. ohmy.gif

Hrmm.. but how relevant is this? As far as I can tell, the original poster isn't talking about low bitrates. Furthermore, who is using MPC at low bitrates? It isn't designed for low bitrates, it isn't tuned for low bitrates, and it's not recommended for low bitrates by the developers who have worked on it. Saying that it may not be good at low bitrates (on harpsichord music or not) doesn't really mean much in this context.

QUOTE
QUOTE
There is only one thing i want to know:
Is there any relation between music style and compression format? You know different music has different behavior. So is it your opinion that we can encode POP and Classical and Jazz and all the other music using Lame's standard switch(alt-presets) in order to get the transparent audio quality? The well-known critical songs shouldn't be included of cource.


As you can see from the above example, you really can't generalize that much, although almost everyone here seems to like it that way.

[...]

The most important thing is that you should not trust anything someone tells you (especially me), but rather test it for yourself before making generalizing comments.


I do certainly agree with the idea of listening on your own before drawing conclusions, but at the same time I think that it is safe to make certain assumptions about overall performance of an encoder in various situations based on past listening tests and other data available on the forums here. Of course, anything goes on a case by case situation and one should always verify results if they need a precise judgement of quality, but it is also well known (through continuous testing and verification) that certain encoders perform well in certain areas. I don't think there is necessarily a need to disregard all public knowledge, but only to be wise in your application of such.
Go to the top of the page
+Quote Post
twlkwind
post Dec 2 2002, 13:46
Post #15





Group: Members
Posts: 24
Joined: 19-May 02
Member No.: 2076



to hans-jürgen:

1. I have been here for quite a long time and maybe i know which person you're talking about. IMO, mpc is the best encoder untill now. I heard FhG had a aac-based demo encoder which seemed to be the best one, but i can't get it.

2. On this forum there must be someone i would trust such as Dibrom, johnV...etc. They have more experience than i.

3. I know AAC has good audio quality. But it's so slow when encoding (Only 1x avg). Furthermore I like open source.
Go to the top of the page
+Quote Post
Annuka
post Dec 2 2002, 15:22
Post #16





Group: Members
Posts: 333
Joined: 2-February 02
Member No.: 1233



Lossless is a realistic alternative for classical music. FLAC has reduces most of my classical music to ~30% of original size. Still 430 kbps, but how much classical music do you have.
Go to the top of the page
+Quote Post
twlkwind
post Dec 2 2002, 16:05
Post #17





Group: Members
Posts: 24
Joined: 19-May 02
Member No.: 2076



QUOTE (Annuka @ Dec 2 2002 - 02:22 PM)
Lossless is a realistic alternative for classical music. FLAC has reduces most of my classical music to ~30% of original size.  Still 430 kbps, but how much classical music do you have.

i'm leading a music related forum which contains many FTPs, and many people share their classical music using lossless format. So there will be at least 1G ape files everyday, not including the other style music.
Now u can understand what i'm worrying about.
Go to the top of the page
+Quote Post
hans-jürgen
post Dec 2 2002, 18:28
Post #18





Group: Members
Posts: 573
Joined: 2-August 02
From: Hamburg, Germany
Member No.: 2898



QUOTE (twlkwind @ Dec 2 2002 - 01:46 PM)
1. I have been here for quite a long time and maybe i know which person you're talking about.  IMO, mpc is the best encoder untill now. I heard FhG had a aac-based demo encoder which seemed to be the best one, but i can't get it.

Don't get me wrong, I think guruboolez is probably right with his opinion on MPC and harpsichord music at low bitrates. I only used this as an example how fast you can go wrong if claiming one format or another to be always good or bad on a certain kind of music.

And the FhG demo isn't available for the public, that's right. I only have a short test file from the recent c't comparison, and I noticed that Ivan had encoded at least one test file with version 2.2 and uploaded it on his server.

QUOTE
2. On this forum there must be someone i would trust such as Dibrom, johnV...etc. They have more experience than i.


Maybe, but as far as I know they also favour own comparisons instead of plain believing everything they say. I think you should always take an expert's opinion as an invitation or suggestion to check if it might be true for yourself and your own preferences, especially when talking about something as subjective as music or even psychoacoustics.

