What is better, LAME or Fraunhofer MP3 Encoder?

Welcome.

Open your eyes before posting, please. There are two sticky topics "List of recommended LAME settings/compiles"... go figure.

?

LAME has a much better proven track record, since a LOT of public testing has been done on it. It is possible that other encoders could be better, but they don't have the testing behind them than LAME does. In other words, if you wanted a gun, I could offer you an M-16 or something in a black box. LAME is the M-16 - at least you know what you're getting.

You didn't read a single post, did you? In the MP3 section, you can read LAME everywhere. As Silversight said go figure...

I mean Fraunhofer MP3 Encoder is very professional.

and? i think you are trolling.

Ah no, i mean what is better for archiving, LAME or Fraunhofer MP3 Encoder?

I agree with beto. They have answered your question in previous posts. FhG may be used by programs like MusicMatch and Windows Media Player but Lame has the proven quality. Besides, I don't see any "professionals" using the FhG mp3 encoder.

Because LAME is free.

LAME Options:


Fraunhofer MP3 Encoder Options:

Would you pay $10'000 for a car that keeps coughing up smoke, dies extremely frequently in the highway,

Or pay $5'000 for a car that, well, runs well?

Paid software does not guarantee a quality product.

IMHO, the actual LAME encoder options are quite professional:

http://lame.cvs.sourceforge.net/*checkout*/lame/lame/USAGE

Fraunhofer is developer of MP3 format.
Why LAME is better?

>There are two sticky topics "List of recommended LAME settings/compiles"... go figure.

Because Lame has more coverage on HA does not mean it is any better by default. Again better for what? low / mid or high bitrates? an encoder which is good at low bitrates might not shine at high bitrates. Only listening tests completed with latest versions of each could determine (btw I am guessing that most people could not tell most mp3 encoders apart 160 kbps and above)...

Ok, Franhofer doesn't update MP3 codec, but LAME does.

Jimpin, you're killing me man
The best thing you can do is read more

Me too. Looks like trolling.

Behaviour and style reminds me a bit of the banned user "Sina".

Football was first played in England. Why are Brazilians better?


Sina was a terrible case.

I'd rather go with the AK-47 Kalashnikov

Because LAME makes encodes that are far better in audio quality AND tends to produce smaller files?

I searched in google & found this forum:
http://www.adobeforums.com/cgi-bin/webx?128@@.3bc3fc76

Sina

BTW, there IS a new Fraunhofer MP3 encoder - dated May 25 2007 - found by Alex B. It supports VBR and their own gapless implementation called OFL (Original File Length).

EDIT: Damn, this was old news.

STOP! STOP! STOP! Men, don't you see this guy is trolling?

I'm not trolling, just stop it, ah, forget about this post.

They have answered your question. You are the one that refuses to forget about it. Oh wait, now I did it.

Funny how this question comes up from time to time even if it may be trolling. I didn´t do any encoding with an Fraunhofer one for a while cause of its poor behaviour on transients but had to try this new eig sample floating around here.
Wow! mp3sencoder at highest vbr quality produces sounds like you press the buttons on an old typewriter were just short impulses should be

Jimpin if you're still reading this thread, check this MP3 Listening Test

I think Xing rules!!

OMG...couldn't resist. I feel so ashamed

Ok, thanks for infos.

well, the thing is that Franhofer has a ton of patents on their encoder, meaning that there are likely a bunch of compression techniques that they can use but the LAME team can't.

And this leads to... ?

Lame 10 x 0 Fraunhofer

[citation needed]

I think the majority rules here LAME has had much more development and offers excellent results. Stick to Lame 3.97 and you can't go wrong

MP3 is a patented technology, LAME necessarily uses patented methods.
Also, take in mind the broader range of the Fraunhofer product versus LAME, or even take in mind that it didn't have VBR mode until a few years ago. (Xing and LAME had it before)

Lame overall score : 3.74
FhG overall score : 3.27

Yeah that's the best way to settle this. BTW that test used Lame3.95 so it's improved a bit since then. In a later test Lame 3.96 scored 4.18 in a similar test (See http://www.rjamorim.com/test/multiformat128/results.html ). I don't think 3.97 or 3.98beta have had public listening tests at 128kbs yet, but I think there might be even more improvement now.

