Please don't ask for a finite date as that is not cast in stone but it will almost certainly be within a couple of weeks.
Full Version: LAME 3.98 Release
As it would appear that we are quite close (probably a week, or two) to a release of 3.98 final, I have decided that it makes no sense continually to update the beta compiles at Rarewares, but to await the final for the next set of compiles. The CVS is changing quite rapidly, as you would expect, and daily updates make no sense at all. If this should be delayed in any serious way, I'll maybe think again, but at the moment continual updates are a bit pointless.
Please don't ask for a finite date as that is not cast in stone but it will almost certainly be within a couple of weeks.
Please don't ask for a finite date as that is not cast in stone but it will almost certainly be within a couple of weeks.
QUOTE (john33 @ Jun 27 2008, 11:58) 
As it would appear that we are quite close (probably a week, or two) to a release of 3.98 final ...
Great to hear 3.98 will be final soon.
Excellent 
I have a question for you John ; will you add flac support as input for Lamedrop (as oggdrop) ?
That would be nice.
Anyway, thanks for your apps
I have a question for you John ; will you add flac support as input for Lamedrop (as oggdrop) ?
That would be nice.
Anyway, thanks for your apps
QUOTE (le_canz @ Jun 27 2008, 13:04)
.....
I have a question for you John ; will you add flac support as input for Lamedrop (as oggdrop) ?
That would be nice.
.....
I have a question for you John ; will you add flac support as input for Lamedrop (as oggdrop) ?
That would be nice.
.....
I know I've been promising that for some time, but I'm afraid it's not there yet.
Well, that's not a big deal, Lamedrop is already a time saving app anyway.
Thanks for your fast answer
Thanks for your fast answer
QUOTE (john33 @ Jun 27 2008, 05:58)
Please don't ask for a finite date as that is not cast in stone but it will almost certainly be within a couple of weeks.
If you go to the LAME Project homepage, click the About LAME box and then click on the "History and ChangeLog" link, you can see that LAME 3.98 will be released on July 4: History (of LAME)
Question of that day will be... is the new recommended Lame version going to be 3.98 or 3.97?
Good work on rarewares, I always get the FLAC, Lame and Ogg compiles from it
Good work on rarewares, I always get the FLAC, Lame and Ogg compiles from it
Time for a new round of listening tests, methinks. Either that, or everybody buys halb27 and guruboolez (and several other people! and the LAME team!) a few cases of beer.
QUOTE (Axon @ Jun 27 2008, 19:21)
Time for a new round of listening tests, methinks. Either that, or everybody buys halb27 and guruboolez (and several other people! and the LAME team!) a few cases of beer.
Yes, I will do my usual tests again, though I'm more and more aware that they are so easily misleading. Those samples with the clearest results for judging about improvements/regressions (for instance eig as an extreme pre-echo sample) potentially have the least practical impact (at least I encounter extremely rarely a pre-echo problem in practice). Those problems however which are most significant to me (my personal tonal problems) I often cannot repeatedly ABX though I'm sure meanwhile the issue exists (it's subtle of course though).
So I hope a lot of members will do their listening tests with 3.98, and of course it would be great if guruboolez would return here for listening tests.
QUOTE (le_canz @ Jun 27 2008, 09:04)
Excellent
I have a question for you John ; will you add flac support as input for Lamedrop (as oggdrop) ?
That would be nice.
Anyway, thanks for your apps
I have a question for you John ; will you add flac support as input for Lamedrop (as oggdrop) ?
That would be nice.
Anyway, thanks for your apps
I don't know about John's Windows utilities, but our current version of lame supports flac input (and many other input types like those IMA ADPCM files that you get from the popular S1MP3 players when recording). It does this due to new libsndfile1 support (as opposed to libsndfile0 that lame used to have up to 3.97).
In other words, you may get flac input "for free" if there is a libsndfile1 for Windows. Sorry that I don't know anything about Windows and the utilities you guys use.
Hope this helps, Rogério Brito.
QUOTE (Axon @ Jun 27 2008, 14:21)
Time for a new round of listening tests, methinks. Either that, or everybody buys halb27 and guruboolez (and several other people! and the LAME team!) a few cases of beer.
It will be a bit easier for Unix users to perform abx listening tests with the command line utility provided with lame (in the misc directory). I have fixed some of the problems that prevented it to work on my Linux systems and I think that the fix is general enough for other flavors of Unix.