QUOTE
3. I know AAC has good audio quality. But it's so slow when encoding (Only 1x avg). Furthermore I like open source.


There is a fast version of AACEnc called FastEnc, you can get it at Roberto's website. It is much faster than AACEnc, almost as fast as FhG's FastEnc for MP3. And Ivan has already stated here that the new Nero plugin will also be faster than PsyTEL AACEnc v2.15.

And the open source thing: at least for me it is not important if the people who developed a codec did it in their spare time or if they were paid for it, I'm only interested in the resulting sound quality. My former experiences in another open source project have further convinced me that a chaotic environment isn't always the best condition for a good result.


--------------------
myspace.com/bluezzbastardzz
myspace.com/indigorocks
Go to the top of the page
+Quote Post
guruboolez
post Dec 2 2002, 19:53
Post #19





Group: Members (Donating)
Posts: 3474
Joined: 7-November 01
From: Strasbourg (France)
Member No.: 420



QUOTE (Garf @ Dec 2 2002 - 08:08 AM)
Heavy metal and the likes are known to produce higer bitrates, but the resulting files are still transparent.

This is true for mp3. MPC has a different behavior. I noticed it immediatly when I switched to mpc, and not only on harpsichord : violin (esp. ancient instruments), organ and exotic instruments like glass harmonica. Of course, a piano encoding is going under the usual bitrate value. Metal encoding are more stable (170-180 kb/s for mpc --standard, 200-210 for --extreme, etc...)
Go to the top of the page
+Quote Post
guruboolez
post Dec 2 2002, 20:11
Post #20





Group: Members (Donating)
Posts: 3474
Joined: 7-November 01
From: Strasbourg (France)
Member No.: 420



For twlkwind > The only flaw I found to --alt-preset were :
· sharp and plain harpsichord recordings
· strong attacks (the 'classic' pre-echo problem)

Musepack is more flexible on natural instruments : between piano and harpsichord, there is a big difference in bitrate. The same Goldberg variations are going with --extreme from 185 kb/s (piano) to 250 kb/s (hapsichord) : +30/35 % ! VBR behavior in Mp3 is totally different, and bitrate never explode (I rarely reached 200 kb/s with --aps).

And don't forget anything else : gap between tracks. Mp3 is not gapless. Mp3cue doesn't work very well with lame VBR ; crossfading are not perfect. It's a real pain when listening to an opera, an orchestral suite, a piano sonata, an oratorio, etc... Classical needs gapless playback.

This post has been edited by guruboolez: Dec 2 2002, 20:13
Go to the top of the page
+Quote Post
hans-jürgen
post Dec 3 2002, 12:43
Post #21





Group: Members
Posts: 573
Joined: 2-August 02
From: Hamburg, Germany
Member No.: 2898



QUOTE (Dibrom @ Dec 2 2002 - 01:38 PM)
QUOTE (hans-jürgen @ Dec 2 2002 - 04:52 AM)
QUOTE (twlkwind @ Dec 2 2002 - 08:05 AM)
I've been pushing the mpc format on my forum since i know it's better audio quality and faster encoding speed.

OK, but have you heard it for yourself? wink.gif For example there is one user here who even can't find any adequat words in his own language to describe how bad MPC sounds to him with his favorite harpsichord music at low bitrates. ohmy.gif

Hrmm.. but how relevant is this? As far as I can tell, the original poster isn't talking about low bitrates.

But twlkwind talked about "pushing MPC" in general on his forum, connotating this to his original question about better suited codecs for classical music where he did not speak about bitrates first. OK, but I don't want to start nitpicking, because I noticed in the new "Monty" thread that you've already read the postings of Case and myself about MPC at low bitrates in the MP3 board, and I used guruboolez's opinion only as an example, like I already mentioned in my answer to twlkwind.

QUOTE
Furthermore, who is using MPC at low bitrates?  It isn't designed for low bitrates, it isn't tuned for low bitrates, and it's not recommended for low bitrates by the developers who have worked on it.


