#Brush Mapping again?
15 messages in this thread
Stan, I tried your suggestions contained in message # 2897 and sorry to
say, that didn't take care of the mapping/gnippam problem. Instead of 2
discrete images though, the forward/backward brush was connected. I had a
sphere of 100 units, axis of 200 Z and 320 Y. The axis was at -100 x, -100
z, and 0Oy.
Oddly enough this resulted in the axis bisecting the sphere when viewed
from the Right View….. i.e. (|). I had done this accidentally before
and noticed that it didn't seem to matter, I originally thought the axis
had to be between the camera and the object.. i.e. – |(), where – is the
camera, | is the brush axis and () is the object.
Re-reading the message, I see something that might be confusing. When I say
"sphere of 100 units, axis of 200 Z and 320 Y" I am of course talking about
2 objects, the sphere and the IFF brush axis. I'm still lost as to how to
do the mapping without getting mirrored brushes, any suggestions? Thanks
for all the help, any clues as to what's in store from Impulse? How about
an easy way to convert IFF images (outlines?) to Silver objects? If I
could do that, I'd forget about mapping brushes to objects and just work on
the objects themselves.
I too tried the sphere and still got mirrored text. I then even tried just
a simple surface with an IFF wrap, it too had the reversed text on the back
of the surface.
Phil – I must be missing something here. Yes from the back side of a
surfface or plane the brush will be mirrored. or if the your camera is
placed inside a sphere or other object, or if you can see in side an object
like a tube. Other wise, you should only be getting one copy of the
picture on the outside of the object. I make spheres and other things all
the time and rotate them, and ALWAYS have only one copy of the brush on the
object. <–Stan
I am also continuing to have the mirror image sphere problem. The
mirror image appears on the outside of the sphere, not the inside. I have
moved the IFF brush all over the place, changed its size, and all it does
is move where the mirror image starts and finishes. I had the same
experience as Keith when following your suggestions in message #2897. It
looks like the mirror image merges with the forward one in that case
because the width and position of the brush moves the center of the brush
off-center on the sphere.
I tried "solving" the problem by splitting a sphere in half and wrapping
both halves, but a seam shows.
By the way, the problem I mentioned some time ago in which a perfectly good
solid model object disappeared when ray-traced seems to have been related
to object size. Since it was stated that larger objects trace quicker, I
was making objects in the 6000 unit range, and it seems that one light
works fine in solid model on an object that size, but only illuminates a
small portion on a trace. I have solved that by making objects smaller.
But I hope you can work something out on this mirror-image situation.
Thanks for your help.
In solid model mode, there really is no size to the world, so you should be
able to make things large, and/or move all over the place. In Full Ray
Trace mode, however, you are limited to the 1000 units from 0,0,0 for
objects. Lights and the camera though can be placed anywhere. I'll get
with the programmer in the morning, and we'll start going through the code,
to find out why you are getting the mirror effect with your brush wraps. If
it's a bug, you two guys must be doing something that we haven't
encountered. (like a series of settings that trigger the error) We need to
find and correct it as soon as possible.
Speaking of brush-mapping, I was wondering what the effect would be,
if I were to make the various parts of an amiga, and map a brush of the
details onto the objects. Do you think it would look real, or would I need
to actually make the drives, for instance, part of the object? –Eet–
That's one of the fun things about Turbo Silver. You get to experiment
just to see what happens! <–Stan
Would it help if I mailed you a disk with the complete cell/setup for the
mirrored IFF problem?. One thought, what do you use to generate your IFF
brush? perhaps its the brush not the sphere thats funny.
That's an idea…… I've used Dpaint, and have also manipulated the brush
images with Pixmate. I was originally using a full IFF screen as a brush,
then yesterday tried an actual Dpaint brush, with the same mirrored
results.
I've tried a Dpaint full screen and some partial screens created with
Photon Paint. Same mirrored results….
What I wonder is, are we the only ones with this problem? Brush mapping
was one of the things I intended to use heavily, I have stuff I've used as
static screens in industrial videos that are begging to be jazzed up with
some motion.
Keith & Phil, it kind of sounds like you are not wrapping X and Z but
rather are using FLAT X and FLAT Z. This would produce te results you're
describing. Just a thought. Kirk
Nope, sorry. I have tried all combinations of Flat vs wrap x/z.
Phil – Yes that would help. ……….Come to think of it, I always use
our paint program, Diamond, to do our brushes. Perhaps if you are using
DPaint II you may have save the file with the stencil option on that can
cause some wierd results, also I think Digi-View 3.0 has some funny info in
the header that could cause TSilver =et confused. Also earlier versions
of Photon Paint)vd{hMome offbeat header info, but we tried to catch the
Photon chunk . Stan
I have been using IFF files from DPaint II, some of which have
originated in Digiview. I also tried some sample IFF files from other
programs…same results. By any chance are the IFF pics you are using made
of symmetrical designs? Perhaps you are getting the mirrored effect
without realizing it. All of mine are logos and lettering, so it shows up
quickly.