Displace with "tight bounds" and "shift"??

[RSF]

Hi!
Is there a way to work use "tight bounds" and "shift" at the same time?
I need to displace a surface both ways, up and down but the areas that need to be displaced are very small and sparsed so "tight bounds" is a must.
The problem is that in order no dislpace both ways I need to use a 50% gray to the vast area that I don't want displaced but then "thight bounds" has no use.
I'm I correct? If so anyone knows a workarround?
Thanks

[vlado]

It should be fine; what makes you think that you cannot use both options together?

Best regards,
Vlado

[RSF]

Hello Vlado! Thanks for the fast response as always.
Well...I'm guessing that because all the area that is to stay in place (not displaced) at a 50% gray and not black.
I'm displacing a golf course terrain to create bunkers (the surface follows the general topografy) so I have a all black map with litle brighter areas to form the bunkers and of course I have a negative displacement distance.
Now I'm trying to displace some bushes near the bunkers (not everywhere) so I made another map at 50% grey where the bunkers are darker and the bushes brighter and used a positive distance (3meters) and set shift to -1.5. I thing I doing this right.
The problem is that is seams to take more time presampling the displacement.

Ok...I'm making same tests...here it goes..
The map is a all white except the darker bunkers and I'm using 3d displacement, view dependant and tight bounds

1# situation
map is inverted(all black except bunkers)
Displace amout -3
Shift 0
Render time 15sec (without tight bounds...10sec?? ok...render is to fast to have advantage in presampling)

2#
Maps rgb level at 0,5 (not inverted)
(I didn't use offset because displace seams to clamp it so I doubled displace amounts)
Displace amout 6
Shift -3
Render time 31sec (whitout tight bounds 38secs)

The result is exactly the same, no diference in the final render what so ever.

Well...what is the right way displace both ways when the displaced area of the surface is small?

I'm I missing something?? Wouldn't be the first time

[vlado]

Hmm, I'll try to reproduce this, but I don't think it has anything to do with the "Tight bounds" option...

Best regards,
Vlado

[RSF]

Hi again...
Well...you're right, it as nothing to do with the "tight bounds" options but that's exactly the problem because all the surface is beeing displaced and shifted back to the same place.
If "tight bound" would filter out the exact color value calculated with both "displace amount" and "shift" would be great!
I think it wouldn't be necessary any major pragrame changes...would it??

[vlado]

Works fine for me here with a simple Noise texture; if you want, you can email me the scene to vlado@chaosgroup.com for investigation.

Best regards,
Vlado

[RSF]

Simple noise won't reproduce it because in either case the surcafe is being displaced everywhere.
The problem is that I have a map thats largely black.
The "tight bounds" option is an optimisation for this kind of situation but the situation itself (map being largely black so no displace to be computed) would be better than having to displace all the surface. This is what happens when you have to make the black areas 50%gray in order to displace both ways(up&down)

Here's the map I'm using http://clientes.netvisao.pt/ruquita/...rs_disp_LR.png

[vlado]

Ok, I will take a look; however the tight bounds check looks for flat regions in the map (at any level) across a triangle, not for black areas specifically.

Best regards,
Vlado