DGDemux development
Re: DGDemux development
Oh, such a whirlwind. First Boltzmann told me it's all about distributions and then Newton told me it's all about initial conditions. Who to believe?
Anyway, I re-based my speedup code on slipstream 43, which seems pretty stable. I still reduce demux time for the video from 3:26 to 0:58 for a representative disk. I would like to put the code back under an INI option so that it doesn't diverge from the main line again. Is that all right, Rocky?
Anyway, I re-based my speedup code on slipstream 43, which seems pretty stable. I still reduce demux time for the video from 3:26 to 0:58 for a representative disk. I would like to put the code back under an INI option so that it doesn't diverge from the main line again. Is that all right, Rocky?
Sherman Peabody
Director of Linux Development
Director of Linux Development
- Bullwinkle
- Posts: 338
- Joined: Thu Sep 05, 2019 6:37 pm
Re: DGDemux development
Distributions and initial conditions are the same thing. There is no abstract distribution divorced from a specific sampling of reality, and such a sampling can itself be taken as a set of initial conditions given at a specific time.
Go ahead and integrate your stuff (use Commit on the VisualSVN pulldown). When you have everything supported, we can publish the INI setting so that our great users can test it out. 3:26 -> 0:58 sounds great. Git 'er done!
Go ahead and integrate your stuff (use Commit on the VisualSVN pulldown). When you have everything supported, we can publish the INI setting so that our great users can test it out. 3:26 -> 0:58 sounds great. Git 'er done!
Re: DGDemux development
What Bullwinkle said.
Re: DGDemux development
Let me guess:
Speedup=0/1
Speedup=0/1
- Mr. Peabody
- Posts: 45
- Joined: Tue Dec 24, 2019 9:20 am
Re: DGDemux development
New_Single_Level_Parsing_Architecture=false/true
Curly Howard
Director of EAC3TO Development
Director of EAC3TO Development
Re: DGDemux development
Brain_Dead_Design_From_The_Past_Finally_Excised=duh/huh
Re: DGDemux development
Silly billies, only DGDemuxGUI reads the INI file, so there has to be an option to DGDemux: maybe -fast.
And then in DGDemuxGUI, if you hold down the shift key when pressing the Show or Demux buttons, the -fast option is issued. That way we don't have to create a new option line in the GUI (which would just have to be removed down the road) while still being able to go fast or slow without needing to edit the INI file or exit the GUI.
And then in DGDemuxGUI, if you hold down the shift key when pressing the Show or Demux buttons, the -fast option is issued. That way we don't have to create a new option line in the GUI (which would just have to be removed down the road) while still being able to go fast or slow without needing to edit the INI file or exit the GUI.
Re: DGDemux development
Great idea, Britney. I implemented that and hopefully Rocky will include it in the next slipstream. I'll keep adding to it until all the stream types are supported in fast mode. Then when we switch everything over we will have fast mode by default.
BTW, I demuxed the AC3 stream from 1917:
old mode: 3:46 secs
fast mode: 0:32 secs
Seems like a useful development. :belly-laugh:
BTW, I found a little bug in build 43. DV streams are not marked as DV (regression from the fix to not show PIP as DV). Say thank you.
BTW, I demuxed the AC3 stream from 1917:
old mode: 3:46 secs
fast mode: 0:32 secs
Seems like a useful development. :belly-laugh:
BTW, I found a little bug in build 43. DV streams are not marked as DV (regression from the fix to not show PIP as DV). Say thank you.
Sherman Peabody
Director of Linux Development
Director of Linux Development
Re: DGDemux development
Talk about exposing people's early careers. Still, not quite in your league:
https://youtu.be/76pYZjhVylw?t=40
Re: DGDemux development
Let's expose Natasha. Sunlight kills vampires. Nyuh nyuh.
Curly Howard
Director of EAC3TO Development
Director of EAC3TO Development
Re: DGDemux development
Spin in circles, doughboy.
Re: DGDemux development
Settle down children! Love and respect one another.
Sherman, thank you for checking in your code and for reporting the DV marking bug. I noticed that gaps correction is not working for DTS in fast mode (AC3 is OK). Can you look into it please? Great speedup BTW.
Sherman, thank you for checking in your code and for reporting the DV marking bug. I noticed that gaps correction is not working for DTS in fast mode (AC3 is OK). Can you look into it please? Great speedup BTW.
- Bullwinkle
- Posts: 338
- Joined: Thu Sep 05, 2019 6:37 pm
Re: DGDemux development
Thank you, Sherman, for the DTS fix for fast mode. Now I see that LPCM is demuxing differently between fast and slow modes. Please check it out. Looking good!
Re: DGDemux development
I LOVE the preview !
Many thanks !
Many thanks !
Re: DGDemux development
Sorry I took so long for the LPCM fix. It's all better. Bullwinkle had me laying floors for a week!
Typical disk demuxing the LPCM track:
slow mode 3:45
fast mode 0:35
Anyway, I just have to add the following and then you can let it out to the wild:
* THD and embedded AC3
* PGS
* preallocation
* garbage stripping option
Shouldn't take too long. However:
But Mousie, thou art no thy-lane,
In proving foresight may be vain:
The best laid schemes o’ Mice an’ Men
Gang aft agley,
An’ lea’e us nought but grief an’ pain,
For promis’d joy!
-- Robert Burns
Typical disk demuxing the LPCM track:
slow mode 3:45
fast mode 0:35
Anyway, I just have to add the following and then you can let it out to the wild:
* THD and embedded AC3
* PGS
* preallocation
* garbage stripping option
Shouldn't take too long. However:
But Mousie, thou art no thy-lane,
In proving foresight may be vain:
The best laid schemes o’ Mice an’ Men
Gang aft agley,
An’ lea’e us nought but grief an’ pain,
For promis’d joy!
-- Robert Burns
Sherman Peabody
Director of Linux Development
Director of Linux Development
Re: DGDemux development
Sounds great, Sherman. I confirmed that the LPCM is fixed in fast mode.
As they say in the swimming world: Finish Strong!
As they say in the swimming world: Finish Strong!
Re: DGDemux development
I don't see any problem. Would you be able to explain why you think there is a problem?Bullwinkle wrote: ↑Thu Sep 24, 2020 7:55 pmSherman, did you solve the quantum measurement problem yet?
@Rocky
I can't swim.
Sherman Peabody
Director of Linux Development
Director of Linux Development
- Mr. Peabody
- Posts: 45
- Joined: Tue Dec 24, 2019 9:20 am
Re: DGDemux development
Propping up straw men is a not-so-subtle form of rhetoric to be eschewed. I refer of course to the quantum measurement straw man.
Re: DGDemux development
Does that mean hard to swallow?
@Rocky
Got PGS demuxing working in fast mode. The fast and slow demuxed SUP files are binary identical. Now have to do THD and embedded AC3. Then a little cleanup and we can give it out.
There is enough supported to fully demux La Notte (Criterion). Here are the timings for demuxing everything:
slow: 5:00
fast: 1:19
That's a 3.8 times speedup!
Sherman Peabody
Director of Linux Development
Director of Linux Development
Re: DGDemux development
Give the kid a break. He's doing his best.
Curly Howard
Director of EAC3TO Development
Director of EAC3TO Development