Do you have a link to a thread where Frank Klemm wrote this recently? I got an answer from him yesterday in de.comp.audio (a german newsgroup) where he states that he also plans to improve Musepack at low bitrates having some new ideas that could help in this issue. Furthermore he also repeated the same impression that Case already gave in relation to MP3 at low bitrates.

QUOTE
Saying that it may not be good at low bitrates (on harpsichord music or not) doesn't really mean much in this context.


That's what guruboolez said, Case and me said the opposite, and that's what I wanted to point out and what does matter in this context: there are no quick all-time truths, even (or maybe especially) if they are repeated over and over again. But I think you would probably agree on that, because you already said that there is no general guideline which codec sounds best on classical music.

By the way, guruboolez will send me some samples with classical music encoded at low bitrates where he finds MPC to be inferior to other formats, so I can compare them for myself. But I probably can't comment on them before Thursday, because I have to go on working now... wink.gif

QUOTE
QUOTE
The most important thing is that you should not trust anything someone tells you (especially me), but rather test it for yourself before making generalizing comments.

[...]
Of course, anything goes on a case by case situation and one should always verify results if they need a precise judgement of quality, but it is also well known (through continuous testing and verification) that certain encoders perform well in certain areas. I don't think there is necessarily a need to disregard all public knowledge, but only to be wise in your application of such.


Right, and I did not suggest to "disregard all public knowledge", but rather to trust your own ears after doing the best you can to find an answer to your specific question, i.e. reading comments and making own thorough comparisons with different codecs. And in the end you may even come up with a decision that differs from the one that the favorite guru and/or his zealots praise, because the sensitivity to certain kinds of artifacts is probably even more subjective than to music styles in general.


--------------------
myspace.com/bluezzbastardzz
myspace.com/indigorocks
Go to the top of the page
+Quote Post
Dibrom
post Dec 3 2002, 13:01
Post #22


Founder


Group: Admin
Posts: 2958
Joined: 26-August 02
From: Nottingham, UK
Member No.: 1



QUOTE (hans-jürgen @ Dec 3 2002 - 04:43 AM)
QUOTE
Furthermore, who is using MPC at low bitrates?  It isn't designed for low bitrates, it isn't tuned for low bitrates, and it's not recommended for low bitrates by the developers who have worked on it.


Do you have a link to a thread where Frank Klemm wrote this recently? I got an answer from him yesterday in de.comp.audio (a german newsgroup) where he states that he also plans to improve Musepack at low bitrates having some new ideas that could help in this issue. Furthermore he also repeated the same impression that Case already gave in relation to MP3 at low bitrates.

I don't know if Frank has ever said this explicitly, but I know that Andree (the original creator of MPC) has certainly implied it if not said it outright. The codec was not designed from the start for low bitrates either, so that suggests this as well.

QUOTE
QUOTE
Saying that it may not be good at low bitrates (on harpsichord music or not) doesn't really mean much in this context.


That's what guruboolez said, Case and me said the opposite, and that's what I wanted to point out and what does matter in this context: there are no quick all-time truths, even (or maybe especially) if they are repeated over and over again. But I think you would probably agree on that, because you already said that there is no general guideline which codec sounds best on classical music.

By the way, guruboolez will send me some samples with classical music encoded at low bitrates where he finds MPC to be inferior to other formats, so I can compare them for myself. But I probably can't comment on them before Thursday, because I have to go on working now... wink.gif


The bit about pointing out that "there are no quick all-time truths" would be fine, but nobody is saying that MPC is really supposed to be good at low bitrates. AFAIK, it's not considered a "truth" by any stretch. How then is pointing out that it may be problematic at low bitrates (assuming for a second we're ignoring the fact that it is actually showing a lot of potential apparently) exactly relevant? This is why I mentioned context btw. In the other case, you're right in that truths shouldn't be taken for granted (that MPC is necessarily bad at low bitrates), but in this case, I don't think there is a commonly held conception that it is good at low bitrates.. so if it doesn't perform well on some harpsichord music at lower bitrates it doesn't change a lot.
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 1st August 2014 - 05:56