My first post...
I'm trying to get images for an animation in a tutorial. The frames will consist of a moving LED beam.
I've got it working, but I'd like the gradient used for the beam to be more representative of the the actual light dispersion.
The density of the gradient should fade from left to right, AND
the density of the gradient should fade as a function of the "off angle". Brightest on the sensor pole immediately adjacent to the pole, and fading off as the angle get larger. For example, 50% intensity at +-7 degrees.
That's what the actual radiation pattern of the LED is. Decreasing with the square of the distance, and decaying to 1/2 intensity at +-7 degrees. Having the graphic reflect something like the actual dispersion of light will help make some of the points of the tutorial more intuitive.
I think I should be able to get what I want with a gradient/mesh/transparency and or mesh fill. Maybe combining multiple objects somehow. But, I've played with all those tools and can't get what I want.
Ignoring the need to also fade from left to right, I can get this using the fountain fill with linear gradient. But, even this subset of the task isn't what I want. The gradient is vertical, versus radial around the emitter point:
I tried elliptical gradient, but the gradient starts from the center of the object. Maybe I could somehow combine and object with an elliptical gradient with its center over the emitter point source. But, I couldn't figure out to do that. I kept loosing the fill all together when I did virtual segment delete after combining the curves
I see a tutorial on how to do what I want in PhotoShop, so I'm about ready to give up and try that. I'm not that skilled in CD, but much better in CD than PhotoShop, So I'd like to stick with CD since this is just a small piece of the graphics I need to do for the animation.
Any help would be greatly appreciated!
DaveThomas said:I tried elliptical gradient, but the gradient starts from the center of the object.
The gradient does not have to start from the center of the object. Try using the Interactive Fill tool with an elliptical fill, and see if that gets you something closer to what you want.
Hey, that IS closer to what I want!
I guess I didn't try elliptical fountain fill using interactive fill--just from the fill setting in the object property window. The ellipse was always centered on the object and I could not figure out how to move it.
While this is close, I'd like the intensity to decrease more rapidly off access. Pulling the circular handle down does that, but when the ellipse intersects the triangle the radiation intensity is totally white:
Is there a straight forward way to get steeper gradient off axis, independent of the left to right gradient?
You can get a steeper gradient off axis by blurring the entire image.Keep in mind that one does not normally see a beam of light -- just where it hits a wall or the ground,
Blurring does help some, but it really just fades the edges a bit.
On the PS tutorial, they used two, overlaid shapes. The in side shape was smaller and a brighter color of white
The shapes are then "blended" (might not be the correct terminology) so the off axis gradient could be controlled as desired (at least, that's my hope).
I thought I'd give that a try in CD. But, CD blending is NOT what I want. That seems to create shapes between the shapes to be blended that evolve from one to the other. Maybe I need to do something with transparency.
"Keep in mind that one does not normally see a beam of light -- just where it hits a wall or the ground"
Well, that is true if the medium was free of moisture and dust particles. Probably true enought on an Arizona desert night or most places out West or high altitude.
But just take a flashlight outside on typical, hot muggy, night here in the South East. The beam is certainly visible!
Kids love to play "Star Wars" with powerful flash lights out in the back yard.
But, it really doesn't matter for the points I need to make in the video It's IR "light", which isn't visible at all. But, the video is about where and why IR receive devices are located and how they are scanned to minimize interference from the many emitters that are sequentially turned on.
Actually, blending does look like it might get me what I want. I'm still playing with it, I'll post what I end up with. Hard to describe what I'm looking for without a picture. And, when I have the picture, I'm done!
This is close enough to what I need.
I first created a beam clone and used the interactive fill tool with elliptical fill to get the left to right gradient. Then I made a copy of the same beam and enlarged it slightly. Here are the two beams before I removed their outlines and chose
After removing the outlines of both shapes, I used the blend tool to get the top image. From there, I can convert to a bitmap and add blur or maybe other effects.
II think I can play with the difference in color between the two cones and their relative size to get something a little close to the actual emitter's radiation pattern (50% intensity roll-off at 7 degrees off axis).
I was hoping I could blend N different cones to get very close to the actual off-axis intensity roll-off, but it seems you can't blend more than two objects, or use an object group that's already been blended and blend again with another object. Please let me know if I'm wrong about that.
Regardless, what I have is very close to what I need. I'm spending so much time on getting it as good as possible, since it will be repeated hundreds of times in animation frames. So, I'll wait for any additional suggestions before I do that (tedious) work.
Thanks for the help!
Thought I'd post what I'm going with, lacking any other suggestions.
The optics on the emitters "focus" the beam to +/-7%. But, it's not like the intensity is zero once your off axis more than 7 degrees. That's just where the intensity is 50%. That's part of the point I need to make.
Anyway, the image doesn't reflect the actual off-axis roll off, but it's good enough for illustration purposes. It would be "nice" if the outside edge wasn't a line at all, but just faded all the way to white, but I don't know how to do that.
"I was hoping I could blend N different cones to get very close to the actual off-axis intensity roll-off"
Actually, you CAN do this.
The "trick" for me was to create the objects with the same left to right gradient. Then, I change the start gradient color to progressively lighter colors for the wider angle cones.
Starting at the center, I blended the the most acute angle clone to the next wider one. Then blended that "blend group" to the next wider cone, etc.
So, you can get the change in intensity versus off-axis as close as you are willing to bother--just more objects to blend.