PDA

View Full Version : noob rendeman/RAT ..texturing..


samel
03-14-2006, 02:14 PM
i saw someone do this in slim..

surface<--image file<--mayaUV

with this tcl in the maya UV STmatrix slots

[st0 myUV]
[st1 myUV]

where "myUV" would be referring to my (renamed) place2Dtexture node..

and with this telling prman to use my maya uv's instead of it's own st space thingee

this seems like something i'd want to do very much ;) well, how come it's not in the docs
..all they talk about is how to project your texture with a renderman coordinate sys.. and yeah, i might want to do that aswell sometimes but why no mention of the mayaUV node..
anyway, the above is absolutely not working and prman pretty much doesnt give a crap about what i do to my uv's.. so question..is this the way to do it or am i missing something.. (and why on earth arent there any slim/RAT tutorials on the whole www)

//s

samel
03-14-2006, 04:32 PM
ok.. i think i managed to get it working after some messing around.. but still..
slim/RAT shader tutorials please...

//s

rendermaniac
03-14-2006, 08:07 PM
Just wait until you try doing projections! hint - the MayaProjection template is a lot better than the Projection one. The docs could be a lot better - saying that they have improved a lot - and are stil a lot more complete than the Mental Ray ones.

Simon

samel
03-16-2006, 10:34 AM
Just wait until you try doing projections! hint - the MayaProjection template is a lot better than the Projection one. The docs could be a lot better - saying that they have improved a lot - and are stil a lot more complete than the Mental Ray ones.

Simon

yeah, i hear you.. the mental ray docs are a disaster.. but there are still a lot of helpful tutorials out there on hooking up your stuff in the right way such as jozvex's stuff
here, http://www.jozvex.com/tutorials.html and the lamrug site is quite in-depth..
anyway, i'm moving on to applying some occlusion to my test-scene.. and, i'm having some trouble with the sampling.. i have a trace occlusion through a spline to a constant..
sampling is up to 512 and im still getting some artifacting.. doesnt really look like the sampling is too low but more like a ray bias error.. (blocky, square blotches in the occlusion)
any ideas..
//s

playmesumch00ns
03-16-2006, 11:46 AM
If you could post a pic it would help.

512 is still quite low. There's a fernickety trade-off between sampling and interpolation. What is your max error set at? You can either reduce that, or increase your samples. For really clean occlusion, you want to be looking at about 2000 samples.

samel
03-16-2006, 02:37 PM
hmm.. 2000 made the occlussion really smooth.. however, i'm experiencing some really weird results as shown here..

http://i14.photobucket.com/albums/a303/le_bonbon/artefacting.jpg

now.. the poly-mesh is split as the colors indicate to the far left..

second from left has "visible in reflections" on for every piece which produces some
kind of blotchiness on the floor..

third from left has "visible in reflection" off for the middle piece (label) which produces weird white areas.. rays not "reaching" all the way in and creating erroneous occlussion?

last, fourth from left has "visible in reflection" off for the bottom piece (label is back on) which eliminates the white stuff but the blotchiness from nr. 2 is still there but much more subtle and, ok one might let that slip in production..

now, the max error is currently at default -1 and i have messed with and it didnt do jack..
this is at shading rate 5 but i took it down to .5 and the artefacts wont budge..
also notice how turning off v_i_r on the label (third fr. left) completely alters the lighting on the floor
which has a "matte" -surface attached to it(!!)..

WTF is going on..

thank you for answering ;)
//s

tweeeker
03-16-2006, 08:21 PM
i'm not too familiar with the slim templates, but i think in the latest version of prman, the new maxvariation paramater overrides maxerror. So that might be why your not seeing anything when you change maxerror. The artifacts in your image do indicate that maxerror/maxvariation might be to high though.

samel
03-17-2006, 09:54 AM
yeah umm.. ok so the max variation did the trick..turns out i had been adjusting the max error for the irradience cache *stupid* , i just went for the first max error i saw.. sorry
..does anyone know where i can read up on these kinds of things..?

thanks //s

tweeeker
03-18-2006, 09:10 AM
In the prman docs theres a whole application note dedicated to gi and ambient occlusion. I'd certainly have a read through that.

CGTalk Moderation
03-18-2006, 09:10 AM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.