re-encode question...
Nov 23, 2003 at 11:36 AM Thread Starter Post #1 of 24

linnem

100+ Head-Fier
Joined
Jun 1, 2002
Posts
214
Likes
0
Is it possible to reencode my eac/lame 256kbs vbr mp3s to 128kbs vbr without ripping all my cds again? (using eac/lame) -And if yes, will it be as good as if I started all over again?

Thanks.
 
Nov 23, 2003 at 11:44 AM Post #2 of 24
Quote:

Originally posted by linnem
Is it possible to reencode my eac/lame 256kbs vbr mp3s to 128kbs vbr without ripping all my cds again? (using eac/lame) -And if yes, will it be as good as if I started all over again?

Thanks.



linnem... no, not really. transcoding is just about always a bad idea. you are taking a lossy-type audio file and making another lossy-type file from it. I would expect the results to be noticeably worse than if you ripped at 128kbs directly from a CD. However, since you are really considering a downgrade to 128kbs (which I think is unlistenable), maybe you standards are not as high as mine and it would in fact be OK for you. Why not give it a try, and see what you think? And if you do, let us know!
biggrin.gif
 
Nov 23, 2003 at 1:02 PM Post #3 of 24
Yes, it's possible, no, it won't sound as good
 
HiBy Stay updated on HiBy at their facebook, website or email (icons below). Stay updated on HiBy at their sponsor profile on Head-Fi.
 
https://www.facebook.com/hibycom https://store.hiby.com/ service@hiby.com
Nov 23, 2003 at 3:22 PM Post #4 of 24
Quote:

Originally posted by linnem
Is it possible to reencode my eac/lame 256kbs vbr mp3s to 128kbs vbr without ripping all my cds again? (using eac/lame)


yes, dbpoweramp music converter will do the job easily.

Quote:


-And if yes, will it be as good as if I started all over again?


No, not as good, but it will be pretty close. I usually discourage transcoding, but this is one of the few cases where I feel it won't hurt a whole lot. You're going to lose a lot of quality just by using 128, starting with 256 mp3s instead of originals isn't going to make it all that much worse.

I'd suggest using --alt-preset ABR 128 (ABR is average bit rate) instead of CBR.

But try it out for yourself and see how it sounds with your music, to your ears, on your equipement.
 
Nov 24, 2003 at 1:21 PM Post #5 of 24
hanks for all your answers. Looks like I have to start all over again. I will keep my 256kbs files for my harddrive player and use the 128kbs for my solidstate player.
Damn.
 
Nov 24, 2003 at 5:06 PM Post #6 of 24
I'm going to go against the grain and say transcode. Save the time. If it was about quality, you wouldn't be using 128. The difference will unlikely be great. Certainly not enough to do a quality EAC rip. Does your player support any other format (Ogg, WMA, etc.)? Also is LAME even the best encoder for MP3's at 128? Either way do you have the space for "--alt-preset standard -Y", "--alt-preset normal", or "--r3mix". All should be better quality with little additional space.
 
Nov 24, 2003 at 6:20 PM Post #7 of 24
The concept of using LAME for 128kbps MP3 is lost on me as well. It's like saying "I want the best quality 128kbps MP3 possible"... when 128kbps is already considered unlistenable by just about any veterans here.

If 128kbps is needed (for size reason, seeing that it's a solid state player), may as well just use one of the faster encoders. LAME is not fast even encoding just 128kbps MP3's... using other encoder will just make it work a lot faster, and there's really not much quality to be gained by using LAME at 128kbps anyway.
 
Nov 24, 2003 at 7:41 PM Post #8 of 24
For my ears, I use --alt-preset standard with lame is the best quality to size compromise.

Scott
 
Nov 24, 2003 at 8:54 PM Post #9 of 24
Quote:

Originally posted by blessingx
I'm going to go against the grain and say transcode. Save the time. If it was about quality, you wouldn't be using 128. The difference will unlikely be great. Certainly not enough to do a quality EAC rip. Does your player support any other format (Ogg, WMA, etc.)? Also is LAME even the best encoder for MP3's at 128? Either way do you have the space for "--alt-preset standard -Y", "--alt-preset normal", or "--r3mix". All should be better quality with little additional space.


