Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion.

Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules.

 

Go Back   Doom9's Forum > Capturing and Editing Video > VapourSynth

Reply
 
Thread Tools Search this Thread Display Modes
Old 29th July 2018, 14:14   #3201  |  Link
LigH
German doom9/Gleitz SuMo
 
LigH's Avatar
 
Join Date: Oct 2001
Location: Germany, rural Altmark
Posts: 5,546
There is indeed an ANI file format. But hardly compressed, only a stack of DIBs, I believe. It was mainly intended for animated mouse cursors.
__________________

New German Gleitz board
MediaFire: x264 | x265 | VPx | AOM | Xvid
LigH is offline   Reply With Quote
Old 30th July 2018, 21:33   #3202  |  Link
poisondeathray
Registered User
 
Join Date: Sep 2007
Posts: 3,501
Quote:
Originally Posted by HuBandiT View Post
concept #2: python dudes in reduced size in lower right corner of icon (as customary when an icon references a technology employed), with blue python dude's eye shining white (read: consider making "white" darker in other areas of the icon) as it is projecting (with some kind of visible light rays) an image onto one frame a film strip which occupies the main area of the icon; if enough space available, show three (or 0.5 + 1 + 0.5) frames of the film strip, top frame already has a fully formed image, middle frame (where blue dude is projecting with his eye) is forming, bottom frame blank (awaiting to be imaged)
Like austin powers / dr. evil - sharks with fricken laser beams attached to their heads!

"visible light rays" can be done as volumetric lights, but it's difficult to retain as an ico or small logo. A colored light example in the video below. And I don't think you'd be able to fit all that in a small ico.

Also why only blue python with fricken laser beams? It seems yellow python gets no love

But you can have different graphics for the logo (e.g. something that you put on a webpage, or use in a video), vs the ico . But I think they should share some themes and traits. eg. Python dudes with their official colors at least. I like them better than Lord Mulder's cute green guy


Quote:
Originally Posted by Sparktank View Post
If this was going to be a movie studio logo before the picture, I can see this turning around to project on the silverscreen: "aVAPOURSYNTHproject"
Maybe something like this, with the python symbol/logo as the "o" some people misspell "vapOursynth". Not suitable for an ico , but an example of how you might use a possible logo. Colored light projection/ volumetric rays
https://www.youtube.com/watch?v=GKELqibwR8A

I don't like the text. But you need something that is synonymous with "vapoursynth". I also apologize for the font, its myriad pro, but I don't know what the official "font" is. Python uses Flux Regular "core.text" uses Terminus but it's an ugly font.



Quote:
Originally Posted by ChaosKing View Post
idk why but I immediately thought of this: uuuh yeahhhh, nice infinite ass wipe xD
Maybe it's just my dirty mind
maybe it was that ffms2 logo's subliminal effect ? or something Freudian

gives new meaning to "clean python code"
https://s22.postimg.cc/4s2tzj6vl/clean_python_code.jpg
poisondeathray is offline   Reply With Quote
Old 5th August 2018, 13:29   #3203  |  Link
hajj_3
Registered User
 
Join Date: Mar 2004
Posts: 827
It would be nice if you could add support for Python 3.7
hajj_3 is offline   Reply With Quote
Old 5th August 2018, 23:17   #3204  |  Link
ObenS
Registered User
 
Join Date: Jul 2018
Location: Desert
Posts: 2
Quote:
Originally Posted by hajj_3 View Post
It would be nice if you could add support for Python 3.7
r45 will support Python 3.7
ObenS is offline   Reply With Quote
Old 10th August 2018, 12:35   #3205  |  Link
Wolfberry
Helenium(Easter)
 
Wolfberry's Avatar
 
Join Date: Aug 2017
Location: Hsinchu, Taiwan
Posts: 56
Documentation Update

Internal functions like Binarize() Deflate()/Inflate() Median() Minimum()/Maximum() work in float in R44, but the doc still says they can only work in interger, I think this will confuse some users.

