CompuServe Thread

#Moving Clouds

14 messages in this thread
#46437From: John EllisJun 16, 1993 10:33 AM
Hi John, actually I was thinking more along the lines of fractal clouds or something like the SMOKE (sxp) that would create random cloud patterns. I also use DGS which has a shader which can be applied with opacity that generate random clouds, the problem is that it is very labor intensive as DGS uses a "Style" sheet which in this case would have to be modified for every frame to get random motion in the "shader generated" clouds. Something like the "NOISE" sxp would work but the pattern is fixed even though you can move it on the xyz plane. SMOKE looks like it might be able to do it but I would like to know more about it and how it works. I will check out Jamie's clouds though as of course cloud maps are very useful for other projects. Thanks – John
#46465From: Don LandisJun 16, 1993 12:50 PM
I tried to use smoke to create a puff of smoke and the effect didn't look real because the IPAS module makes the smoke look like it's in a bottle or any shapped glass container. Real clouds don't have any bounding borders. You could retouch the image in photostyler but if it's a flic that could be a bear. Depending on your layout you could bound the smoke beyond the image safe zone (overscan). I was told that smoke.sxp was designed to simulate a light beam through a fog or smoke.
#46575From: John EllisJun 17, 1993 12:31 PM
Don, my thought was to create a long rectangle very thin into the Z axis and use that as a bounding box then as you travel under the box you would have some depth, and using smoke as an opacity map render with a blue background, but it sounds like the SMOKE.SXP doesn't quite do it. We need something that can do fractal clouds. Technically it shouldn't be too difficult as the NOISE.SXP would almost work, if you could randomize the values and change the pattern. Well anyway, thanks for your response, I'll keep looking. – John
#46584From: MARTIN G FOSTERJun 17, 1993 1:59 PM
San Diego John! Hey mate, remember me? FWIW, I have a totally impractical idea: Why don't you do the old movie cloud-tank trick. You set up an aqaurium filled with coloured water (or similar) and your camcorder at an appropriate perspective and then mix in the appropriate fluid which will billow like clouds. Some research might be necessary on this – a trip to the library and looking for a film special effects book. The trick would be to get the correct fluids and good lighting You could then use the video for a background. Told you it was impractical but if you get it right it would be a killer effect <g>.
#46589From: John TissavaryJun 17, 1993 2:30 PM
Martin– Excellent idea. I really like the idea of compositing traditional film trickery with modern 'byte' induced hallucinations.
#46625From: MARTIN G FOSTERJun 17, 1993 6:25 PM
John, I get lots of ideas from reading books on film effects – we can learn alot from the old-timers <g>.
#46724From: John EllisJun 18, 1993 2:10 PM
Martin, I think you've been in L.A. too long… or you've been hanging out with the special effects guys instead of the animators! We'll have to get you down to San Diego for some fresh air and sunshine. All kidding aside, it would be fun to try. You've got alot of imagination, that's probably why your such a good animator but don't let it go to your head… LA's got too many of those types already. By the way I noticed you've been checking out VTR's. I just got my EVO-9650 and love it, what a sweet little machine. Took me a couple of days to write the code to control it, and its humming along like a champ. If you get anymore wild ideas pass them along, they're entertaining and some of them might even work. Just kidding. CIAO.
#46749From: MARTIN G FOSTERJun 18, 1993 7:42 PM
John, Have fun with the EVO-9650 – all the people I know who have one really like them. I'm really envious since I don't have any of that type of equipment in my setup 🙁
#46800From: John EllisJun 19, 1993 11:30 AM
Well Martin, its taken me four years to finally put a system together that does what I want and I am very happy with it. To give you a little history not that it matters much, but I bought 3D Studio when I already had Digital Arts, because I saw it was the wave of the future, instead of putting the money towards a deck. I got Digital Arts second hand but it still cost me $6000 and that of course was before the price dropped because 3DS came in and kicked the big soccer ball. And of course 3DS is so much easier to work with, things that took me weeks to work on, I can do in days. And I'm sure with your talent and devotion to the art you will not have to wait so long. CIAO and no more 🙁 I appreciate your command of the English language it must be Native.
#46737From: Larry MintonJun 18, 1993 5:15 PM
Just don't forget to take the fish out first…. LAM
#46750From: MARTIN G FOSTERJun 18, 1993 7:43 PM
Larry, Oh no! You leave the fish in to give that abstract Dali-look <g>.
#46599From: Don LandisJun 17, 1993 3:34 PM
Might work if the bottom of the box could be hidden behind some buildings or other visual obstruction and the top and sides of the box would be outside the viewing picture area. I'll try it this weekend. If it works you can buy the module with assurance.
#46725From: John EllisJun 18, 1993 2:10 PM
Don – You got the idea.. I'll be interested to see what you think. It should be workable, just don't know how it'll look. Thanks – John
#46666From: John DowdellJun 18, 1993 2:53 AM
I've run into a similar "surface smoke" situation in RenderMan…fwiw, I've had luck with the workaround of duplicating the object a few times, successively smaller, with the same central point, and applying the smoke at various opacities…looks better when moving than still. Another workaround was to create a second smaller/larger duplicate object and fill it with a fairly transparent color…RMan "glow" (different method than IPAS glow, I think) works well, too. There's a package on the Amiga called "Aladdin 4D" that apparently handles turbulence- and image-controlled volume shading…examples I've seen have had some surface artifacts, though…anyone have more info on true volume rendering on the desktop? Do the algorithms and CPUs meet yet, or are we still a bit aways from that goal? jd