hey guys. are there any new better compiles of lame other than dibrom's 3.90.2? any athlon xp optimized compiles? cuz aps takes a minute a song and thats sloowwwww

i think they are already athlon optimized
if you want faster encoding you should give mpc a try. it's amazingly fast but doesn't help you if you need MP3

already tried . I admit speeds and transparency is excellent and by #'z its more accurate than LAME's APS. However, i just got an AMAZING iPOD and i love it and due to compatibility issues i still use mp3. Also, i will use Mp3 just because it has a wide compatiblity and imo is just as transparent as mpc. How long does it take you guys to rip a song on aps (not apfasts)? I'm jw. and please post your processor

Try Gogo-no-coda if you want speed. Version 3.11 is based on LAME 3.90. I don't think it's got any of the --alt-presets in it, but still, it's close (in terms of quality) to the default settings of LAME and very fast. You'll be able to find it at mitiok's site.

gogo is based in Lame 3.88.

I get about 4X in my Athlon 1.33 GHz

I'd rather use APFS or APFE than Gogo.

3.88 just wont do when it comes to aps 3.90.2 is BEASTa!!!!!!!!!!!

Try out Easy CD-DA Extractor. I haven't used it for ages, but about a year or two ago it ripped my CDs to 192kbps CBR (good enough for me) with about 16x speed. Processor was (and still is) a 1.4GHz Athlon Thunderbird.

Why do people care about ripping/encoding speed that much?
Think abou it, you encode once and listen to it many many times, so what's the problem with investing 10 minutes more? I wouldn't want to sacrifice any quality for speed.

dev0

Dude,

His nickname is "feces1223"...

Ruairi

I use 3.93.1 with --preset medium -q 0. I get 0.9x encode speed on my 1.8ghz Athlon That's a saving of ~2kb/s but takes 4 times as long!

"Golden" words.

Feces1223, at Apr 8 2003 - 04:25 PM, you claimed APS took a minute a song (it takes me five). By Apr 8 2003 - 08:47 PM, upon claiming 3.90.2 is "BEAST", 258 minutes had passed.

You could have encoded 258 songs in that time. Thats ~20 albums. How many songs do you have to do!? How big is your harddrive where the WAV's are at the moment?

Surely you must go out or go to sleep for 4 or more hour periods, when you could leave the computer on and do another 20 albums...

the .wavs are temporary and get overwritten by the compressed aps tracks. i have 28 albums currently.

wise words dev0

patience is key if you want better quality... i would wait any amount of extra time if it meant i was getting a better quality encode

True Dat. I'll keep encoding in APS. Im 110% satisfied with results. and what else do i have to do other than wait a minute?

Why is there so much discussion about an optimized Lame 3.90.2 when at least Lame 3.93 is out now.

Is this a hint that 3.90.2 is the latest version that is really stable and well tested while anything newer is experimental. Or am I reading too far into things?

Postul8or:

New here?

3.93.1 is not tuned for the --alt-presets, 3.90.2 is the last one that was properly tuned. No one here really uses the latest verison of Lame. I understand there is some devolpment team friction and whatnot these days... Looks like 3.90.2 is the way to go for a while, till at least 3.94 comes out in final. Even then I'm doubtful.

I was around r3mix in the past, but still never as hardcore as other people in the scene.

Where should I head to get the super duper Dibrom compile of 3.90.2 then?

You should go back one step to the "General - (MP3)" and choose there "Sticky: List of recommended LAME compiles" topic. It's allmost first topic in that list. There you can find Lame.exe ja lame.dll in firs post.

3.93.1 is the latest stable release. although no excessive testing has been done with it, the presets and everything should be alright, because no major changes have been done to the psycho-acoustics.

and there is still no proof that sound quality is worse than before. but you can say for sure, that it isn't better.

"more tested" dosn't necessarily mean "better quality". 3.93.1 seems stable and quality wise equal. the same applies to 3.92.

cheers

askoff: Thanks for the advice on where to get the Lame Compiles.

amano & everyone else: Why would new Lame versions 3.91, 3.92, 3.93 etc. be released if there was no real evidence that they were a move forward? Either bug fixes, better quality better speed would be a reason for a new release in my mind. What's the case here?

No troubles here with the speed, I was run in a PIII 500 Mhz 128 MB RAM, the last year and the encoding speed with Lame 3.90.2 (APS) were at 1.3x, now I update my pc and runs a Atlhon 1800 XP (1533 Mhz), 384 MB RAM and the speed up to 6.5x when I encoding mp3 with LAME 3.90.2 (APS). I'm very satisfied with the speed now but with my old PIII I was happy too becouse the speed doesn't matter very much for me, only the quality.-

if your using the Command line mode versions. make sure you turned the Dos Prompt windows - IDLE SENSITIVITY - to "LOW" it might give ya a slight boost. or try Razorlame and use the REALTIME setting.
i just opt for CLI and adjust the IDLE, gives me a little more boost.

You'll find some 'inyeresting' timings here: http://www.hydrogenaudio.org/forums/index....80&hl=athlon+xp



I have a XP2000+

Definately new here

This has been the topic of long discussions before. For example, check out these threads:

http://www.hydrogenaudio.org/forums/index....=ST&f=15&t=3836
http://www.hydrogenaudio.org/forums/index....=ST&f=15&t=3920

newbie alert: how do you see the speed of encoding? Where is the 6.5x you say you encode at? i am using razorlame for encoding.

thank you and excuse me for my lame questions!

normaly it is output in the command prompt window. With Razorlame you can find it in the menu view - Last Log.


Pun intended of course
--
Ge Someone

Definitely a good thing to know...just because you see a new version of LAME doesn't mean it was tested properly and doesn't mean you should trust it until you hear about the tests being thorough!

Thanks for the info, I guess it's back to 3.90.2 for me.