Skip to main content ↑ ↓
Topic: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ? (Read 10325 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Since i only collect mp3 encoders which sounds unique when comparing their MP1/MP2/MP3 encodings against each other.



If all versions of:
(FASTENC/FRAUNHOFER/LAME) ,  uses the same psychoacoustic models and settings.


Then only keeping one version for each encoder makes sense.

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #1 –
"Which mp3 software can replace every other mp3 frame with the frame before it? "
"I can't use libav like this:"
"Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?"

You don't have a track of reasonable questions...

First, let's clarify your sentences:
Of course, LAME does not use the same pyschoacoustic model than fraunhoffer, so I guess you mean to ask if each brand keeps the same psychoacoustic model in all its versions.

Next, obviously, you are asking if all the versions of the encoders have the same quality, which has an obvious answer: False.
In fact, LAME has had many important changes in its model ( GPSYCHO ) since it was first introduced in version 3.0 in 1999.

So whatever you want to achieve, you cannot reduce it to a single version of a single product.

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #2 –
Quote from: [JAZ] on
In fact, LAME has had many important changes in its model (GPSYCHO)
IIRC LAME also has second psychoacoustic model called nspsytune.

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #3 –
Quote from: [JAZ] on
"Which mp3 software can replace every other mp3 frame with the frame before it? "
"I can't use libav like this:"
"Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?"

You don't have a track of reasonable questions...
What ?

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #4 –
@lvqcl : yes, you are right. I don't remember if the whole nspsytune is currently present in the main branch, or only some of the improvements were backported from naoki's work.

@Jedinhopy : I just looked at your previous posts and all of your questions have been strange... Asking for something unusual for an MP3 file, asking about how to use a not recommended encoder and now about psychoacoustics and putting MP1/MP2/MP3 all in the same bag as if it was all the same.

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #5 –
Quote
and now about psychoacoustics and putting MP1/MP2/MP3 all in the same bag as if it was all the same.

And also "collect mp3 encoders which sounds unique" phrase... What does it mean?

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #6 –
Quote from: lvqcl on
collect mp3 encoders which sounds unique" phrase... What does it mean?
That does not produce 99% identical results.

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #7 –
I am not sure where you could find the information to get exactly what you want, but you sure can make a distinction in this way:

- dist10 based, minimally modified code  ( bladeenc, 8hz,  lame previous to 3.0 etc).
- lame gpsycho, previous to "alt-presets" (pre 3.92)
- lame previous to VBR "mtrh" tunnings (pre 3.96)
- current lame
- buggy fastenc (pirated code)
- official fastenc (i think this was used as a plugin in some wave editors, DAWs, aside of the command line application)
- official fraunhoffer.
- apple's itunes mp3 encoder (from older versions )
- apple's quicktime mp3 encoder

This is a rough list but can give you some idea on what to look for

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #8 –
Is:
(mpeg Encoder v0.07a - by SoloH)
, based on the (official fraunhoffer) ? and/or which version of the (official fraunhoffer) ?

Does the:
(Internal MP2 Encoder (version 1.12, engine 1.12) )
, . In (CDex Version 1.30 Beta 8), sound unique when compared to:
(twolame) and the MPEG encoders in: (SCMPX Version 1.5.1) and: (Electronic Cosmo's MPEG Suite 1.5) ?

And:
(8hz-mp3-011b)
, uses:
(Psychoacoustic Model: AT&T)
.

And does the:
(Xing MP3 Encoder (C) 1999 Xing Technology Inc.)
(x3enc.exe)
, sound unique apart from excluding stereo at low bitrates?

And does:
(Mthree Development WAV to MP3 Encoder Version 2.0.8)
, sound unique?

And does the:
(mp3-tvq-e22)
(TestEnc.exe)
, sound unique too?

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #9 –
You will have to find out on some of those questions, but a few things should be obvious.

Soloh MPEG encoder is obviously not fraunhoffer.
Quote
Soloh MPEG Encoder v0.07a (slow, uses ISO Code)
  ISO = dist10 = demonstration code (simple, to show people how to generate such files). Many initial third-party mp3 encoders started from this code. Fraunhoffer code did not.

8hz-mp3, as i said, is a dist10 based encoder. Not sure where you found that "AT&T", but if anything, it might imply dist10.


I completely forgot about Xing.  Xing would be one of the encoders that you want to keep, but there are two quite different versions:  1.x, and 2.x, most importantly, the one released with the product "AudioCatalyst". Most of the problems existed in 1.x, although 2.x is not a high quality encoder at moredate bitrates. In comparison, it was up to 8 times faster than the competition.

