Vector drawing into video - help please
 

Ok - I'm not sure of the technical stuff so I'll just describe what I'm trying to do here.

I'm producing a ski instructional DVD.
For some sections I need to be able to draw curves and then place stills of these into my video.

I have ZeusDraw which is a vector drawing program that allows me to draw bezier curves.

It only allows me to export in certain formats, and when I import those into Final Cut Express they seem fine in the viewer but a little "fuzzy" or "jagged" in the browser. I've been using JPEG format to export/import as both programs recognise those... tried TIFF also but no better...

Any ideas? Anyone done this stuff and can point me in the right direction it would be greatly appreciated.

Obviously you can only use a raster image in FCP- try saving an EPS as a high-res PSD and importing to FCP. You can do this in Illustrator or Photoshop.

I open vector images in Photoshop, because it will make you specify the size first. You want 72dpi, and just figure out the resolution you need. You can bring PSD files into FCP, which is good because they can maintain transparency.
As far as image quality there's a lot of problems with determining that in Final Cut. If you are seeing low quality in the canvas window, view it at 100%. If you notice that your image looks crisp at first, but when you render it looks fuzzy: remember that video just isn't perfect anyways. But the only real test is how it looks in the end. You have to see it on DVD before you decide how it will look on DVD.
There are a lot of things that can make images look bad on video, if it's made of thin lines, if it's red, if its a gradient etc... But be sure that it really looks bad first, and not simply video quality.

Meaning no disrespect but 72 dpi is a myth. Video is made of pixels. Use the correct number of pixels to create your raster image from the vector file to fill the screen. If you want to fill an NTSC DV screen with a vector image that is 4:3, use 720 x 480 pixels. Done.

DPI is specific to printing and has no place in video, except when scanning. Then use the appropriate dpi resolution to ensure you have enough pixels to work with.

You have to specify the resolution in Photoshop. Video resolution is 72, how is that a myth? There's no point in it being 300, the only thing that could possibly do is slow things down. In the past, high resolutions have created a lot of glitches for animations I've done in After Effects, so I always set my PSDs to 72 pixels/inch.

72 dpi myth
 

http://www.scantips.com/no72dpi.html
http://www.earthboundlight.com/photo...2dpi-myth.html
http://www.ekdahl.org/test_dpi/test2dpi.htm
http://www.maureenalbright.com/digit...h-72%20dpi.htm
http://www.woram.com/temp/woram.htm

Aric-

PPI (pixels per inch) is irrelevent to video because nothing is measured in inches in video. It is the number of pixels wide by the number of pixels high that matters.

In standard video, a frame is 720 x 480.
At 72 ppi this would be 10" wide x 6.667" high.
At 300 ppi it would be 2.4" wide x 1.6" high.

BOTH have exactly the same pixel dimensions, which is the only thing that matters in video. In fact the image is identical in every way except by a dimension (PPI/DPI)) that only matters for printing purposes.

Really about the only reason you'd need larger pixel dimensions would be if one were to want to do some pan & scan (the so-called Ken Burns effect, even though he in no way pioneered the technique).

After Effects, Premiere, Final Cut Pro, etc don't care about how many pixels per inch are used, they only care about actual pixel count. You can very easily make a wrong image at 72 ppi. You can just as easily make one that is correct. But at the end of the day, the PPI has NOTHING to do with whether it's right or wrong, it is only the actual pixel dimensions that matter.

Dammit, this is so confusing. I've only glimpsed at all those links so far, and what I am getting is: "Say no to 72dpi it is a myth"
I do various types of animation, sometimes drawing on the computer, sometimes printing video frames to rotoscope on paper and then scan. I have to specify the ppi in Painter and Photoshop, etc. Obviously I'd like to be able to print any illustration or video element I've created. How would it impact my work if I start making the resolution 7 as opposed to 72 or 300? I need to do more research on resolution, but until now I was sure that lower resolution would screw me, and higher would slow me down and create larger files. But it's only the pixel dimensions that slow programs down?