blessingx
i'm kinda in the same boat as linnem here and have been re-encoding my --aps mp3s as --ap 128 mp3s using razorlame. i'm going to try out a flash player with 256mb and want to squish a bit more music on it than i can with my --aps mp3s. so, my questio to you is: can you define the basic parameters for --apn and --r3mix, please. i'm assuming both are vbr. what's the target bit rate for them (as in 190kbps for --aps)? how much of a size difference will i really see? i've tried --aps -y, and don't see that it squishes the music down too much more than --aps. are the others better?
thanks!
nikolaus
 
Nov 24, 2003 at 9:12 PM Post #10 of 24
Quote:

Originally posted by matheis
blessingx
i'm kinda in the same boat as linnem here and have been re-encoding my --aps mp3s as --ap 128 mp3s using razorlame. i'm going to try out a flash player with 256mb and want to squish a bit more music on it than i can with my --aps mp3s. so, my questio to you is: can you define the basic parameters for --apn and --r3mix, please. i'm assuming both are vbr. what's the target bit rate for them (as in 190kbps for --aps)? how much of a size difference will i really see? i've tried --aps -y, and don't see that it squishes the music down too much more than --aps. are the others better?
thanks!
nikolaus


I think I can answer part of that question.

--r3mix was an earlier version (VBR) preset that's not so highly refined as the newer --alt -preset standard. IME, -aps works out to around 200k usually, --r3mix was around 180 or 190, if I recall correctly.

I encoded about 20 CDs with --ap 128 for my flash player and was satisfied with the quality for use while exercising.

Lame, --ap 128 just makes the music kinda "flat" sounding. When I tried Fhg at 128 the overall quality wasn't so flat, but there were really annoying artifacts occasionally. So at 128, I think it's a "pick your poison" situation.

[edit: fixed last paragraph to read --ap 128 not --aps]
 
Nov 24, 2003 at 9:25 PM Post #11 of 24
--r3mix has not been updated for years, and does not work as well as --aps. --r3mix usually boasts an average bitrate of ~185kbps, but it's extraordinarily higher or lower on some tracks.

- Chris
 
Nov 24, 2003 at 9:29 PM Post #12 of 24
earwax
thanks for answering part of the question. doesn't sound like --r3mix will save much room, with that little of a difference in kbps. i'm converting an album's worth of --aps mp3s to --r3mix to see what the difference is. from the visuals, it looks like the r3mix setting puts most of the frames in the 160 and 128 kbps range, with a little on either side. i'll post the size difference after it's complete.
on to the --ap 128 - sounded alright to you? doesn't have to be stunning, i guess. i'm working towards having a less and more portable set-up and am willing to accept the risks involved, ie. lower quality. i'm planning on the --ap 128 mp3s for headphone-based background music while i'm working.
later
nikolaus
 
Nov 24, 2003 at 10:43 PM Post #13 of 24
I don't have a LAME version that supports --apn (I use 3.92), but I did a test with Radioheads HTTT. Keep in mind it's relatively complex and long at nearly an hour, but this is what I found...

Original: Radiohead's Hail To The Thief 56.6 minutes 571.3 megs

LAME "--alt-preset fast standard" (each songs average kbps was 154-237) 77.8 mb BASE
LAME "--r3mix" (each songs average kbps was 137-232) 72.2 mb -7%
LAME "--alt-preset standard -Y" (each songs average kbps was 147-206) 70.0 mb -10%
Any codec 160 kbps CBR/ABR ~65.5 mb -16%

These differences may not seem much except on flash players, etc. where it may mean the difference between an extra album.
 
Nov 24, 2003 at 11:06 PM Post #14 of 24
well, i re-encoded gas: koeningsforst (some minimal, ambient electronic off the mille plateaux label)

here's what i came up with...

--aps (emusic) 91.3 MB ~200kbps on avg

--r3mix (cdex) 69.7 MB 21.6mb less 24% less ~152kbps on avg

--ap128 (razorlame) 58.0mb less 36% less

blessingx: thanks for the info. perhaps i'll try --ap160. that should allow some extra and not be too compromising, if --ap128 seems abhorrent ;p

later
nikolaus
 
Nov 24, 2003 at 11:18 PM Post #15 of 24
It's interesting to see differences based on complexity, etc. I really do think r3mix can still be useful in some cases, though it's possible -apn will finally replace it. In the ~128-160 range Ogg (q4-5) and AAC (QT), if possible, should really be looked at also. Matheis, any of your iRivers support the former?

Also remember the 128 bitrate test.

plot12z.png
 

Users who are viewing this thread

Back
Top