vade

im wondering about why the following patch has different output 

Im sending the same geometry matrix to jit.gl.render, sketch and 

Id love to be enlightened as to why. Im assuming it has something to 

I LIVE! I LIVE! I LIVE! I LIVE! I LIVE! I LIVE! I LIVE! I LIVE! I 

You will not be saved by the Holy Ghost. You will not be saved by the 

im wondering about why the following patch has different output  
depending on my rendering method

Im sending the same geometry matrix to jit.gl.render, sketch and  
mesh, and each one looks different.

Id love to be enlightened as to why. Im assuming it has something to  
do with normals.

I LIVE! I LIVE! I LIVE! I LIVE! I LIVE! I LIVE! I LIVE! I LIVE! I  
LIVE! I LIVE! I LIVE! I LIVE!

You will not be saved by the Holy Ghost. You will not be saved by the  
God Plutonium.

opengl-output-issues-with-varying-rendering-methods

Id love to get a thought from the hard core Jitter folk...

No one has any clue about this?
Id love to get a thought from the hard core Jitter folk...

I took a look at your patch. The obj works nicely because it takes

it's normals from the text file. So, whatever program generated the

obj file also generated the normals to make it look proper.

The mesh object doesn't do a good job of "smooth shading" when in

@draw_mode triangles. If you don't give it a normals matrix, and set

@auto_normals 1 it jit.gl.mesh will make it look bizarre because it

uses the triangle vertices (a guess here) to calculate the normal. 

Depending on vertex winding and thus which way the normal is facing,

you will get a jagged, checkered look. If however, you explicity set

the vertex matrix by sending a 3 plane matrix in the 3rd inlet, it

will look just like jit.gl.model's results. For some reason, sending

an 8 plane matrix in jit.gl.mesh's first inlet doesn't set the normal

matrix like I thought it was supposed to. So, you have to break the

matrix up and set the matrices on a per inlet basis. Personally, I

think the jit.gl.mesh object could use this feature as it would be

very simple to add as it is a pain in the ass to unpack and repack

As far as the sketch results are concerned, I haven't a clue.

The patch below shows how to get good results from jit.gl.mesh.

Hey there,
I took a look at your patch.  The obj works nicely because it takes
it's normals from the text file.  So, whatever program generated the
obj file also generated the normals to make it look proper.

The mesh object doesn't do a good job of "smooth shading" when in
@draw_mode triangles.  If you don't give it a normals matrix, and set
@auto_normals 1 it jit.gl.mesh will make it look bizarre because it
uses the triangle vertices (a guess here) to calculate the normal. 
Depending on vertex winding and thus which way the normal is facing,
you will get a jagged, checkered look.  If however, you explicity set
the vertex matrix by sending a 3 plane matrix in the 3rd inlet, it
will look just like jit.gl.model's results.  For some reason, sending
an 8 plane matrix in jit.gl.mesh's first inlet doesn't set the normal
matrix like I thought it was supposed to.  So, you have to break the
matrix up and set the matrices on a per inlet basis.  Personally, I
think the jit.gl.mesh object could use this feature as it would be
very simple to add as it is a pain in the ass to unpack and repack
matrices for opengl geometry.

Ah, Wesley, thanks. For some reason I hadn't thought of manually 

packing/unpacking the normal array from the model output. Thanks - 

I humbly submit a request to have jit.gl.mesh take the normals from 

the 8 or 13 plane matrix that would 'normally' (no pun intended) come 

out of any gl object with matrixoutput 1 enabled. perhaps it would 

only be overridden if one manually entered a textcordinate array or 

Ive shunted in your version wes to my main patch and it works fine. 

as for jit.gl.sketch.. anyone? Ive tried playing with the syntax 

drawmatrix triangles 0 0 0 0 or 0 0 2 2 etc etc, with no luck.

Ah, Wesley, thanks. For some reason I hadn't thought of manually  
packing/unpacking the normal array from the model output. Thanks -  
that makes total sense.

I humbly submit a request to have jit.gl.mesh take the normals from  
the 8 or 13 plane matrix that would 'normally' (no pun intended) come  
out of any gl object with matrixoutput 1 enabled. perhaps it would  
only be overridden if one manually entered a textcordinate array or  
normals array to the proper input.

Ive shunted in your version wes to my main patch and it works fine.  
Thanks.

as for jit.gl.sketch.. anyone? Ive tried playing with the syntax  
drawmatrix triangles 0 0 0 0 or 0 0 2 2 etc etc, with no luck.

OpenGL output issues with varying rendering methods