Fraunhofer's MP3 encoder
Mats Peterson
http://www.all4mp3.com/tools/sw_fhg_cl.html does a very good job, and
sounds much better than LAME in my book. It retains much of the depth
and air in the music, while the LAME output sounds comparatively flat
and lifeless. After all, Fraunhofer IIS has been a major participant in
the development of the MP3 format.
--
Mats Peterson
http://www.geocities.com/matsp888/
UnsteadyKen
> http://www.geocities.com/matsp888/
Wot! No Spam song in your Monty Python downloads?
--
Ken
Mats Peterson
> Mats Peterson wrote...
>
>> http://www.geocities.com/matsp888/
>
> Wot! No Spam song in your Monty Python downloads?
>
>
http://www.geocities.com/matsp888/python/spam.txt
;)
Mats Peterson
> UnsteadyKen <unstead...@gmail.com> wrote:
>> Mats Peterson wrote...
>>
>>> http://www.geocities.com/matsp888/
>>
>> Wot! No Spam song in your Monty Python downloads?
>>
>>
>
> http://www.geocities.com/matsp888/python/spam.txt
>
> ;)
>
Don't know if it's a "song", but that's what I've got...
Industrial One
UnsteadyKen
> http://www.geocities.com/matsp888/python/spam.txt
Doh! QRK 5
--
Ken
UnsteadyKen
> Don't know if it's a "song", but that's what I've got...
Oh, have you not heard it Mats? The Vikings are singing at
the top of their voices which is why they keep getting told
to shut up, then the sketch ends with "Lovely spam" etc
sung by a massed choir with bagpipes!
I'm going to try out the Fraunhofer you posted so I
could make you an mp3 if you would like a copy,
if you don't mind couple of pops and clicks.
--
Ken
Mats Peterson
> Mats Peterson wrote...
>
>> Don't know if it's a "song", but that's what I've got...
>
> Oh, have you not heard it Mats? The Vikings are singing at
> the top of their voices which is why they keep getting told
> to shut up, then the sketch ends with "Lovely spam" etc
> sung by a massed choir with bagpipes!
> I'm going to try out the Fraunhofer you posted so I
> could make you an mp3 if you would like a copy,
> if you don't mind couple of pops and clicks.
>
I have heard it. Regarding pops and clicks, that belongs to the older
versions.
UnsteadyKen
> Regarding pops and clicks, that belongs to the older
> versions.
Slap head again, not my day for thinking, yes my copy is 1973
and of course they are out on CD now, but you lose
the surprise of the double groove on Matching Tie.
Now that was a masterful piece of mastering.
I also see that the new fraunhofer is command line, I
don't think I could manage that.
_._
--
Ken
John Phillips
> The latest MP3 encoder from Fraunhofer IIS at
> http://www.all4mp3.com/tools/sw_fhg_cl.html does a very good job, and
> sounds much better than LAME in my book. It retains much of the depth
> and air in the music, while the LAME output sounds comparatively flat
Are you able to say what parameters you used with the two encoders
(and the LAME version) for the comprison? (Sorry to ask but I'm rather
interested in quality / bit-rate trade-off in codecs.)
--
John Phillips
Mats Peterson
> Mats Peterson wrote...
>
>> Regarding pops and clicks, that belongs to the older
>> versions.
>
> Slap head again, not my day for thinking, yes my copy is 1973
> and of course they are out on CD now, but you lose
> the surprise of the double groove on Matching Tie.
> Now that was a masterful piece of mastering.
>
> I also see that the new fraunhofer is command line, I
> don't think I could manage that.
>
> _._
>
Ah, I'm sorry Ken! I thought you were referring to the encoder when
you mentioned the pops and clicks... Anyway, I'm familiar with that
sketch, but thanks anyway :)
Mats Peterson
There aren't many parameters to talk about. Just give the -br and
possibly -q parameter to it, e.g. -br 256000 -q 1 for 256 kbps with
high quality (the quality switch only seems to make a difference for
bitrates below 192 kbps).
For LAME, I have used --preset cbr 128/160/192/256 etc. which I assume
will give the best possible constant bitrate encoding.
Mats Peterson
I should add that I have encoded a great deal of my FLACs into 256 kbps
MP3s, which is a decent quality/bit-rate trade-off for me. It certainly
sounds acceptable through speakers as background music.
