From: Philip Hood (email suppressed)
Date: Wed Apr 12 2006 - 10:31:39 PDT
avisynth gives me a substantial control over digital
video data. I have been working with it for a bit,
but by no means claim any specific expertise or
erudition, or even understanding, but I am _always_
willing to share what I know and work with others.
should someone have something more to share here, let
them please do.
now, having said that:
in most cases, each frame of video is
actually 2 "complete" image frames, which we can
call fields a & b. this is not true for all video,
but most. these two fields are then
interlaced together. using avisynth, it is
possible to take a video stream and break it up into
its constituent fields.
so, for instance, if you have 30 frames of ntsc (720,480)
video, and you break it up into its costiuent fields,
you will then have 60 frames of (720,240) video. the
even and odd fields (frame 1, 2) & etc are generally
woven together to create a total image of 720x480 pixels
and then rendered to the screen. As to more technically
why this is, I have read a handful of radio and television
engineering books, but I fail to totally understand.
In the example I gave of sarah krouse's work, I took to
separate clips, we will call them clip 1 and clip 2,
and separated the fields into a & b, so we 1.a & 1.b and
2.a & 2.b. I then wove together 1.a with 2.b, and
this is all quite simple using avisynth, a tool that is
free and has a number of other very helpful filters.
the catch is that its interface is quite like a
programming language, and not point and click, which may
or may not hinder some more used to an interface like
final cut or somesuch. I find this manner far more
flexible and easier to use with a more fine grained
level of control. This tool is free and easily found
on the internet.
but there are other things I'd like to do. I'd personally
love to spend some time building a tool to create a profile
of an image, with brightness and r g b characteristics
mapped somehow, and then mix and match these images with
a representation of sound and tone color, producing a
similiar effect as we saw in a post yesterday, but one that
would be, in my estimation more geared toward the visual
aspect, and possibly produce far more "scrupulous" imaging.
finally, I am always happy to work with _anyone_ on just
about _anything_. I am in New York and always looking to
sharpen my skills. Michael, if you send me your clips,
I'll be happy to show you exactly what I mean here ... and
welcome bouncing other ideas off of you towards the future.
On Wed, 12 Apr 2006, Michael Betancourt wrote:
> This is quite interesting--its an effect that I''ve been wanting for
> some of my work too.
> Can you be more specific about how this interlacing was created? This
> kind of thing could be done with analog equipment (the Experimental TV
> center can do this) but it's not an option with DV generally. (As the
> DV tool get more "PRO" they become less useful to experimentation in
> my opinion.)
> If you're working on a tool pack or somesuch for this kind of effect,
> I'm interested!
> On 4/8/06, Philip Hood <email suppressed> wrote:
>> hi, I did make a small interlacing of two clips example
>> for sarah kanouse aka http://www.readysubjects.org
>> and the more I look at it, the more I like the
>> idea, and wanted to share to get some feedback of
>> what some others think, especially since there seemed
>> to be some misunderstanding in terms of exactly what
>> she was looking for. Its at:
>> I used avisynth to make it. Is anyone else familiar
>> with this tool ? I've been thinking of all
>> the various things that could be expressed w/ this
>> in a palette. Does anyone know of some lengthier
>> and more woven examples of this idea ?
>> I hope sarah doesn't mind sharing this small clip
>> to spark discussion. The source was files she's
> Michael Betancourt
> Des Moines, IA USA
> the avant-garde film & video blog
> For info on FrameWorks, contact Pip Chodorov at <email suppressed>.
For info on FrameWorks, contact Pip Chodorov at <email suppressed>.