Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Lame 3.97 beta3 released (Read 81089 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Lame 3.97 beta3 released

Reply #100 –
Oh, well... No problem. It's not THAT much slower...  I am still going to use 3.97b3 from RareWares.
lame -V 0

Lame 3.97 beta3 released

Reply #101 –
Thanks for looking into it John. Here is another test on a Pentium M Centrino 1.8 notebook using the following commandline:
Code: [Select]
-V 2 --vbr-new --nohist --noreplaygain


3.97.b2

Code: [Select]
LAME 3.97 (beta 2, Nov 29 2005) 32bits (http://www.mp3dev.org/)
CPU features: MMX (ASM used), SSE (ASM used), SSE2
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding C:\dub_aug_9.wav
      to C:\dub_aug_9.mp3
Encoding as 44.1 kHz VBR(q=2) j-stereo MPEG-1 Layer III (ca. 7.3x) qval=3
    Frame          |  CPU time/estim | REAL time/estim | play/CPU |    ETA
  8424/8424  (100%)|    0:12/    0:12|    0:12/    0:12|   18.149x|    0:00
Writing LAME Tag...done


3.97.b3

Code: [Select]
LAME 3.97 (beta 3, Aug 22 2006) 32bits (http://www.mp3dev.org/)
CPU features: MMX (ASM used), SSE (ASM used), SSE2
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding C:\dub_aug_9.wav
      to C:\dub_aug_9.mp3
Encoding as 44.1 kHz VBR(q=2) j-stereo MPEG-1 Layer III (ca. 7.3x) qval=3
    Frame          |  CPU time/estim | REAL time/estim | play/CPU |    ETA
  8424/8424  (100%)|    0:14/    0:14|    0:14/    0:14|   14.919x|    0:00
Writing LAME Tag...done


3.98a6

Code: [Select]
LAME 3.98 (alpha 6, Sep 13 2006 18:54:54) 32bits (http://www.mp3dev.org/)
warning: alpha versions should be used for testing only
CPU features: MMX (ASM used), SSE (ASM used), SSE2
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding C:\dub_aug_9.wav
      to C:\dub_aug_9.mp3
Encoding as 44.1 kHz VBR(q=2) j-stereo MPEG-1 Layer III (ca. 7.3x) qval=3
    Frame          |  CPU time/estim | REAL time/estim | play/CPU |    ETA
  8424/8424  (100%)|    0:16/    0:16|    0:16/    0:16|   13.004x|    0:00
Writing LAME Tag...done

Lame 3.97 beta3 released

Reply #102 –
The tests on intel M processors give same results as tests on P-III processors because AFAIK intel M CPU derived from P-III and not P4... So maybe for the sake of centrino laptop users (and us poor P-III users too  ) it would be nice if there would be two different biulds with every release. 
Meanwhile, Julien, if you encode maaany files and every little speedup is welcome to you, you could use compiles from http://lame.bakerweb.biz/.

P.S.: Anyone interrested in results with VIA C3 1 Giga PRO CPU?
lame -V 0

Lame 3.97 beta3 released

Reply #103 –
Quote from: psycho on
The tests on intel M processors give same results as tests on P-III processors because AFAIK intel M CPU derived from P-III and not P4... So maybe for the sake of centrino laptop users (and us poor P-III users too  ) it would be nice if there would be two different biulds with every release. 
Meanwhile, Julien, if you encode maaany files and every little speedup is welcome to you, you could use compiles from http://lame.bakerweb.biz/.

P.S.: Anyone interrested in results with VIA C3 1 Giga PRO CPU?


Thanks for the heads up. It is not so much that I encode a lot of files, but I was originaly surprised by the pretty noticeable decrease in speed.

Lame 3.97 beta3 released

Reply #104 –
Thanks again for your feedback. 

I've just uploaded some fresh alpha6 compiles that have been tested on Athlon64, P4 Prescott and PIII Coppermine. All provide a similar speed to the 'Bakerweb' compiles but with the ICL9.1 compiler rather than the 4.5 compiler. I don't have a laptop to test on, so you'll have to tell me how that goes!

Lame 3.97 beta3 released

Reply #105 –
Quote from: john33 on
Thanks again for your feedback. 

I've just uploaded some fresh alpha6 compiles that have been tested on Athlon64, P4 Prescott and PIII Coppermine. All provide a similar speed to the 'Bakerweb' compiles but with the ICL9.1 compiler rather than the 4.5 compiler. I don't have a laptop to test on, so you'll have to tell me how that goes!



