Feature Requests
Re: Feature Requests
OK, I tried to duplicate this but failed. I tried both cropping in DGI and cropping using the DGDecodeNV parameters. So it seems I do not understand what you are doing. Please provide a step-by-step process to see your issue. Thank you.
Re: Feature Requests
Code: Select all
DGsource ("movie.mkv", debug=true, crop_b=148, crop_t=140) #crop is mod4
Code: Select all
DGsource ("movie.mkv", debug=true) #do not crop here since we're not doing mod4
ConvertToYV24 (matrix="Rec709") #to make crops that are mod1
## crop examples
# for wallpapers and image_processing, not critical video_processing but where all picture data can be used
#Crop (0, 139, 0, -138) # top crop is mod1 # resulting height=mod1
#for movies that are 1.85:1
#Crop (0, 21, 0, -21) # top+bottom crop is mod1 # resulting height=mod2
Just for exapmle, clip with black bars to keep debug on clean area:
DGSource("F:\Clips\Inception\Inception CH1.dgi", debug=true)
now same screenshot with mod1 crops to achieve 1.85:1 resolution (for sake of test):
DGSource("F:\Clips\Inception\Inception CH1.dgi", debug=true)
ConvertToYV24 (matrix="Rec709")
Crop(0, 21, -0, -21)
The bigger the Crop() values, the more the debug info gets cropped.
Re: Feature Requests
You're cropping after DGSource() in your script and then expecting the debug info to be repositioned or something? Tell me I have missed something, because there is no way for DGSource() to know you have a crop later in the script.
Best thing I can think of is to comment out your crop line when you need to see the debug ouput. Do you have another idea?
Best thing I can think of is to comment out your crop line when you need to see the debug ouput. Do you have another idea?
Re: Feature Requests
Here's what I'm thinking of: Add an INI file option called Force_Zoom. It can have values 0 (default), 1, 2, or 4. If it is zero the current behavior is used, i.e., after a file is loaded a reasonable zoom factor is automatically chosen. Otherwise the specified value is forced, i.e., Force_Zoom = 2 forces zoom to be 1/2. This allows me to have my favorite auto zoom mode but allows you to force a zoom if you want to. In no case would the GUI selected Zoom be saved in the INI file; it would always be controlled by Force_Zoom. My reasoning for the auto mode is that if I load a UHD video on a 1920x1080 display I don't want to have to go into Video/Zoom to make it all visible.Groucho2004 wrote:Don, any chance you could make DGIndexNV remember the last zoom factor, i.e. add an INI option?
Unless you refute that idea I'll go with it.
Re: Feature Requests
I was hoping maybe for something like a separate call for the info at any time later.admin wrote:Do you have another idea?
Similar to Info() and FFInfo() for the frame information.
Re: Feature Requests
Hmm, thinking... I understand what you are asking for, and it's not unreasonable, the problem is to satisfy it. Let me sleep on it.sparktank wrote:I was hoping maybe for something like a separate call for the info at any time later.admin wrote:Do you have another idea?
Similar to Info() and FFInfo() for the frame information.
Re: Feature Requests
Keep rested. Ideas usually come best from a dream sequence. If you lose sleep on this, put it on the back burner until something happens in a dream. Must go deeper~admin wrote:Let me sleep on it.
Re: Feature Requests
It is deeply gratifying to contact if even so briefly such profound souls as yourself.sparktank wrote:Must go deeper
Re: Feature Requests
Please get a look http://forum.doom9.org/showthread.php?p ... ost1774770.
AVS+ is having a major upgrade and plugins can declare their MT preferred mode + other nice things.
What do you thing about recompiling DGNV and DGIM to take advantage of AVS+?
AVS+ is having a major upgrade and plugins can declare their MT preferred mode + other nice things.
What do you thing about recompiling DGNV and DGIM to take advantage of AVS+?
Re: Feature Requests
Can you be more specific about I should do and what we would gain from doing so? Thanks.
ultim wrote:
- Avs+ will now automatically recognize source filters. If it sees a source filter which has no MT-mode specified at all, it will automatically use mode 3 instead of the default MT mode. This is great for source filters as in their case it rarely hurts performance and they often need mode 3 anyway.
