Hi all. I am trying to figure out the simplest way to accomplish something with the Vidiot and wondering what additional equipment I would need. I am looking to have a video camera going, while at the same time having a nice color blend or something going. I want the camera to mask out the color so that the color is only visible through the parts of the camera image. In software I could just use a luma mask on one of the layers but I’m having a hard time conceptualizing how that could work with the Vidiot.
The Vidiot keyers work a little differently. You can have a pattern patched in the colorizer section and then take the output of the luma processor (I like to use the rear output) and patch it to the Key 2 input. If you set the Color Negative Gate to Key then the image running into the luma processor will be superimposed on the pattern patched in the colorizer section. The superimposed image will be the color negative of the colorizer settings.
So I could use the camera as the key source in, but how would I take the color out from the Vidiot to put it into the background or foreground in of the doorway?
Honestly, keying is a little confusing (at least to me) in LZX-land, because they tend to use “keying” as a verb to mean creating a mask, whether that be a “hard key” mask (Cadet VIII, Castle ADC using just one output) or “soft key” mask (Doorway, Topogram).
Whereas in all other video situations, I have always used “keying” to mean the process of actually masking out part of an image and, generally, replacing with another (maybe I’ve been doing it wrong?). Generally, in LZX-land, it will be some sort of mixer or fader that will be doing that for you. Thus many LZX compositing modules can do this in different ways (Visual Cortex, Marble Index, Color Chords, 3x Cadet VI, etc.).
Yeah the terms aren’t used here as in conventional video production. I’m very interested in how I could achieve what I’m talking about though if you have any suggestions. Luma only would be fine even.
As said above, there’s no way to do what you are talking about with a Vidiot alone. It simply doesn’t composite multiple images in that way — its keyers only invert. And all other compositing between camera and oscillators involves inverting or mixing (rather than multiplying).
So you need another module (or modules) to do the compositing. And then there are different ways to do it depending on exactly what effect you are looking for.
Here are a couple options using just a Visual Cortex:
But, again, other modules — Marble Index, Cadet Fader x3, Passage, etc. — can achieve similar effects (though you may need an adtl. module to create that hard key mask).
Thanks. But I’m not trying to do it with the vidiot alone, I’m asking what other modules would I need to be able to do it? The picture above is similar to what I’m envisioning but how would I do it with the Vidiot and other modules even? My point is that I can’t separate the color out of the vidiot. There’s no way to just get the color processor out and put it into a key because the key is don’t process color right?
Or, rather, you can get the colorizer out (12/13/14, on the back), but you will create a feedback loop if you do because you will also be putting your composited image back into the colorizer for final encoding and output.
You will have to send the individual outputs from the pattern generator section (16-36, 49-51) out and create your mixed pattern in whatever you use to do the compositing (Cortex, Marble Index, etc.), as well as the output of your luma processor (10) or just the camera output (48), if you don’t need to process it.
And then the resultant composited image would go back into the colorizer CV inputs (52/62/72).
Depending on which module(s) you use to do the compositing, you will probably lose some of the functionality found on the Vidiot colorizer (some of the per-channel inverting for instance, or the solarizing) and so it may be hard to recreate the exact pattern you had on the Vidiot externally without adtl. modules. But it is impossible for me to say exactly what you will need to use in all cases because Vidiot’s option set is unique. You can recreate any given mixture of effects with other modules, but which modules will depend on which exact outputs/inputs you were using to create your initial pattern.
But, if you want my advice on a single module to add, I would recommend a Visual Cortex. It does the compositing you are looking for AND has some colorizing abilities similar to Vidiot. Also has built-in hard keying. Plus has the addition of a ramp generator, which is very useful in pattern-making. And then its multiple outputs can also be handy (for genlocking a camera, for instance).
Of course, now I’m realizing that I was being dumb…
If you use a Visual Cortex, you have that adtl. output encoder. So you can create your pattern using the Vidiot colorizer exactly as you would before and then, yeah, send that output (12/13/14) into the Cortex for compositing AND encoding.
That is perhaps the biggest advantage of a Cortex over something like a Marble Index.