Xing almost dissapeared, but the Helix MP3 encoder is a descendant from it. (AFAIK)

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #10 –
Quote from: Jedinhopy on
I only collect mp3 encoders which sounds unique
...
That does not produce 99% identical results.

Do you collect broken encoders? This is a weird hobby.

Most good encoders strive to be *audibly indistinguishable* from the master and, therefore, indistinguishable from each other. What 99% you are talking about? The only true measure are blind tests.

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #11 –
Sorry if I'm a bit off-topic here, but for learning, can I please ask a question?

As a person who only use LAME for MP3 encoding (and can't even imagine using any other--except I played with L3ENC in the past out of curiosity) (I primarily use WavPack hybrid, BTW), if the OP is looking for an MP3 encoder that sounds unique, then isn't it a bit futile since all MP3 encoders strive to sound the same, I mean, sound transparent? Given enough bitrate, they should sound the same, don't they? They will only sound different at lower bitrates, but even then, they will still have the same artifacts, is that right? Compared to comparing MP3 with Vorbis with AAC with Opus, wouldn't that give 'unique' artifacts at lower bitrates (and even then, will sound the same at high bitrates)?

Pardon the question, just curious.

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #12 –
@OrthographicCube The MP3 format defines a specification for the format, and how is a decoder expected to decode that format.
It does not define a set of tools to use to generate the file, provided that a conformant decoder could decode the file.

Obviously, since the format is fixed as well as how a decoder works, the way an encoder works cannot deviate too much from others, but still give opportunity for doing things differently.

As you say, using a high bitrate should compensate for deficiencies or bad decisions taken on a specific encoder, but not using some tools, or using some other incorrectly can cause distinctive problems:
- First versions of Xing encoder: Preecho/Problems with transients due to the lack of short block usage
- Pirated fastenc Windows codec:  Stereo collapse Bug in joint stereo mode.
Some encoders might also show more pronounced warbling/killing of mid frequencies due to lack of proper filtering and ATH curve.

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #13 –
I can't prove it just yet but my feelings tell me that this is a troll thread. That is, a thread created solely for the purpose of making people angry.
some ANC'd headphones + AutoEq-based impulse + Meier Crossfeed (30%)

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #14 –
@magicgoose I think we managed it well enough back then, when the topic started. (Duh! I didn't even remember that I wrote most of the answers)

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #15 –
Quote from: [JAZ] on
Obviously, since the format is fixed as well as how a decoder works, the way an encoder works cannot deviate too much from others, but still give opportunity for doing things differently.
Ooh, that makes sense, as I was thinking that the way the codec works defines the artifacts that it produces (the most obvious ones I can think of is low-pass filter for MP3, metalic chirping for AAC, lack of "attack" (as in ADSR attack) in Vorbis, and static white noise in Opus). Now that you've said that, an encoder can indeed make unique artifacts based on how it works, be it by design or by bugs.

Thanks for clarifying that for me!
 

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #16 –
Quote from: apodtele on
Do you collect broken encoders?
Not only broken,  but even working ones that are different when compared against each other.
Not only different encoders sounds different, even different versions of the same encoder
can sound different due to the fact that features are both added and removed along
the way when updates are released for that encoder/encoders.
And it does not have to always include bug fixes in every increment update, in order for changing the encoding results.
Like if either (frequency masking) is removed or (intensity stereo) is removed, in favor of higher quality encoding.


And i even collect different sounding encoders for adding transformative flavour for different (audio clips) when
doing sound design within unusual music production projects.



Quote from: magicgoose on
I can't prove it just yet but my feelings tell me that this is a troll thread. That is, a thread created solely for the purpose of making people angry.
I try to ask, and/or write questions the best way i can.
Plus that english is not my first language.

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #17 –
Whatever floats your boat :)

Maybe extend to other ancient codecs, enough to find there, like the Yamaha VQF and TwinVQ codecs etcetera.

Re: Which MP1/MP2/MP3 encoders does not use FASTENC/FRAUNHOFER/LAME ?

Reply #18 –
Today LAME is de facto standard, so if you're looking for differently-sounding encoders, look for older ones.
Have you tried looking here? --> http://www.rarewares.org/rrw/programs.php
There are LOTS of obsolete MP3 encoders.
sox -e float -b 32 -V4 -D gain -3 rate -v 48000 norm -1
opusenc --bitrate 128
 
SimplePortal 1.0.0 RC1 © 2008-2020