ok. here is the situation. i accidentally fdisked my c drive. i was trying to remove linux from a drive to install win98 (i know, i know). i ran fdisk on the d drive, but it fdisked the c. i guess many of you would say i got what i deserve for ditching linux in favor of windows. now onto my problem.

now that i have to rerip 5 gig of mp3s, i need a good encoder. i have heard good things about audiocatalyst i think. any suggestions. i was using musicmatch jukebox and it worked well until the last few cds that i copied. since i have to redo everything, i want to start off fresh with a good encoder.

thanks,
jeremy

oh, i wish i wish i had an empeg...

I use Audiograbber with either Audiocatalyst or the original l3enc encoder from franhouffer (spelling!)


Audiograbber's interface is lovely without being overly complex. It ripps and pulls info from CDDB. it has some nice options for normalization and file naming conventions. It's my favorite. The codec supplied from Franhoffer is foound for free by the way in the download for Windows Media Player. Microsoft Liscenced it for distribution. I don't know if it is less powerful than the one they had for sale themselves.


Empeg Kicks Ass
S/N 00203
http://www.iretro.com

The codec supplied from Franhoffer is foound for free by the way in the download for Windows Media Player. Microsoft Liscenced it for distribution. I don't know if it is less powerful than the one they had for sale themselves.

As I understand it, the command-line encoder that Fraunhofer sells has the high-quality encoding option and the Windows version of the codec doesn't have that option. Am I wrong? Does anyone else have any information about this?

-- Tony Fabris -- Empeg #144 --
Caution: Do not look into laser with remaining good eye.

I use Audiocatalyst and have no problems with it at all.

I have to admit that I'm probably not what you'd call a true audiophile... I bought my empeg based more on the sheer geek impulse to have something running Linux in my dash

But I have to say that I personally can't hear much difference between what Audiocatalyst produces and the original CDs (I use VBR at 'normal' quality for most things, although I would drop the quality a bit for 'speaking' CDs) Audiocatalyst has the added attraction of interfacing with the CDDB database, which saves me a lot of typing

Add to that the fact that Hugo and the Empeg guys recommend Audiocatalyst... that's good enough for me

Geoff
---- -------
Reg No. 554, s/n 00064 - It's mine I tell you.... all mine :)

Your question was well timed! In its February issue, the German computer magazine "c't" (the most professional computer magazine on the German market) did a big test of several MP3 encoders. To judge quality, they both used measurement equipment and subjective listening test. For those listening tests, they invited some of the best audiophile professionals.

They compared the Fraunhofer codec (professional version - the Microsoft version cannot encode to high bitrates), Xing, LAME, and as an implementation of the ISO source code, BladeEnc.

Result: If you encode with fixed bit rate, Fraunhofer is still best, very closely followed by LAME. Xing is somewhat weaker here. If you encode to VBR, Xing is best, again very closely followed by LAME. Fraunhofer has its problems with this. BladeEnc produces much worse results under any circumstances, due to serious error in the psycho-acoustic model it's based upon.

If you set Xing's VBR so that the file size is about the same as a Fraunhofer-encoded file with fixed bit rate, the Xing sounds a little better, they say.

Most people they asked couldn't tell if they were hearing a Fraunhofer 128kb-MP3 or the original CD. They used top-notch audio equipment for the tests. (Personal note: "most people" isn't good enough for me, and in my personal direct A-B tests with 128-kb-LAME encoded MP3s, I could hear a difference, so I decided to go with 160kb.)

The Fraunhofer codec (with high bitrate encoding) is available as part of MusicMatch Jukebox above version 4.2. The Xing codec was in MusicMatch until version 4.2, and is in AudioCatalyst. LAME comes in a DLL as part of AudioGrabber, which is an identical twin of AudioCatalyst (except for the codec) or as a stand-alone program for free.

So, if you're on a tight budget, get LAME. Otherwise try Xing with VBR or Fraunhofer with fixed bit rate.