I know that there is the Java abx utility, but since we still don't have Java for all platforms, this seems to be a good substitute, so that more people can perform tests (me for instance
).Regards, Rogério Brito.
I wanted to carry out the MP3 listening test once 3.98 is available as final version. Maybe we could start discussing samples now.
QUOTE (rbrito @ Jun 29 2008, 00:32)
I don't know about John's Windows utilities, but our current version of lame supports flac input (and many other input types like those IMA ADPCM files that you get from the popular S1MP3 players when recording). It does this due to new libsndfile1 support (as opposed to libsndfile0 that lame used to have up to 3.97).
In other words, you may get flac input "for free" if there is a libsndfile1 for Windows. Sorry that I don't know anything about Windows and the utilities you guys use.
Hope this helps, Rogério Brito.
In other words, you may get flac input "for free" if there is a libsndfile1 for Windows. Sorry that I don't know anything about Windows and the utilities you guys use.
Hope this helps, Rogério Brito.
Thanks for this, I'll try to find some more informations about this
I just noticed that the official LAME History page has a release date for 3.98 final as July 4, 2008.
Hopefully this is an actual date, as I finally have my CDs ripped to lossless and would love to set my computer up over the long weekend to transcode my library (currently various versions of LAME, iTunes AAC and FhG) to LAME 3.98 final.
Hopefully this is an actual date, as I finally have my CDs ripped to lossless and would love to set my computer up over the long weekend to transcode my library (currently various versions of LAME, iTunes AAC and FhG) to LAME 3.98 final.
Am looking foward see the speed improvements on the stable non-debug version, I know the beta version is a tad faster then 3.97 (24x vs 25x or sometimes 26x). Also the new tagging features on 3.98 or great, such as album art and adding any tag frames from the ID3 spec.
I will try to test both 3.97 and 3.98 after 3.98 is released, but I sorter have mixed results with my new Sennheiser HD 215 (almost no bass at all) and my old Sennheiser HD 202 (sounds better then my new ones). On my HD 215, I can pick up warbling artifacts very easy or hard to hear artifacts. On my old HD 202, I am more likely to pick up precho artifacts then on my HD 215.
I will try to test both 3.97 and 3.98 after 3.98 is released, but I sorter have mixed results with my new Sennheiser HD 215 (almost no bass at all
) and my old Sennheiser HD 202 (sounds better then my new ones). On my HD 215, I can pick up warbling artifacts very easy or hard to hear artifacts. On my old HD 202, I am more likely to pick up precho artifacts then on my HD 215.
QUOTE (Sebastian Mares @ Jun 29 2008, 02:27)
I wanted to carry out the MP3 listening test once 3.98 is available as final version. Maybe we could start discussing samples now.
That would be a smart idea. The sooner we end the discussion about samples, the sooner the actual test can start.
QUOTE (/mnt @ Jun 30 2008, 18:04)
I will try to test both 3.97 and 3.98 after 3.98 is released, but I sorter have mixed results with my new Sennheiser HD 215 (almost no bass at all ) and my old Sennheiser HD 202 (sounds better then my new ones). On my HD 215, I can pick up warbling artifacts very easy or hard to hear artifacts. On my old HD 202, I am more likely to pick up precho artifacts then on my HD 215.
) and my old Sennheiser HD 202 (sounds better then my new ones). On my HD 215, I can pick up warbling artifacts very easy or hard to hear artifacts. On my old HD 202, I am more likely to pick up precho artifacts then on my HD 215.I have noticed something differently than you: with headphones that have no bass or which inflate the higher frequencies, I am more likely to pick up the artifacts.
But since I am no "golden ears", I can't ABX the original to some of the samples encoded with "lame --resample 44100 -V 7" (lame 3.98, that is---that's one of the reasons why I fixed the abx tool that comes with lame), much to my shame.
Regards, Rogério Brito.
QUOTE (rbrito @ Jun 30 2008, 23:22)
QUOTE (/mnt @ Jun 30 2008, 18:04)
I will try to test both 3.97 and 3.98 after 3.98 is released, but I sorter have mixed results with my new Sennheiser HD 215 (almost no bass at all
) and my old Sennheiser HD 202 (sounds better then my new ones). On my HD 215, I can pick up warbling artifacts very easy or hard to hear artifacts. On my old HD 202, I am more likely to pick up precho artifacts then on my HD 215.I have noticed something differently than you: with headphones that have no bass or which inflate the higher frequencies, I am more likely to pick up the artifacts.
