mesa - coloring and texture testing

Started by folder, July 29, 2009, 12:00:42 PM

Previous topic - Next topic

folder

a while ago Njeneb graciously gave me some of his tgd files to I could learn how to colour mesa's and work on creating such terrains. reading thru the old forum files. I found where TU said he used merge shaders and duel strata  here i have used merge
shaders and 2 power fractals for colour. this was an attempt to create some of the colouring i have seen in Az and new mexico. displacement however has been an issue - even after studying TU canyon tutorial I have not had any real luck - so any tips or hints will be appreciated. while not as good as TU lastest canyon, I am learning

(Njeneb again thanks so much)
folder

buzzzzz1

Not sure what look you are trying to achieve but in this image http://buzzzzz.deviantart.com/art/Southwest-Red-Rocks-100033448 much of the displacement was simply done in the Base Color Node. Might work with what you have in place already. Try this:

Go to Base Color displacement node and displace "along Lateral normalized" set displacement to somewhere between 15. or 30. Don't worry with the requires "Compute Normal"
Then go to the Scale Tab and set the scales,  Feature scale to 30. and Lead-in scale to around 3000. Do a crop and see what you get.  Pretty easy eh?  This is best for distant shots BTW.
XFX 750i SLI  nFORCE Mobo
Intel Core2 Duo E8400 3.00 GHz Wolfdale
8 Gig Corsair DDR2 6400
Nvidia GForce 9600 GSO 768 MB
XP Pro 64 bit OS

folder

buzzzzz1

thanks much - this is the displacement effect i wanted - something different than just the straight horizonal lines - will try now

folder

Tangled-Universe

Hi folder, you have a good start on colors and pov here. The displacements could use some work indeed.
To have less "staircased" strata you need some overhanging structures / lateral displaced features.
You can do this quite easily by using a redirect-shader. Feed a powerfractal to its X- and Z-ports and set the displacement to a couple of 100s.
Below that redirect you could also use serial strata-shaders instead of merged one, because they can be a bit tricky.
In order to get not too much staircasing the hard layer steepness and plateau buildup settings are key.
Further I suggest to look at FrankB's advanced cracks in the file-sharing section.

Good luck and show us your progress please :)

Martin

Henry Blewer

I sometimes just use a strong power fractal set to ridged for displacement. Use the measure tool to find out what size scale you want. At the bottom of the tab there are three inputs for x,y,and z stretching. Increase the y input by up to 3. This can give nice cracks vertically on a surface.
http://flickr.com/photos/njeneb/
Forget Tuesday; It's just Monday spelled with a T

folder

Hi, TU and Njeneb - thanks for the advise, just got home will try tonight. I have attached screen shots of the terran and shader nodes. I am assuming I should make the changes in the shader section. TU if you look at the terrain section , you will see that some of the connections mirror what you were doing in your tutorial, as I mentioned I have not had any success. but was wondering if some settings were cancelling other out.

david

Henry Blewer

I like to make my 'seams/cracks' in the terrain tab, but the shader tab will also work well.
http://flickr.com/photos/njeneb/
Forget Tuesday; It's just Monday spelled with a T

Tangled-Universe

You can send me your project-file if you want and I will make some little adjustments which should point you in the right direction.
You can find my e-mail in my profile.

Cheers,
Martin

folder


folder

cracked strata effect

playing a bit more i seem to have come up with a cracked/chipped effect or al least a step in that direction. it does seem a bit more layered

david

Tangled-Universe

I just started with the tgd you sent me and I see you have a gradient patch size of 2009 in you compute terrain.
Also, some of the powerfractal you're using have crazy values for displacement roughness and spike limit.
Generally you keep this between the default values and not >100.

Anyhow, I will add some changes and things to it and will send it over to you this night.

folder

thanks again

actually i got the 2009 number from a earlier posting of yours in another thread -  :-

do you think this colouration pattern is a step in the right direction

david

Tangled-Universe

Quote from: folder on August 03, 2009, 02:11:24 PM
thanks again

actually i got the 2009 number from a earlier posting of yours in another thread -  :-

do you think this colouration pattern is a step in the right direction

david

Really??? Where?

folder

actually :-[ you are right it was Volker in an compute terrain thread

david

Tangled-Universe

Quote from: folder on August 03, 2009, 04:03:00 PM
actually :-[ you are right it was Volker in an compute terrain thread

david

hmmm...that surprises me as well...anyhow for the sake of clarity: the gradient patch size determines at which patch size the normals will be calculated.
So with a setting of 2009 that will mean the normals will be calculated in patches of a size of 2009 metres. Logically, that doesn't offer much detail normally.
Still surprised you did see some detail and stuff.