hello. i was wondering if i can make a cloth  that has different colors.
you know on one side it is green on the other side red.

is that possible in poser?

thank you.

i think i found the answer right?
http://www.renderosity.com/mod/forumpro/showthread.php?thread_id=2692518

Assuming that the camera is pretty much looking towards the negative Z axis, then we can make an assumption that is trivial to use in a shader. Any polygon whose normal is facing along the positive Z axis is pointing roughly toward the camera. Otherwise it is pointing away from the camera, and we're rendering the back side.

Therefore, if we use the Z component of the normal (N node) and compare that to 0 (Step node) then we can use that in a Blender to pick one of two colors. I've demonstrated that here. You can use a Blender node to drive any of the inputs to the shader. For example, if you wanted the front side to be shiny and the back side to not be shiny, you could use a blender node to drive the specular color - WHITE for the front, BLACK for the back.

thanks. i will try this

ok i tryed now this but it didnt work.
can you pelase writte step by step how do i have to do?

thanks 

I don't know what to tell you. Is the problem you don't know how to edit a material or what?

Are you getting only red or only green?

Is your prop truly one-sided?

The steps are to simply go into the material room and add a Blender node, a Math:Step node, and an N node. The ImageMap node above was just to give some variation to the surface and is not required.

Perhaps you could post a screen shot showing how you set up the material, and a render to show what you mean by "it didn't work".

this is how it looks
http://img441.imageshack.us/img441/7905/testms9.png

Are you sure those are single-sided polygons? It looks like classic poke through of a double-sided polygon. If that is the Poser "Cloth Plane" then you shouldn't be using this trick. That prop is double sided - there are different polygons on each side and you can assign separate materials to them and just color them like normal. For double sided props, set up two material zones, one for each side, and apply a small amount of constant displacement to both materials so there is no poke through.

Quote - Are you sure those are single-sided polygons? It looks like classic poke through of a double-sided polygon. If that is the Poser "Cloth Plane" then you shouldn't be using this trick. That prop is double sided - there are different polygons on each side and you can assign separate materials to them and just color them like normal. For double sided props, set up two material zones, one for each side, and apply a small amount of constant displacement to both materials so there is no poke through.

i can not find material zones.

Hold on - you still haven't told me what prop you're using. We're going to get mighty confused if I start talking about two-sided props versus one-sided props and I don't know what we're dealing with.

Is this the cape that you were asking about in the Blender forum?

If so, Blender defaults to double-sided polygons. select all of the cape and uncheck the Double-sided option in the mesh tools and re-calculate the normals outside (CTRL N) before exporting.

Quote - Hold on - you still haven't told me what prop you're using. We're going to get mighty confused if I start talking about two-sided props versus one-sided props and I don't know what we're dealing with.

i made the cape in blender. but i dont know the settings.

Quote - > Quote - Are you sure those are single-sided polygons? It looks like classic poke through of a double-sided polygon. If that is the Poser "Cloth Plane" then you shouldn't be using this trick. That prop is double sided - there are different polygons on each side and you can assign separate materials to them and just color them like normal. For double sided props, set up two material zones, one for each side, and apply a small amount of constant displacement to both materials so there is no poke through.

i can not find material zones.

yeah i think it is a double sided.
i have been now looking the whole day for those material zones.
i can not find them. i looked at google and nothing.

p.s.: my english is very bad. so i am trying very hard to understadnd you. about the double sidde: the cape is vissible only from one side. if you rotateit you can not see it. so is it double sidded or not?