PhattyMo
> The latest MP3 encoder from Fraunhofer IIS at
> http://www.all4mp3.com/tools/sw_fhg_cl.html does a very good job, and
> sounds much better than LAME in my book. It retains much of the depth
> and air in the music, while the LAME output sounds comparatively flat
> and lifeless. After all, Fraunhofer IIS has been a major participant in
> the development of the MP3 format.
>
I've used their codec ever since MP3 started out,and it was all *gasp*
DOS command line based!
And I agree,better than LAME. I'm not terribly impressed with the LAME
codec.
Mats Peterson
This one is command line based too, but that's what I prefer anyway. I
use it in a Perl script to convert from FLAC. I thought I was the only
one in the world thinking that LAME isn't the holy grail of encoders...
John Phillips
> John Phillips <news...@DontUseThis.mainly.me.uk> wrote:
>> On 2008-08-29, Mats Peterson <mats...@yahoo.com> wrote:
>>> The latest MP3 encoder from Fraunhofer IIS at
>>> http://www.all4mp3.com/tools/sw_fhg_cl.html does a very good job, and
>>> sounds much better than LAME in my book. It retains much of the depth
>>> and air in the music, while the LAME output sounds comparatively flat
>>> and lifeless. ...
>>
>> Are you able to say what parameters you used with the two encoders
>> (and the LAME version) for the comprison? (Sorry to ask but I'm rather
>> interested in quality / bit-rate trade-off in codecs.)
>
> There aren't many parameters to talk about. Just give the -br and
> possibly -q parameter to it, e.g. -br 256000 -q 1 for 256 kbps with
> high quality (the quality switch only seems to make a difference for
> bitrates below 192 kbps).
> For LAME, I have used --preset cbr 128/160/192/256 etc. which I assume
> will give the best possible constant bitrate encoding.
I'm afraid that gives too little precision for me to test your assertion
by experiment.
And I now see you posted the same initial message to two forums where
the reaction was just as might have been predicted. It's not good form
to engage in puffery in places where critical questions are likely to
be asked.
--
John Phillips
Mats Peterson
Now now, Einstein. This is about what we hear, not some figures. And I hope
you have understood that these are my subjective opinions. I can't
explain *why* I think it sounds better, ok? You might think otherwise,
so be it.
Bill Ruys
news:pltuk.1940$U5....@newsb.telia.net...
Well, I have no scientific data either, but I have always found the
Fraunhofer encoder to be sonically superior to the LAME encoder. The
differences in quality are marked at low bit-rates, such as 128 Kbps. The
high frequencies are less distorted and flanged. As you go up in bit-rate,
it gets harder to tell the difference, but they are perceivable up to about
256 Kbps. At 320 Kbps, I can't tell the difference.
Bill.
** Posted from http://www.teranews.com **
Schöön Martin
>
> Well, I have no scientific data either, but I have always found the
> Fraunhofer encoder to be sonically superior to the LAME encoder. The
> differences in quality are marked at low bit-rates, such as 128 Kbps. The
> high frequencies are less distorted and flanged. As you go up in bit-rate,
> it gets harder to tell the difference, but they are perceivable up to about
> 256 Kbps. At 320 Kbps, I can't tell the difference.
>
> Bill.
>
--
Martin Schöön <martin...@gmail.com>
"Problems worthy of attack
prove their worth by hitting back"
Piet Hein
Richard Crowley
> "Bill Ruys" writes:
>> Well, I have no scientific data either, but I have always found the
>> Fraunhofer encoder to be sonically superior to the LAME encoder. The
>> differences in quality are marked at low bit-rates, such as 128 Kbps.
>> The
>> high frequencies are less distorted and flanged. As you go up in
>> bit-rate,
>> it gets harder to tell the difference, but they are perceivable up to
>> about
>> 256 Kbps. At 320 Kbps, I can't tell the difference.
>>
http://en.wikipedia.org/wiki/Flanging
Schöön Martin
Mr.T
> UnsteadyKen <unstead...@gmail.com> wrote:
> but you lose
> > the surprise of the double groove on Matching Tie.
> > Now that was a masterful piece of mastering.
Or in fact just very old hat by that time. Disks with many seperate grooves
were common years before.
My grandfather had a horse racing disk with about 9 or 10 different winners.
I thought it was very clever when I was a child.
I've seen quite a few similar disks since then.
> > I also see that the new fraunhofer is command line, I
> > don't think I could manage that.
There are GUI front ends of course, and many rippers can pass the necessary
parameters. That's how many people use LAME in fact.
Whether the new Fraunhofer is any better is another matter requiring further
testing.
MrT.