Aric,

Perhaps it will help if you think of it this way: The dpi setting is simply instructions to a printer telling it how large to print the image.

Open a document in Photoshop. Go to Image>Image Size. You will notice that if the "Resample Image" box is UNCHECKED, your pixel dimensions (and file size) are greyed out and unavailable. You will also notice that the width, height and resolution (or dpi) are all locked together. That means that as you increase the dpi, your printed size goes down. Conversely, if you decrease the dpi, the size of your printed image gets bigger.

Keep in mind that as you do this, your width and height in PIXELS remains the same. People get confused about this because Photoshop conveniently allows you to resample the image to another pixel dimension.

This is fine if you're downsampling a huge DSLR photo for the web, for example. Or downsampling the same photo for fullscreen video, which amounts to 720x480 or 640x480 (that's non-square vs. square pixels, but that's a whole other discussion).

Now, the 3 common misconceptions people have when using stills for video are the following:

1. People have heard that 720x480 is video resolution. That's fine if all you want to do is show the still image fullscreen. But, if you want to zoom in on the image at all, you need much higher PIXEL resolution. Say you want to zoom in to double the image size - 200% - you'd need 1440x960, which is double 720x480. This leads to:

2. You can't get pixel resolution back once you've downsampled. Say you created your 720x480 pixel photo, art, whatever, then you do your 200% scale to it. FCP (or Photoshop, for that matter) will be happy to scale up the image, but it will look pixellated/blocky. It's garbage in, garbage out, and you can't get it back. It gets even worse with JPEG's, which have macroblocking/compression issues inherent to the format. Ever look at a JPEG at 400% or 800%? You get the drift.

3. dpi doesn't matter, but pixel dimensions mean everything. This is because dpi is ignored by every device on the planet except printers.

You have to start thinking about pixels dimensions and dpi as 2 separate animals. They are only linked together by virtue of being in the same menu in Photoshop.

Of course, the nice thing about vector artwork is it can scale to any size and remain crisp - vectors by definition have no set size. When in doubt, always open eps or any vector files at a larger size than you think you'll need, then scale down. The ideal would be to use a vector file at it's maximum needed size and no larger, so as to avoid huge, unwieldy files.

With raster images (.psd, .tif, .jpg, etc.) the file size gets bigger and all apps slow down by devoting more horsepower to downsampling as the pixel dimensions get bigger. Adjusting dpi does not change that at all.

Perhaps it's better to think of dpi as PRINTER resolution, not TRUE resolution.

The only thing that complicates this is that if you have fine line detail in your vector object, then make it twice the size you end up needing and then decrease it's size in your NLE, you are risking losing the fine line detail to anti-aliasing. While I agree that it is better to downscale than upscale, if you DO happen to know EXACTLY what size you'll need the finished product at, convert to THAT resolution. You'll end up with the LEAST picture degradation.

Until the day video becomes vector (yeah, right... although a couple of video games back in the day WERE vector based ie. The Original Star Wars arcade game and the original Asteroids), we'll always have scaling issues. It's how we manage them that makes the difference to the overall quality.

I'm trying to connect a television screen to the computer via my camera now so I can see what the graphics REALLY look like...

I have firewire plugged to camera and camera to tv screen... but no picture and the "displays" part of preferences is not showing up the other display...


help again please... what am I doing wrong here?

Ok got TV going.... but it has a flicker when it displays the graphics...

I found how to draw bezier lines in keynote... and I can export Quicktime movies from Keynotes...

If I can solve the flicker issue then it may be the end of the problem...

(colours are a bit funny but that is easily fixed by changing colours)

Thankyou for the help so far everyone...

Janis: trying throwing a De-Interlace filter (should be under Video in the Video Filters dropdown). It WILL soften up your graphic but it SHOULD stop the flicker.