I am having  difficulties too. I do understand what you suggested,bugginsbill, and I managed to  get it to work on a one sided plane that I got from the props/primitive section that came with p6.
However I didnt get anything usefull when applied to cloths.I applied it first to a dress I d/l from poserfashion and then tried it on simple Jessi cloths from the P6 library (the conforming vest and the grey dress) with no luck. On the grey dress,for example,  I expected  the outside to be red, and inside green, what happend was the front come out red,the back  green and the inside is just black. It may well have to do with the 1-sided/2-sided issue,but is there a way to tell,in Poser, if an oprop  is 1or 2 sided?
Oh what a shame. I don't even need it particularly at the moment, it just seemedlike  such a neat trick I couldnt help myself  but play with it!

ok i now downloaded the two capes from the freestuff section.
long cape and fantasy cape. doesnt work. can not get the right results.

Your English is fine - don't worry about that!

What I meant about material zones is that you can assign new materials to polygons on a prop using the Group Tool, selecting individual polygons and assigning a new material to it.

The long cape is one-sided and does work with this shader. Here I have made the outside blue and the inside tan.

It appears that you can't tell the difference between one and two-sided props in the preview. Even two-sided polygons disappear when you rotate them so that they are facing away from the camera.

I don't actually understand how two-sided polygons are represented in the geometry, so I can't tell you what to look for. I just know that the Poser one-sided square comes out red on one side, and green on the other, whereas the hi-res square comes out red everywhere.

Artpearl:

The gray Jessi dress works as well. It is one-sided, and I applied the red-green scheme to it as you can see. I exported it as an OBJ file and reimported it so it was no longer a conforming figure. Then I added a Wind Force object and did a cloth simulation to get the front of the dress to flip up.

When you say the back was green, that is correct. Remember, this trick is not paying attention to the camera. It is simply using the rule that any polygon facing North is green, South is red. If you move the camera, then you will be looking at green polygons. You can't move the camera. You must rotate your props or figures instead. The camera must be looking straight north, toward the negative Z axis, or it will "see" the front of polygons that are supposed to be facing the other way for the trick to work.

To both of you, I see problems with this technique besides the camera issue. The cloth simulation sometimes makes folds that pass through each other. When this happens, a you sometimes can see a North-facing polygon that should be hidden by a South-facing polygon, and you get the green where you should get red. This shows on part of the dress. I don't know what to do about that.

so if i rotate it then the colours will change right?

Yes - keep the camera in place, and rotate the object. As the polygons rotate they will pick new colors based on whether they are facing north or south.

By the way,are you sure the little bits of tan in the blue areas and vice versa are poke through problem due to the dyn.? I get something similar on the conforming dress,when I rotate the camera, (see pic). Maybe it occurs in the regions facing west or east,to use your analogy,  where roundoff inaccuracies determines if it gets assigned a 0 or a 1 almost randomly.

Anyhow,I think I got it now so I'll be able to utilize it sometime.This is so cool.

Thank you very much for sharing your knowledge.

thanks

Yea the roundoff error can happen too. But, with the correct viewing angle, you shouldn't be able to see those.

I haven't tested this shader yet and I don't know shader math, so this question may be foolish, but here goes.

If the camera angle can be a problem with this shader technique, might there be something that can be done with a blend node to combine similar shaders for X,Y,Z?  Then perhaps the blend node(s) could be animated, linked to parameter dials.  Then the parameter dials could be altered in the pose room.  Or with a Python script.  Or with ERC (if it works with such dials).  Maybe a callback in Python could check the camera angle and combine the three N shader blend parms appropriately.

Am I totally way off-base?

No you're dead on. I didn't bring it up because it is somewhat more complicated to explain - I figured rather than go into the math - people should just build the scene around the camera pointing straight north.

However, to be honest, all you need to do is this: You want to know if the dot-product of the normal (N) is aligned with the direction to the camera. To do that, just plug in the x,y,z vector to the camera from the current point on the surface. How you get that is a pain, but a Python script could do it easily. It is one of the features I'm working on with Parmatic (another one of mys cripts), putting the camera position into a node automatically and constantly.

Then you subtract that node from the P node (which gives the coordinate of the currently rendered point) and that gives you a direction. Decompose that into three values and plug into N - which will compute the dot product. If it's positive, then the point is facing the camera.

