Skip to main content
Topic: Lame 3.98 beta 6 (Read 138575 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Lame 3.98 beta 6

Reply #100
Can we get 3.98 final as a Christmas gift?

It's very unlikely that it will be released this year, because there are some issues with Debian and Cygwin/MSYS, which our confi-GURU wants to solve first.

Lame 3.98 beta 6

Reply #101

Can we get 3.98 final as a Christmas gift?

It's very unlikely that it will be released this year, because there are some issues with Debian and Cygwin/MSYS, which our confi-GURU wants to solve first.

That's good news as I'm away from Friday, 21 Dec until 2 Jan with only very limited Internet access.  Happy hols everyone!!
John
----------------------------------------------------------------
My compiles and utilities are at http://www.rarewares.org/

Lame 3.98 beta 6

Reply #102
Quote
It's very unlikely that it will be released this year, because there are some issues with Debian and Cygwin/MSYS, which our confi-GURU wants to solve first.


so AFTER that, it will be released ?!?! ;-)

Lame 3.98 beta 6

Reply #103
so AFTER that, it will be released ?!?! ;-)

There is a lot of time left AFTER one particular point in time 

... So dear LAME devs have a nice Christmas and a happy New Year! And thanks a real lot for your hard and excellent work

Lame 3.98 beta 6

Reply #104
I get the impression that only LAME project is the most active against other codecs, followed by NERO Digital. Vorbis seems so abandoned, and not mentinoning the slug MPC...

Would it be the case that LAME heads on top against its competitors at some point?

Lame 3.98 beta 6

Reply #105
Would it be the case that LAME heads on top against its competitors at some point?

thanks to the lame devs' hard work, it could very well happen. at 128 kbps, lame moved from last place here to  tied at first place here. lame also tied with former champ mpc here.


Lame 3.98 beta 6

Reply #107
I guess active members on HA also expect activities on the most promising codecs, and if there are none many members consider a codec to be dead. Look at the discussion about MPC before the current MPC activities.
But this has nothing to do with the quality of a codec. Quality simply may be fine without real need for improvement (thinking about Vorbis for instance).

Anyway the great quality improvements of Lame are most welcome as mp3 is the most universally usable standard of audio compression.
Thanks a lot to the Lame devs who have made Lame mp3 competitive with newer codecs.
lame3995o -Q1.7
opus --bitrate 140

Lame 3.98 beta 6

Reply #108
I just tried tried the official 3.98b6 at V0, and though my last test was with a version pretty close to this one eig has improved one more time. The 2 prominent impulses that have been a problem for so long were a pretty negligible problem with my last test, but now there is no problem at all with them for practical listening situations. I had a hard time to abx them - I had to listen over and over again before daring to abx.
Trained this way I was also able to abx harp40_1, but from previous tests I have the experience that with mp3 it only takes sufficient time to be able to abx harp40_1 with any mp3 encoder, even at very high bitrate. Again this is academic, for practical listening situations it's no problem at all.
I was also able to abx the tremolo problem of the artificial lead-voice sample. I could not do so with my last test, so I guess this is also due to my current abx training with eig. lead-voice isn't important however for practical purposes.
My other tremolo problem sample trumpet_myyPrince was fine, and this sample has a practical impact.
My other problem samples (Birds, trumpet, herding_calls) were all not abxable to me.

We'll get a great 3.98 version.
lame3995o -Q1.7
opus --bitrate 140

Lame 3.98 beta 6

Reply #109
Any news about the B6 eating up the cpu bug? Would be nice if that was solved then i could use it.

Lame 3.98 beta 6

Reply #110
I installed B5 and now the cpu is at max 2..4% again, so it seems to be a B6 specific problem.

Are there things so good in B6 that i better stay with B6 or is it ok to go back to B5? I can use the extra freed cpu time. Dont want to cripple an mp3 when recording in B6 and opening firefix.

That doesn't make much sense, B5 and B6 should behave the same. Are you using lame.exe or lame_enc.dll?

Lame 3.98 beta 6

Reply #111

I installed B5 and now the cpu is at max 2..4% again, so it seems to be a B6 specific problem.

Are there things so good in B6 that i better stay with B6 or is it ok to go back to B5? I can use the extra freed cpu time. Dont want to cripple an mp3 when recording in B6 and opening firefix.

That doesn't make much sense, B5 and B6 should behave the same. Are you using lame.exe or lame_enc.dll?

The dll.

More info:
https://sourceforge.net/tracker/?func=detai...mp;group_id=290

Lame 3.98 beta 6

Reply #112
The lame_enc.dll frontend code hasn't changed from b5 to b6. The lame_enc.dll doesn't do any file IO, it's the calling application -- in your case totalrecorder -- that does all reading and writing from and to the files. The application calls lame_enc in a loop and passes the data in memory. So, it seems to be more of a compiler / compiling issue, than a change in LAME itself.

Lame 3.98 beta 6

Reply #113
The lame_enc.dll frontend code hasn't changed from b5 to b6. The lame_enc.dll doesn't do any file IO, it's the calling application -- in your case totalrecorder -- that does all reading and writing from and to the files. The application calls lame_enc in a loop and passes the data in memory. So, it seems to be more of a compiler / compiling issue, than a change in LAME itself.