Also, MakeDiff() and MergeDiff() behave differently in float, the output is identical to "x y -" and "x y +" instead of "x y - 0.5 +" and "x y + 0.5 -" that one will normally expect.

Is this the intended behavior? I had implemented some workaround in my script due to this issue.
__________________
ロイヤルネイビーユニコーン
Wolfberry is offline   Reply With Quote
Old 10th August 2018, 17:30   #3206  |  Link
Myrsloik
Professional Code Monkey
 
Myrsloik's Avatar
 
Join Date: Jun 2003
Location: Ikea Chair
Posts: 1,851
Makediff is kinda up for debate. The only reason it's offset in integer is to have zero in the middle. With float there are no concerns like that. Discuss!
__________________
VapourSynth - proving that scripting languages and video processing isn't dead yet
Myrsloik is online now   Reply With Quote
Old 19th August 2018, 19:43   #3207  |  Link
Myrsloik
Professional Code Monkey
 
Myrsloik's Avatar
 
Join Date: Jun 2003
Location: Ikea Chair
Posts: 1,851
Just a warning for all of you out there. The accept_lowercase option will be removed in the next version. None of you should be using it anyway and I really don't think it has a place in a case sensitive language like python.
__________________
VapourSynth - proving that scripting languages and video processing isn't dead yet
Myrsloik is online now   Reply With Quote
Old 24th August 2018, 11:52   #3208  |  Link
unix
Registered User
 
Join Date: Aug 2015
Posts: 25
Hi guys, I have question.

What does this command mean"planes=[0, 2, 1], planes=[1] or planes=[1,2,3] ??
and values only from 0 to 3 value ?

Thanx
unix is offline   Reply With Quote
Old 24th August 2018, 15:21   #3209  |  Link
poisondeathray
Registered User
 
Join Date: Sep 2007
Posts: 3,501
Quote:
Originally Posted by unix View Post

What does this command mean"planes=[0, 2, 1], planes=[1] or planes=[1,2,3] ??
and values only from 0 to 3 value ?

This usually is a reference to the channels being used , such as Y,U,V,A or R,G,B,A
poisondeathray is offline   Reply With Quote
Old 24th August 2018, 17:48   #3210  |  Link
DJATOM
Registered User
 
Join Date: Sep 2010
Location: Ukraine, Bohuslav
Posts: 119
As I know, alpha channel in VS stored as second video clip, so for RGB and YUV it's always [0,1,2] or so.
__________________
Me on GitHub | My Telegram
DJATOM is offline   Reply With Quote
Old 26th August 2018, 03:24   #3211  |  Link
edcrfv94
Registered User
 
Join Date: Apr 2015
Posts: 69
Any way change clip format chroma loction?
After nnedi3 Upscaling if use core.resize.Spline36 fix shiftm chroma loction will be set to center.
edcrfv94 is offline   Reply With Quote
Old 26th August 2018, 05:06   #3212  |  Link
WolframRhodium
Registered User
 
Join Date: Jan 2016
Posts: 73
Quote:
Originally Posted by edcrfv94 View Post
Any way change clip format chroma loction?
After nnedi3 Upscaling if use core.resize.Spline36 fix shiftm chroma loction will be set to center.
Use std.SetFrameProp:
Code:
clip = core.std.SetFrameProp(clip, prop="_ChromaLocation", intval=0)
list of frame properties
WolframRhodium is offline   Reply With Quote
Old 26th August 2018, 08:27   #3213  |  Link
edcrfv94
Registered User
 
Join Date: Apr 2015
Posts: 69
Quote:
Originally Posted by WolframRhodium View Post
Use std.SetFrameProp:
Code:
clip = core.std.SetFrameProp(clip, prop="_ChromaLocation", intval=0)
list of frame properties
Thanks

Anyway can get the chip info/props?
c_cl = clip.props._ChromaLocation
c_range = clip.props._ColorRange
Just idea not working
edcrfv94 is offline   Reply With Quote
Old 26th August 2018, 09:03   #3214  |  Link
WolframRhodium
Registered User
 
Join Date: Jan 2016
Posts: 73
Quote:
Originally Posted by edcrfv94 View Post
Thanks

Anyway can get the chip info/props?
c_cl = clip.props._ChromaLocation
c_range = clip.props._ColorRange
Just idea not working
Use text.ClipInfo if you only want to manually read the values. (text.FrameProps is more human-readable.)