Several problems exist with this, not the least of which is I'm frequently dealing with people who A) don't know math very well, B) don't know the nodes very well, and C) can't put it together.

I totaly sympathize with you - people on the forum here seem to cover such a wide range of abilities and skills  it is hard  to know how to address issues. As for me, I am  good at Maths (my first love...),  very new to nodes, and thus would strugle to put it together, but I'll tell you what I am very good at - copying the examples you put up ! (very subtle hint,hint....).

this is a little more of topic.
i noticed that my cape on the inside is very bright. i changed no settings. the lightining is behind the cape. but somehow the cape is on the inside very bright.like a light was under the cape.
it makes no sense.

can anyone help me?

i have a problam. if i used normals forward then the cloth looked fine.

now i am using bagginsbill's GC gamma correction shaders. and this is what i get.

Attached Link: http://www.runtimedna.com/mod/forum/messages.php?ShowMessage=391935

I suspect that we did not turn on normals forward in the GC diffuse node. That's why it would behave differently.

But - given what we just learned in the linked thread, we have a new and better way to do the detection!

no no no . dont you see. i used only one light . and its above the cloth. i am using the basic square hi-res.
and the bottom part is bright red. it should be black.

meh ignore my post. you already answered my question. the normals forward on the node mut be checked. not only on the poser surface but also on the diffuse node itself.

meeh  ice-boy what are you thinking. no beer for you mister.

Quote - Hmmm. Not sure what to say. But go check this post in the node cult. This is a new technique that works great for the two-sided detection.

I suspect that we did not turn on normals forward in the GC diffuse node. That's why it would behave differently.

But - given what we just learned in the linked thread, we have a new and better way to do the detection!

bagginsbill does the link work for you?  i can not open it. and  i  also cna not find it on the forum. did they delete  the thread?

All links to RDNA are dead. All links from within RDNA are gone. The threads have new URL's since they moved to a new server, and the thread contents were not properly migrated.

I would not have done things that way. I would have made a forwarder from the old thread links to the new ones. It's not like the old links are hooked up to anything - it should be perfectly possible to support re-mapping old thread links dynamically. It is a database-backed application server after all.

this explains it.
now i dont remember how the shader was made.i know that in the shader there were two diffuse nodes. for each side. i didnt save it. i remember that when i used i got artifacts with raytraced shadows.

BB do you maybe know for a new way of getting a two side cloth with different colors?

I don't remember the details, but off the top of my head ...

Set one Diffuse for Normals Forward=On, the other Off.

Compare the outputs using a Math:Step node (must be hooked in the right order) - if they generate different information you're on the back side.

thanks .i will try

notice in the render the orange edge? in the render i have only one light and its behind the cape. its a rim light. there should be no edge on the cape. i tryed with ''normals forward'' but it didnt change.

the cape has two parts. the outside and the inside.

what could the problem be?

I don't know why you say there should be no lighting on the edge. The light is tilted up.

I can't tell from just those images if the cape is concave or convex with respect to the light. I really have no idea what the geometry is.

But to understand what the light sees, I suggest you set it to depth-mapped shadows, start a render and stop it. Starting a render sets up the shadow cameras.

Now switch to the shadow camera for that light. You can now see what the light sees. What do you see?

If you can see the rim then that is why it is lit.

in the preview render i used an extra light to make it brighter so that we can see it.

in the first render there is only one light. and it is behind the cape. i tryed with DM shadows and i still get the same problem.

i think it is the geometry.

bagginsbill is it easier in poser 8 and PP 2010 to use different colors ?

I don't think there's anything new to make this easier.

www.runtimedna.com/forum/showthread.php

the problem with raytraced shadows is that we get those artifacts.

i beg you bagginsbill i beg you can we find a solution for this?

Quote -
meeh  ice-boy what are you thinking. no beer for you mister.

meep :(