But since I am no "golden ears", I can't ABX the original to some of the samples encoded with "lame --resample 44100 -V 7" (lame 3.98, that is---that's one of the reasons why I fixed the abx tool that comes with lame), much to my shame.
Regards, Rogério Brito.
Its kinda odd, most of the precho artifacts I can hear and ABX on my HD202 and its not present on my HD215 headphones. But on my HD215, I have picked up and ABX a few tracks that have warbling or some really bad precho that is not present on my HD202. I know that the Sennheiser HD215 has a higher freq response, but hardly any bass at all, but the sound is alot more clear and warbling artifacts sound alot worse then my old HD202.
Can't wait. 
yes, 3.98 will definetly be a nice version. I think -V2.5 will be my usual preset then.
And if Nero AAC next version comes out soon, as is said silently in a few posts, we can include it in the listening test.
No, because it will be an MP3 test.
Will there be any way to determine the fractional VBR quality of a LAME MP3 from its headers? I don't see the value recorded anywhere. Can it be reliably and accurately determined from the lowpass value?
Also, why is the floating point value just casted to an int? Was it intentional to make -V2.9999 have an integer VBR quality of 2?
Also, why is the floating point value just casted to an int? Was it intentional to make -V2.9999 have an integer VBR quality of 2?
The 3.98 final sources are now available!
http://sourceforge.net/project/showfiles.p...;package_id=309
http://sourceforge.net/project/showfiles.p...;package_id=309
A full set of win32 compiles is now available on Rarewares.
Enjoy!!
Enjoy!!
QUOTE (john33 @ Jul 4 2008, 09:07)
A full set of win32 compiles is now available on Rarewares.
Enjoy!!
Well done that man! Congratulations on the new release!!
Enjoy!!
QUOTE (john33 @ Jul 4 2008, 09:07)
A full set of win32 compiles is now available on Rarewares.
Enjoy!!
Enjoy!!
Thanks for the builds, well done.
I just had a quick test on the encoding on 3.98, and it is faster then 3.97. LAME 3.97 encodes at around 25x, while 3.98 encodes at around 29x - 30x, on my computer.
CODE
Microsoft Windows XP [Version 5.1.2600]
© Copyright 1985-2001 Microsoft Corp.
C:\>cd c:\Downloads\lame3.98
C:\Downloads\lame3.98>lame.exe -V 2 C:\Temp\Linchpin.wav "C:\Temp\Linchpin 3.98.
mp3"
LAME 3.98 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:\Temp\Linchpin.wav to C:\Temp\Linchpin 3.98.mp3
Encoding as 44.1 kHz j-stereo MPEG-1 Layer III VBR(q=2)
Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
7850/7850 (100%)| 0:06/ 0:06| 0:06/ 0:06| 29.693x| 0:00
32 [ 28] *
40 [ 0]
48 [ 0]
56 [ 0]
64 [ 0]
80 [ 0]
96 [ 2] *
112 [ 25] *
128 [ 89] %**
160 [ 938] %*********************
192 [3014] %%%*****************************************************************
224 [2513] %%%%*****************************************************
256 [ 822] %%%%***************
320 [ 419] %%%*******
-------------------------------------------------------------------------------
kbps LR MS % long switch short %
210.4 7.3 92.7 92.7 4.1 3.2
Writing LAME Tag...done
ReplayGain: -10.3dB
C:\Downloads\lame3.98>cd "c:\Program Files\LAME"
C:\Program Files\LAME>lame.exe -V 2 --vbr-new C:\Temp\Linchpin.wav "C:\Temp\Linc
hpin 3.97.mp3"
LAME 3.97 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:\Temp\Linchpin.wav to C:\Temp\Linchpin 3.97.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
7850/7850 (100%)| 0:08/ 0:08| 0:08/ 0:08| 24.901x| 0:00
32 [ 28] *
40 [ 0]
48 [ 1] *
56 [ 0]
64 [ 0]
80 [ 0]
96 [ 18] *
112 [ 65] **
128 [ 163] ****
160 [1315] %****************************
192 [3093] %%******************************************************************
224 [2177] %%%*********************************************
256 [ 771] %%***************
320 [ 219] %%***
-------------------------------------------------------------------------------
kbps LR MS % long switch short %
202.6 4.6 95.4 95.8 2.5 1.7
Writing LAME Tag...done
ReplayGain: -10.3dB
C:\Program Files\LAME>
© Copyright 1985-2001 Microsoft Corp.
