Help - Search - Members - Calendar
Full Version: --alt-preset standard VS --alt-preset cbr 256
Hydrogenaudio Forums > Lossy Audio Compression > MP3 > MP3 - General
Tunnel
Hello there! smile.gif

I've been thinking of which --alt-preset of LAME 3.90.3 to use. I've already ripped many CD's in --alt-preset cbr 256 (with EAC) because I know CBR files won't cause any incompatibility issues (wrong length displaying etc.). And I know CBR 256 should be totally transparent to human ear (using a good encoder). But the average bitrate of --alt-preset standard - roughly ~192 kbps - makes a very big filesize difference. Though I have a high-end professional soundcard I don't have great speakers nor headphones (yet).

So I'd like you to tell me which one, --alt-preset cbr 256 or --alt-preset standard, gives better overall sound quality, using LAME 3.90.3 or if they give roughly the same quality at different filesizes or if the quality difference is trifling considering the big filesize difference.

Thanks a million in advance! wink.gif
Tanel
Jebus
Better quality with --alt-preset standard. There are certain bits of just about every song that require 320kbps frames for transparency... even though the bit reservoir will help you when using CBR, it isn't ideal. And for what it's worth, i've never personally used a single program or piece of MP3 hardware which has problems with VBR files.
Single
QUOTE(Tunnel @ Mar 21 2004, 01:38 AM)
Hello there! smile.gif

So I'd like you to tell me which one, --alt-preset cbr 256 or --alt-preset standard, gives better overall sound quality, using LAME 3.90.3 or if they give roughly the same quality at different filesizes or if the quality difference is trifling considering the big filesize difference.


If you care a lot about VBR compatibility the cbr 256 may be the best compromise choice. But I think if you are have not hardware players that have problems with VBR, you don't need to care a lot about VBR mode. Just do't buy player with problems. smile.gif

Another side: "--alt-preset standard" are clear for many people. But if you planning to use some stereo expanders, surround kits, dolby prologic recievers and any matrix decoders don't encode with VBR mode. VBR mode brings artifacts to "surround chanel" much bigger than CBR. Lame VBR encoded mp3s is worst choise with any stereo "expanders" and upmixing technologies. And in what i am hearing ABR mode brings less artifacts than CBR. For me I choose ABR 256. It smaller that 320 about 20% and bigger than aps about 20% and it is sounds better than 256, narly as 320. Even if you try lower abr bitrates you will see that codec fly up to 320kbps with problem samples.

I used 3.93.1. I not nearing any differences between 3.90.3 and 3.93.1 in cbr and abr modes, but 3.93.1 is faster than 3.90.3.
Tunnel
Thanks for replying, guys! smile.gif

But Single, what about using ABR with matrix, stereo expanders etc.? I know ABR is variable bitrate but with given average.. So it should add similar artifacts as VBR, right?
LoFiYo
QUOTE(Jebus @ Mar 20 2004, 05:40 PM)
Better quality with --alt-preset standard. There are certain bits of just about every song that require 320kbps frames for transparency... even though the bit reservoir will help you when using CBR, it isn't ideal.

Hi, I know almost all the oldtimers here have that same opinion, and I tend to agree on that. But I would like to see it for myself. If there are some publicly available samples where APS is clearly superior to CBR256, I would be very interested in ABX-testing them. Do you know of any (and its location)? Or do you personally have such samples? smile.gif
Wombat
@LoFiYo

Try spahm.wav first and then other samples with much transients if you have some sparetime.

Wombat
Single
QUOTE(Tunnel @ Mar 21 2004, 05:04 AM)
But Single, what about using ABR with matrix, stereo expanders etc.? I know ABR is variable bitrate but with given average.. So it should add similar artifacts as VBR, right?

Hm, I am sorry for my english it is not native for me. I am not developer and i can not explain lame behavior. But, ABR mode produces LESS artifacts than CBR and it is easy to abx. VBR mode produce significantly MORE artifacts than CBR (and ABR of course) even in very high bitrates.
Moneo
QUOTE(Single @ Mar 21 2004, 01:50 AM)
But if you planning to use some stereo expanders, surround kits, dolby prologic recievers and any matrix decoders don't encode with VBR mode.

Better yet, don't use lossy compression at all, as it isn't meant to be processed by any filters.

Or apply the effects before you encode.
Single
QUOTE(Moneo @ Mar 21 2004, 09:16 PM)
Better yet, don't use lossy compression at all, as it isn't meant to be processed by any filters.

Or apply the effects before you encode.

It is very hard to apply dpl II or surround reciever before encoding. But it is very easy do not destroy surround by not using VBR mode but CBR or ABR. smile.gif
indybrett
QUOTE(Single @ Mar 21 2004, 01:05 PM)
VBR mode produce significantly MORE artifacts than CBR (and ABR of course) even in very high bitrates.

That is complete utter nonsense.
PowerPigg
@indibrett

I think he meant in context of surround-encoded or stereo-expanded source material, not normal 2-channel music or standard compression artifacts. Might be possible given the differences between how ABR and VBR work.

I remember people mentioning this limitation to LAME VBR in a couple of previous posts, and I think some one even explained why this might be the case, but no clear cut answer was given on that. It was a thread I stumbled into several weeks ago here in HA.
indybrett
@PowerPigg

You're probably right about that, and if that's the case I apologize. I just didn't see the original poster asking about surround sound, but only about which was better. I didn't want anyone to get the impression that CBR256 was better than --alt-preset standard.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2008 Invision Power Services, Inc.