So it seems there's nothing I need to do.
ultim wrote:
- Avs+ will now automatically recognize source filters. If it sees a source filter which has no MT-mode specified at all, it will automatically use mode 3 instead of the default MT mode. This is great for source filters as in their case it rarely hurts performance and they often need mode 3 anyway.
So it seems there's nothing I need to do.
Re: Feature Requests
admin wrote:Can you be more specific about I should do and what we would gain from doing so? Thanks.
Does DGNV dll already declare its MT-mode?plugins should now be able to reliably specify their MT-mode using the SetCacheHints() mechanism (for an example, see the usage of CACHE_GET_MTMODE here)
Re: Feature Requests
No, so avs+ will automatically use mode 3.
Re: Feature Request for x264 lossless
DGDecNV currently fails to decode x264 lossless AVC format profile "High 4:4:4 Predictive Level 3".
Would it be possible to support this format in future?
Would it be possible to support this format in future?
- Aleron Ives
- Posts: 126
- Joined: Fri May 31, 2013 8:36 pm
Re: Feature Requests
I could be wrong, but IIRC nVidia doesn't support it, because no user-facing content is ever lossless, so it's not something DG can fix without nVidia's help.
Re: Feature Requests
Well, if so it's a pity. I wanted to capture video tapes to x264 lossless for later offline conversion to lossy and filtered AVC. DGdecNV would have been convenient for quick inspection/analysis of the captured files.
Anyway, I can view these files with MPC-HC / LAV instead, or capture to quasi-transparent (say CRF=10) lossy format with a very fast x264 settings and use DGDecNV on these.
Anyway, I can view these files with MPC-HC / LAV instead, or capture to quasi-transparent (say CRF=10) lossy format with a very fast x264 settings and use DGDecNV on these.
Re: Feature Requests
Aleron is right, AFAIK. Lossless is not supported by CUVID.
LAV doesn't use CUVID for lossless, yes?
LAV doesn't use CUVID for lossless, yes?
Re: Feature Requests
LAV doesn't use CUVID, as far as I understand.
Re: Feature Requests
Think I responded to you in the MKV demux thread just as you locked it
Re: Feature Requests
No problem. Just make initial suggestions here and then I will split if it leads to development. Happy Sunday night.
The thread is open now at the link I gave, so it's unlocked and visible, with the archived material restored. Information escaped from a black hole. What are we to make of that?
The thread is open now at the link I gave, so it's unlocked and visible, with the archived material restored. Information escaped from a black hole. What are we to make of that?
Re: Feature Requests
Quantum Mechanics at work?Information escaped from a black hole. What are we to make of that?
My feature request, when you finish the present project AND if its not to much effort, would be the ability to select which streams to carry over on a "Output Trimmed TS" project since m2ts loses all language identifiers
A good night to you as well
[admin note -- following gonca post was moved here after a thread split]
When using "Output Trimmed TS" it might be useful to be able to select which tracks to use
The mpls has language indication but m2ts does not, so it would be useful, to me at least, to be able to select which audio and subtitle tracks to carry
Re: Feature Requests
I was thinking in DGIndexNV when creating a trimmed TS
Thanks for considering it
Thanks for considering it
Re: Feature Requests
The TS trim support is literally just a DGSplit that forces to cut on 0x47 transport syncs starting at the beginning of the project range. It would be a big project to make a stream editor. I could display or dump info from the MPLS that might give you the information you need. Maybe if you described your work flow I could be more helpful.
Re: Feature Requests
I open the mpls for the main movie in DGIndexNV
Set the project range and Output Trimmed TS (m2ts extension)
I only need the main English audio track and English subtitles, but since m2ts has no language identifiers I write which are which
The idea is that if the trimmed TS would have only these tracks then it would simplify things downstream
If it is a lot of effort don't worry about it because it only means I have to write down a few track numbers
Set the project range and Output Trimmed TS (m2ts extension)
I only need the main English audio track and English subtitles, but since m2ts has no language identifiers I write which are which
The idea is that if the trimmed TS would have only these tracks then it would simplify things downstream
If it is a lot of effort don't worry about it because it only means I have to write down a few track numbers