Otherwise you might want to use std.FrameEval, std.ModifyFrame or simply
Code:
clip.get_frame(n).props['_ChromaLocation']
.
WolframRhodium is offline   Reply With Quote
Old 26th August 2018, 10:16   #3215  |  Link
edcrfv94
Registered User
 
Join Date: Apr 2015
Posts: 69
Quote:
Originally Posted by WolframRhodium View Post
Use text.ClipInfo if you only want to manually read the values. (text.FrameProps is more human-readable.)

Otherwise you might want to use std.FrameEval, std.ModifyFrame or simply
Code:
clip.get_frame(n).props['_ChromaLocation']
.
clip.get_frame(0).props['_ChromaLocation']
clip.get_frame(0).props._ChromaLocation
both doesn't work
edcrfv94 is offline   Reply With Quote
Old 26th August 2018, 12:05   #3216  |  Link
WolframRhodium
Registered User
 
Join Date: Jan 2016
Posts: 73
Quote:
Originally Posted by edcrfv94 View Post
clip.get_frame(0).props['_ChromaLocation']
clip.get_frame(0).props._ChromaLocation
both doesn't work
Why? Maybe the key is not defined?
WolframRhodium is offline   Reply With Quote
Old 26th August 2018, 21:41   #3217  |  Link
DJATOM
Registered User
 
Join Date: Sep 2010
Location: Ukraine, Bohuslav
Posts: 119
Quote:
for prop in clip.get_frame(0).props:
print(prop)
should print a list of valid props.
__________________
Me on GitHub | My Telegram
DJATOM is offline   Reply With Quote
Old 27th August 2018, 00:16   #3218  |  Link
edcrfv94
Registered User
 
Join Date: Apr 2015
Posts: 69
Quote:
Originally Posted by WolframRhodium View Post
Why? Maybe the key is not defined?
text.ClipInfo show Chroma Location: Unknown
But text.FrameProps or print show _ChromaLocation not defined.
edcrfv94 is offline   Reply With Quote
Old 3rd September 2018, 07:06   #3219  |  Link
Yrosma
Registered User
 
Join Date: Mar 2015
Posts: 2
Large page speed hit in ffmpeg

OK this is not a vapoursynth issue but as I read about large pages here maybe some people know.

I few pages back some people mentioned large pages and that it could give some better performance. As I'm doing some 4K material now which is quite slow I thought about trying this out as some people did get some improvement. My systems do have 32GB of memory so that should be fine.
But when I enabled large pages in windows and did some tests only using ffmpeg to convert to h264 and h265 (so no vapoursynth yet) I got a pretty big performance hit. Avarage around 20% slower conversions times just in ffmpeg conversions.

Did other people run into the same issues? Seen it on intel system with 32GB and an AMD system with 32GB

Because if it's like this even if there is an improvement in speed in vapoursynth, if ffmpeg conversion is this much slower there won't be a netto win.

Or could I be doing something wrong?
Yrosma is offline   Reply With Quote
Old 3rd September 2018, 08:23   #3220  |  Link
LigH
German doom9/Gleitz SuMo
 
LigH's Avatar
 
Join Date: Oct 2001
Location: Germany, rural Altmark
Posts: 5,546
According to StackOverflow, this is a quite complex topic, much related to the sizes of CPU internal caches (and thus the CPU model), and video encoding may not even deal in general with the kind of RAM access which would benefit from large pages.
__________________

New German Gleitz board
MediaFire: x264 | x265 | VPx | AOM | Xvid
LigH is offline   Reply With Quote
Reply

Tags
speed, vaporware, vapoursynth

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 18:45.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2018, vBulletin Solutions Inc.