John,

Here is a test using the same notebook(Pentium M Centrino 1.8) and the same command line (V2, no hist, no replaygain):

Lame 3.98a6 from Rarewares

Code: [Select]
LAME 3.98 (alpha 6, Sep 14 2006 12:09:42) 32bits (http://www.mp3dev.org/)
warning: alpha versions should be used for testing only
CPU features: MMX (ASM used), SSE (ASM used), SSE2
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding C:\dub_aug_9.wav
      to C:\dub_aug_9.mp3
Encoding as 44.1 kHz VBR(q=2) j-stereo MPEG-1 Layer III (ca. 7.3x) qval=3
    Frame          |  CPU time/estim | REAL time/estim | play/CPU |    ETA
  8424/8424  (100%)|    0:14/    0:14|    0:14/    0:14|   15.393x|    0:00
Writing LAME Tag...done



Lame 3.98a6 from http://lame.bakerweb.biz/ [Wednesday, September 13, 2006]

Code: [Select]
LAME 3.98 (alpha 6, Sep 13 2006 06:00:51) 32bits (http://www.mp3dev.org/)
warning: alpha versions should be used for testing only
CPU features: MMX (ASM used), SSE (ASM used), SSE2
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding C:\dub_aug_9.wav
      to C:\dub_aug_9.mp3
Encoding as 44.1 kHz VBR(q=2) j-stereo MPEG-1 Layer III (ca. 7.3x) qval=3
    Frame          |  CPU time/estim | REAL time/estim | play/CPU |    ETA
  8424/8424  (100%)|    0:13/    0:13|    0:13/    0:13|   15.985x|    0:00
Writing LAME Tag...done

Lame 3.97 beta3 released

Reply #106 –
Thanks a lot, Julien.  I think I call that close enough.  Basically, I'm trying to move away from using the old 4.5 compiler. There will be small difference in speed - a number of the compile options used with 4.5 are now obsolete, particularly in the area of optimising for the older processors, but overall, I think these compiles are sufficient unto the purpose!!

I'll rework the beta3 compiles along the same lines and upload again later.

Lame 3.97 beta3 released

Reply #107 –
Quote from: john33 on
Thanks a lot, Julien.  I think I call that close enough.  Basically, I'm trying to move away from using the old 4.5 compiler. There will be small difference in speed - a number of the compile options used with 4.5 are now obsolete, particularly in the area of optimising for the older processors, but overall, I think these compiles are sufficient unto the purpose!!

I'll rework the beta3 compiles along the same lines and upload again later.


John,

thank you again for having looked into it.

Lame 3.97 beta3 released

Reply #109 –
Quote from: john33 on
Quote from: Julien on

John,

thank you again for having looked into it.

 

Revised beta3 compiles now at Rarewares.



The ultimate test:)

Rarewares compile

Code: [Select]
LAME 3.97 (beta 3, Sep 14 2006) 32bits (http://www.mp3dev.org/)
CPU features: MMX (ASM used), SSE (ASM used), SSE2
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding C:\dub_aug_9.wav
      to C:\dub_aug_9.mp3
Encoding as 44.1 kHz VBR(q=2) j-stereo MPEG-1 Layer III (ca. 7.3x) qval=3
    Frame          |  CPU time/estim | REAL time/estim | play/CPU |    ETA
  8424/8424  (100%)|    0:13/    0:13|    0:13/    0:13|   16.413x|    0:00
Writing LAME Tag...done
-------------------------------------------------------------------------------


Bakerweb compile

Code: [Select]
LAME 3.97 (beta 3, Sep 14 2006) 32bits (http://www.mp3dev.org/)
CPU features: MMX (ASM used), SSE (ASM used), SSE2
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding C:\dub_aug_9.wav
      to C:\dub_aug_9.mp3
Encoding as 44.1 kHz VBR(q=2) j-stereo MPEG-1 Layer III (ca. 7.3x) qval=3
    Frame          |  CPU time/estim | REAL time/estim | play/CPU |    ETA
  8424/8424  (100%)|    0:12/    0:12|    0:12/    0:12|   17.133x|    0:00
Writing LAME Tag...done
-------------------------------------------------------------------------------



It's still slower than b2, but it may be attributable to the tuning rather than the optimizing.

Thanks again.