C:\>cd c:\Downloads\lame3.98
C:\Downloads\lame3.98>lame.exe -V 2 C:\Temp\Linchpin.wav "C:\Temp\Linchpin 3.98.
mp3"
LAME 3.98 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:\Temp\Linchpin.wav to C:\Temp\Linchpin 3.98.mp3
Encoding as 44.1 kHz j-stereo MPEG-1 Layer III VBR(q=2)
Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
7850/7850 (100%)| 0:06/ 0:06| 0:06/ 0:06| 29.693x| 0:00
32 [ 28] *
40 [ 0]
48 [ 0]
56 [ 0]
64 [ 0]
80 [ 0]
96 [ 2] *
112 [ 25] *
128 [ 89] %**
160 [ 938] %*********************
192 [3014] %%%*****************************************************************
224 [2513] %%%%*****************************************************
256 [ 822] %%%%***************
320 [ 419] %%%*******
-------------------------------------------------------------------------------
kbps LR MS % long switch short %
210.4 7.3 92.7 92.7 4.1 3.2
Writing LAME Tag...done
ReplayGain: -10.3dB
C:\Downloads\lame3.98>cd "c:\Program Files\LAME"
C:\Program Files\LAME>lame.exe -V 2 --vbr-new C:\Temp\Linchpin.wav "C:\Temp\Linc
hpin 3.97.mp3"
LAME 3.97 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:\Temp\Linchpin.wav to C:\Temp\Linchpin 3.97.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
7850/7850 (100%)| 0:08/ 0:08| 0:08/ 0:08| 24.901x| 0:00
32 [ 28] *
40 [ 0]
48 [ 1] *
56 [ 0]
64 [ 0]
80 [ 0]
96 [ 18] *
112 [ 65] **
128 [ 163] ****
160 [1315] %****************************
192 [3093] %%******************************************************************
224 [2177] %%%*********************************************
256 [ 771] %%***************
320 [ 219] %%***
-------------------------------------------------------------------------------
kbps LR MS % long switch short %
202.6 4.6 95.4 95.8 2.5 1.7
Writing LAME Tag...done
ReplayGain: -10.3dB
C:\Program Files\LAME>
Congratulations on the release!
Please do continue the test preparations. We have done so much good preliminary work that it would be a pity if the actual test would not take place.
I have not been very active at HA during the recent months (been very busy with my work and real-life), but I am willing to sacrify some of my time for this good cause.
I posted a new reply in the test thread: http://www.hydrogenaudio.org/forums/index....showtopic=57322
QUOTE (Sebastian Mares @ Jun 29 2008, 11:27)
I wanted to carry out the MP3 listening test once 3.98 is available as final version. Maybe we could start discussing samples now.
Please do continue the test preparations. We have done so much good preliminary work that it would be a pity if the actual test would not take place.
I have not been very active at HA during the recent months (been very busy with my work and real-life), but I am willing to sacrify some of my time for this good cause.
I posted a new reply in the test thread: http://www.hydrogenaudio.org/forums/index....showtopic=57322
awesome~! john33, can you also upload debian package?
QUOTE (Nick.C @ Jul 4 2008, 09:09)
Well done that man! Congratulations on the new release!!
Seconded! Great work John.
QUOTE (ZinCh @ Jul 4 2008, 11:35)
awesome~! john33, can you also upload debian package?
I'd love to help but I'm afraid I have the skills of an aardvark when it comes to that!!
I'm sure xmixahlx will take care of that before long.
Thanks for the build !
Fantastic!
My sincere Thankyous for all the developers, contributors and john33.
My sincere Thankyous for all the developers, contributors and john33.
Thanks for all your dedication and hard work!
Thank you!!!
Special treat; especially on the 4th!
Thanks to everyone involved. A lot of work went into this release.
What they all said. More thanks for the altruistic dedication.
Woohoo!
3.98 has been coming long time and it's finally here. Thank-you for all developers for your dedicated and hard work.
3.98 has been coming long time and it's finally here. Thank-you for all developers for your dedicated and hard work.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.