| |
| |
DVD |
|
| |
Animation |
|
| |
Output
to tape |
|
| |
Output
to film |
|
| |
Exhibitions |
|
| |
Special deal for Television Producers
Traditionally you have had to go
to four different companies for design in a production, the
logo design, print (stationary / letterhead), opening title
sequences,
and web pages. Each has totally different demands and expectations.
HONEYcom designs once and can produce solutions for all the above
- you pay for the design only once!!
|
|
 |
|
|
Output to Film
Outputting video to film can be extremely complex. Here are the
hurdles:
1. Film is 24 frames per second and video is 25fps
2. Film, in video terms, is progressive, while video is interlaced.
Interlaced Video?
When television was being invented, the engineers found you couldn't output an entire frame on a CRT (Cathode Ray Tube) without getting flicker. What was happening was that by the time the scanning of the CRT reached the bottom of the screen; the beginning, the top, had now faded to black. To resolve this, they decided to simulate how a film projector actually outputs film - a projector plays the film at 24fps but has a two blade shutter that shows the frame twice which means each frame is actually shown twice - 48fps.
In television, the frame divided into two by taking all the odd horizontal lines
and calling it the "odd field" and outputting that field (lines 1-625)
then going back and filling in the even lines - the "even field" (lines
2-624). When a television set outputs, it "interlaces" the two fields
- the first odd lines and then in the gaps between it fills the even lines and
this is where the term "interlaced video" comes from.
Unfortunately, the computer world has managed to rename the fields as upper and
lower fields, odd = upper and lower = even so as to cause maximum confusion.
Now, when you transfer film to video via telecine, it creates the odd and even
fields
by scanning a film frame twice. But adding them together as a video freeze
frame, you see the entire "progressive" frame.
However, a video sees things different. By the time you scan the first field,
the world has moved on a 50th of second and the second field can be completely
different to the first.
And this is not a bad thing - It means that video is 50-half resolution-frames
a second and as any animator will tell you, the more frames you can output per
second, the more "natural" the motion. |
Back to outputting video to film. If the
picture was originally shot on film, telecined and compiled, then
it is a (almost) straight
forward process.
But video shot on Beta SP or DVCAM will be interlaced. If you output
a interlaced still frame to a television screen, you get a wobbly
picture. When you output the same to film you get ugly unnatural
edge artifacts.
So, the quickest thing to do is to throw away a field. This will
give the most natural motion but the picture is now only half the
resolution and when projected on the big screen, the lower resolution
becomes very noticeable.
HONEYcom has developed a series of procedures to perform intelligent
adaptive de interlacing.
Some of the arsenal of tools we have developed are as follows:
1. We use BlackMagic Decklink SDI cards to get pictures in and out of our computers. Decklinks also lets us preview how de-interlacing will look.
2. We create masks in Commotion or Photoshop and process parts
of the picture differently.
3. We use some "in-house" developed convolution filters.
4. We have a range of filters such as CineMotion and CineLook.
|