Since I freely admit that I basically haven't a clue what I'm talking about I'm glad to see that this reinforces my choice of VBR with the Xing encoder (I'd hate to look like a complete idiot only 30 seconds after posting my opinion )

Geoff
---- -------
Reg No. 554, s/n 00064 - It's mine I tell you.... all mine :)

Most people they asked couldn't tell if they were hearing a Fraunhofer 128kb-MP3 or the original CD.

Interesting. I'm not much of an audiophile but I can clearly tell the difference. I haven't tried a blind test of myself on this, but all of my MP3's are done with Fraunhofer at 128kb, and I can hear the difference in certain kinds of high-frequency artifacts. Mostly white-noise stuff like cymbals or crowds cheering. The encoding adds kind of a swishy sound to the high stuff. I guess it's because that's technically random noise, and compression algorithms can't handle randomness very well.

I'll bet if I pointed it out to "most people", they'd be able to hear what I was talking about, once they knew what to listen for. Fortunately, it's very subtle and you have to listen closely to hear it, so doesn't bug me that much. Once you get used to it, it's not a problem.

Did that magazine mention whether or not Xing's VBR encoding sounded better on the high stuff like cymbals? Was there a magic bitrate in the Fraunhofer encoder where those artifacts disappeared?

Thanks for that report. I was looking for another side-by-side encoder comparison. Does this magazine have an online version of that article?




-- Tony Fabris -- Empeg #144 --
Caution: Do not look into laser with remaining good eye.

After encoding mp3 files using AudioCatalyst (VBR high quality) and being happy with that, I recently tested encoding with the WMA format using Audiograbber. To my ear this encoder deliver superior sound, most notably the high frequencies seem much crisper - and on the samt time smaller file sizes.

I mostly listen to classical music, and to a lot of piano music where the difference to my ear is considerable in favor of the WMA.

I sent a mail to the empeg team and got the answer that this codec hopefully will be supported when the new model ships.

So at present I encode both to mp3 and to WMA. Perhaps something to test?

Regards,
jhr

Empeg #315 - 20GB blue

I mostly listen to classical music, and to a lot of piano music where the difference to my ear is considerable in favor of the WMA.

My brother is a serious audiophile and a classical music buff, and he says that solo piano music is the single best way to test for compression artifacts.

Question for those "in the know"...

If JHR is noticing better quality on the WMA format, is the reason A) that the WMA format itself is actually better, or B) audiograbber's WMA encoder happens to have a better encoding algorithm?



-- Tony Fabris -- Empeg #144 --
Caution: Do not look into laser with remaining good eye.

Tony, to answer your questions: the c't folks found that the Fraunhofer dampens high frequencies more than Xing and LAME, but also produces less pre-echo. They didn't specify a "magic" bitrate where artefacts disappear.

As the article is in the current issue, it's not yet online. They may or may not put it on their web site (only some selected articles make it to the Internet). Theis URL is http://www.heise.de

Daniel

There is only one WMA encoder - MS's own. As far as I'm aware, the standard isn't published so noone can make "compatible" encoders (and it's not really worth it as they're free on the Windows platform, leaving little space for people to make money out of an alternative - eg like Xing's AC cf The Frauhofer encoder).

Hugo

In my personal experience I can definately tell the difference between 128, 128vbr, 160vbr, and 192vbr... 128 sounds really bad to me, while 160 is tollerable and 192 is almost-perfect..

Just my two cents worth.. All my newer cd's are encoded in 160vbr (audiocatalyst) or 192vbr depending on how much I like the cd..
-mark

...proud to have one of the first Mark I units

Hi.

> In my personal experience I can definately tell the difference between 128,
> 128vbr, 160vbr, and 192vbr... 128
> sounds really bad to me, while 160 is tollerable and 192 is almost-perfect..

Just out of interest, why is there a difference between 128vbr and 192vbr?
I use LAME, and I tell it to use 64kbps with vbr (default q (I'll have
to try some tests with different qs and see if I can distinguish them)).

Virtually all of the frames are encoded at 128kbps or above, which is
what I'd expect.

Doesn't the vbr in your tool decide what is the most appropriate bit rate?

Richard.