3.97b2 was tested in this test, and scored 4.60. However, I'm not sure about comparing scores from different tests. It was also approx. 128kbps VBR - not sure if that's relevant.

Yeah good point, different music selections and different anchors makes it pretty hard to be certain of exactly how they correlate. Still I'm pretty sure Lame hasn't gone backwards since 3.95.

Where do you actually get the Fraunhofer encoder from? I can't remember ever downloading it intentionally.

Would be fun to encode some uneasy music at 128kbs CBR with it in order to get that old-skool-MP3-feeling.

I believe software such as Windows Media Player, MusicMatch, and the free version of WinAmp all use the FhG mp3 encoder. You can also download a stand alone exe, it is floating around out there. Still, it is a waste of time. I think there was a time when the FhG encoder actually did a better job than Lame at low bitrates of around 64kbps and below. I don't know if that holds true anymore though as that was a long time ago during the Lame 3.90.3 era.

I have yet to read someone here post that LAME is superior than FhG at 128 CBR... (that is, using the latest FhG ACM bundled with WMP11)

Till then, the broad question of which is better is still unanswered.

Don't bash me on this, but there would be people who wants 128 CBR.

LAME is, of course, better than FhG at VBR.

I believe LAME has the edge over other mp3 encoders in the 128k range (VBR / ABR) and joint stereo. At 192k all decent encoders should do well so its not that clear. At higher bitrate (224~320k) vbr has no great advantage and all encoders should be transparent less a very rare sample. LAME gapless playback is also important.

There is also a degree of correlation in regards to quality and principles of design. LAME was probably the only mp3 implementation focusing on high quality

While it is true that LAME -V5 would have an advantage over FhG 128 CBR, LAME may not be said to be better in terms of forced 128 CBR.

same here, I think so since the first time that I used lame for encoding in CBR, the fhg encoder seems to produce better subjective quality in that setting

And why would you ever expect to? It makes very little sense to test codecs using substandard settings. Generally, you should compare them using optimal settings.



I doubt any of them really care about mp3 quality though.

It would matter for someone running an internet radio station.

I'm still expecting, because up to now it hasn't been proven. Till then, no encoder is better than the other across the full bitrate range. Remember, the OP's question has been very broad... Better at what? It doesn't mean that if one is better in high-bitrate VBR, it is better in in low-bitrate also. There still has been tunings for LAME for the low-bitrate encodings, so it still has room for improvement, even in your so-called "substandard settings."

No doubt I care about quality (LAME -V2, Vorbis -q5) but if you cannot convince other 128 CBR people who are happy with that given bitrate, might as well find the best encoder at that setting. (And don't just tell them encode at --preset standard or -V5).

I must say, up to now even FhG's new Mp3SEncoder at VBR mode is still wanting... It appears to be ABR though.

P.S. I'm not saying one of them is better. BTW, why was Jimpin was banned?

EDIT: Grammar... damn!

Taking into account mp3 limitations:

- For low <100 k encoding use another format
- Using the most common lossy bitrates of 130k~190k LAME vbr is competitive with the other formats. At very high bitrate of 224~320k I believe any half decent encoder is fully transparent less the odd killer samples (mainly pre-echo with mp3) which also plagues other encoders. I don't believe the other formats have any meaningfull advantage at very high bitrates.

We can narrow down the debate to 128~192 k range. One might test LAME vs fhg in this bitrate range. Since most people won't be able to do a non-killer sample test at 192k, It narrows down to just a 128k test.

I agree to shadowking. A head-to-head 128 CBR test between newest FhG ACM and LAME 3.97 would do. A VBR test using FhG fastencc is no match for LAME at ~128 range.

FHG is too. http://www.all4mp3.com/tools/sw_fhg_cl.html

I have done A LOT of encoding with this (and other) FHG encoder, and I think that with CBR it is at least comparable to LAME. I actually prefer FHG over LAME to encode 128k CBR for my portable player (please don't ask why I don't use lame VBR it's mostly because it produces low frequency hiss in certain parts of my music).
FHG VBR indeed sucks in high bitrates, but in the low bitrate range (below 112kbps) I think the results are quite respectable.