Lame 3.97 beta3 released

Reply #111 –
I have installed the beta version and the rip speed is down to 3.9x. Anyone know why this is, and a possible solution please?

Lame 3.97 beta3 released

Reply #112 –
Quote from: Big Fella on
I have installed the beta version and the rip speed is down to 3.9x. Anyone know why this is, and a possible solution please?
Ripping speed? This has probably nothing to do with LAME (which is only the encoder). What ripping application? EAC? Have you made tests with other CDs? But I'm afraid we are highly OT within this thread...

Cheers.

Sergio
Sergio
M-Audio Delta AP + Revox B150 + (JBL 4301B | Sennheiser Amperior | Sennheiser HD598)

Lame 3.97 beta3 released

Reply #113 –
Quote from: smz on
Quote from: Big Fella on
I have installed the beta version and the rip speed is down to 3.9x. Anyone know why this is, and a possible solution please?
Ripping speed? This has probably nothing to do with LAME (which is only the encoder). What ripping application? EAC? Have you made tests with other CDs? But I'm afraid we are highly OT within this thread...

Cheers.

Sergio


Sorry yes I use EAC, same with all other cd's as well, most I get is 3.9, with itunes I get 15x.

Lame 3.97 beta3 released

Reply #114 –
I just set up EAC 0.95 Beta 4 with LAME 3.97 Beta 3 as per the instructions here.

The CD is ripped fine, but when it goes to encode with LAME, a window flashes open for a moment then immediately closes. The resulting files are just the ripped .wavs.

I tried changing the command line switch, with the same result.

Anyone know what's going on?

edit: Also, anyone know where I can find the 3.97 b2 or older lame rars for Windows?  Rarewares only seems to have older versions for Linux or Mac.

Lame 3.97 beta3 released

Reply #115 –
Check your command line options, that usually happens when there is a mistake inside the command

Rarewares has 3.96.1 for Windows, scroll down the page a bit more

Lame 3.97 beta3 released

Reply #116 –
Quote from: CrazyCanuck on
I just set up EAC 0.95 Beta 4 with LAME 3.97 Beta 3 as per the instructions here.

The CD is ripped fine, but when it goes to encode with LAME, a window flashes open for a moment then immediately closes. The resulting files are just the ripped .wavs.

I tried changing the command line switch, with the same result.

Anyone know what's going on?

edit: Also, anyone know where I can find the 3.97 b2 or older lame rars for Windows?  Rarewares only seems to have older versions for Linux or Mac.
Could be the very common genere tag prob: just use --genre-list to see if the one you've selected is okay and/or simply add --ignore-tag-errors to your commadn line.

A few useful LAME version can be found here: http://homepage.ntlworld.com/jfe1205/LAME/
WavPack 5.7.0 -b384hx6cmv / qaac64 2.80 -V 100

Lame 3.97 beta3 released

Reply #117 –
Sorry for this small off-topic. Does anyone know where the trumpet.wav sample comes from?

Thank you.
Liberate tutemae ex inferis

Lame 3.97 beta3 released

Reply #118 –
I noticed the Linux static compile is still on beta1. Any reason for this?

Lame 3.97 beta3 released

Reply #119 –
bleh, can someone summarize this thread? so is this new beta any good?

Lame 3.97 beta3 released

Reply #120 –
Quote from: adlai on
bleh, can someone summarize this thread? so is this new beta any good?

From what I've seen in this thread, LAME 3.97 beta3 is the new Hydrogenaudio-recommended version  .
 

Lame 3.97 beta3 released

Reply #121 –
As we are ahead of a new wavPack version which I am using in lossy mode apart from mp3 I am reconsidering my settings. For my best music (sound quality as well as musical quality) I use wavPack @352 kbps, and for anything else I use mp3 @270 kbps. But with my last encodings I felt I should use a lower bitrate with mp3 for a better scaling within my two encoding schemes. Moreover mp3 is excellent also at a lower bitrate. So I intend to use mp3 @ 230 kbps.
I tried my usual bad samples trumpet, herding_calls, harp40_1 as well as Wombat's Birds.flac using 3.90.3 and 3.97b3, both with --alt-preset 230 --lowpass 17.0 (I personally don't care about such a lowpass - my ears are nearly 57 years old).
I am very content with the results, for 3.97b3 as well as 3.90.3 (which I did not expect for 3.97). No issue for practical listening conditions to me.
lame3995o -Q1.7 --lowpass 17