I just did a check with the official B6 (forgot wich site but it wasn't rareware this time) to see what happened. And gues what the problem is now gone i have a normal cpu % again.

So it could be indeed the compiler that rareware used.
What should i do with the bugreport?

Lame 3.98 beta 6

Reply #114
I just did a check with the official B6 (forgot wich site but it wasn't rareware this time) to see what happened. And gues what the problem is now gone i have a normal cpu % again.
Good to know, where did you get it from?

Quote
What should i do with the bugreport?

Nothing, I'll take care of it.

Lame 3.98 beta 6

Reply #115

I just did a check with the official B6 (forgot wich site but it wasn't rareware this time) to see what happened. And gues what the problem is now gone i have a normal cpu % again.
Good to know, where did you get it from?

Quote
What should i do with the bugreport?

Nothing, I'll take care of it.

Stupid me, i always thought i got my non official beta's 6 builds from rarewares, but in fact it was from http://lame.bakerweb.biz/
I gues it read so much about rarewares that somehow it was printed in my head, sorry about that.

The official and working B6 however was dloaded from rarewares, sorry for the misunderstanding.

Lame 3.98 beta 6

Reply #116
Sorry if this was asked in an earlier post, but around what time will the final version of 3.98 be coming out?

Lame 3.98 beta 6

Reply #117
Sorry if this was asked in an earlier post, but around what time will the final version of 3.98 be coming out?

robert replied here:

Can we get 3.98 final as a Christmas gift?

It's very unlikely that it will be released this year, because there are some issues with Debian and Cygwin/MSYS, which our confi-GURU wants to solve first.

Sorry if this has been discussed before, however from the changelog (here) it states:
Quote
Feature request [ 1811483 ] WAVE_FORMAT_EXTENSIBLE support (PCM)

Does this potentially mean multichannel MP3, or that LAME will be able to read/select certain channel from a WAVE_FORMAT_EXTENSIBLE *.wav file (and prodece a mono/stereo track from those)?

Cheers, Arite.

Lame 3.98 beta 6

Reply #118
I get the impression that only LAME project is the most active against other codecs, followed by NERO Digital.

But there's a big gap between them. Nero digital devs keep on telling there's problems regarding putting up stuff onto the web page, but however. I lost trust in the codec because of some problems with classical, they soon told the source of the problem has been found, but this is some ages ago.

Lame 3.98 beta 6

Reply #119
Quote
Feature request [ 1811483 ] WAVE_FORMAT_EXTENSIBLE support (PCM)
Does this potentially mean multichannel MP3, or that LAME will be able to read/select certain channel from a WAVE_FORMAT_EXTENSIBLE *.wav file (and prodece a mono/stereo track from those)?
No.

Lame 3.98 beta 6

Reply #120
Are there still planned features and bug fixes which affect quality before 3.98 goes stable?

Lame 3.98 beta 6

Reply #121
Quote
Feature request [ 1811483 ] WAVE_FORMAT_EXTENSIBLE support (PCM)
Does this potentially mean multichannel MP3, or that LAME will be able to read/select certain channel from a WAVE_FORMAT_EXTENSIBLE *.wav file (and prodece a mono/stereo track from those)?
No.

OK, thanks for the reply - what does it mean then?

Cheers, Arite.

Lame 3.98 beta 6

Reply #122
You can have normal mono/stereo wave files with WAVE_FORMAT_EXTENSIBLE format definition. Previously LAME did reject them or interpreted them as raw PCM files. Now LAME recognizes and may read them.

Lame 3.98 beta 6

Reply #123
Quote
You're wrong here. Vorbis may be abandoned, but Musepack isn't.


Musepack wears out with the long long lack of constant releases. But it's not just it, we're talking about here audio quality. Guruboolez once stated that he was so frustrated with MPC people that wouldn't even acknowledge a plain clear issue concerning audio quality, and that still was not looked into until today. And more... projects with no-dateline are pretty depressing, put everybody on hold and pain. It's easier to look forward to something when you see that there will be new releases in a time, that everything is moving on. But when a developer says "it's out when it's out" or "I just don't know when it's out... perhaps some day in life" is just so depressing... MPC have sinned in many ways like this.

Lame 3.98 beta 6

Reply #124
I get the impression that only LAME project is the most active against other codecs, followed by NERO Digital. Vorbis seems so abandoned


Wrong about Vorbis. It is absolutely necessary to keep Vorbis I as stable and the codebase as constant as possible. This has to do with the lack of a formal "external" specification for Vorbis similar to the MPEG-specs so that it is necessary to keep everything as frozen as possible to allow hardware-vendors to trust Vorbis and to step in and to support Vorbis I on their devices (MP3-players). The exception is Aoyumi as he really does a fantastic job in optimizing the sound quality - actually without changing the already existing codebase (apart from adding a few possibilities).

As far as I know, the work on Vorbis II has started but I do not know more about the reasons to start it...

 
SimplePortal 1.0.0 RC1 © 2008-2020