cKleinhuis
|
|
« on: December 11, 2009, 02:06:33 AM » |
|
|
|
|
Logged
|
---
divide and conquer - iterate and rule - chaos is No random!
|
|
|
David Makin
|
|
« Reply #1 on: December 11, 2009, 02:49:39 AM » |
|
It's only WIP but blb is using my formula from here: Edit: Please see here: http://www.fractalforums.com/mandelbulb-implementation/update-to-the-mmf-wip-formula-for-ultra-fractal/I should be releasing an updated version soon, though it will be a new formula in the same ufm file rather than an update to the existing formula because I've found loads of optimisations/improvements and adding them to the current version in a backwardly compatible manner would be too messy Note that there are still one or two things that don't work properly and some hangovers from the fact that it was started from my mmf4.ufm:3D IFS formula (such as colour by genetics). Note that for best results set the max iter to a high value (e.g. 400) and to get more detail reduce the "solid threshold" value.
|
|
« Last Edit: January 19, 2010, 04:26:37 AM by David Makin »
|
Logged
|
|
|
|
bib
|
|
« Reply #2 on: December 11, 2009, 11:17:19 AM » |
|
Note that for best results set the max iter to a high value (e.g. 400) and to get more detail reduce the "solid threshold" value.
I've been using your formula quite extensively in the last weeks. I found that for global views (not zoomed), 8 to 12 iterations is enough, with an Accuracy of 2, and a Solid Threshold of 1E-2 to 1E-3. For zooms, I sometimes need to set 100 to 500 iterations : the more you zoom into "whipped cream" areas the more you need iterations. And a solid threshlod sometimes as low as 1E-9, but generally 1E-5 to 1E-7 is enough. To get rid of artifacts, I occasionnally set the accuracy to 3 or 4. When you publish you formula in the database, it means I won't be able to use my existing ufr's if I want to take advantage of the last optimisations? Cheers bib
|
|
|
Logged
|
Between order and disorder reigns a delicious moment. (Paul Valéry)
|
|
|
David Makin
|
|
« Reply #3 on: December 11, 2009, 01:55:03 PM » |
|
When you publish you formula in the database, it means I won't be able to use my existing ufr's if I want to take advantage of the last optimisations? bib
Essentially correct, though I don't know if I'll ever upload it to the UF database, as I've said before it's only a WIP towards creating a set of ray-tracing classes for UF. I will make the updated version available for download from my site though. However all the important parameters will remain the same so manually setting the new formula to the same as an existing UPR using the old one should not be too difficult (just a little tedious).
|
|
|
Logged
|
|
|
|
cKleinhuis
|
|
« Reply #4 on: December 11, 2009, 02:13:47 PM » |
|
how do i set up the coloring/lighting ?
|
|
|
Logged
|
---
divide and conquer - iterate and rule - chaos is No random!
|
|
|
David Makin
|
|
« Reply #5 on: December 11, 2009, 03:12:39 PM » |
|
how do i set up the coloring/lighting ?
The object is coloured as "inside". With the "Lighting+colouring" colouring methods then use mmf4.ucl:3D colouring direct as the inside colouring - the rest of the parameters are basically in the main formula rather than the colouring. If you wish to control the RGB channels in the lighting separately then you can do so by choosing the "advanced" lighting option - but be sure that this is set to match in the formula and the colouring. If using "lighting+colouring" in normal combined light mode then you should set UF's standard maxiter to 4 (or more), if using the advanced method (split RGB) then you need to set UF's maxiter to 8 or more.
|
|
|
Logged
|
|
|
|
David Makin
|
|
« Reply #6 on: December 11, 2009, 06:58:24 PM » |
|
It's only WIP but blb is using my formula from here: http://www.fractalgallery.co.uk/MMFwip3D.zipI should be releasing an updated version soon, though it will be a new formula in the same ufm file rather than an update to the existing formula because I've found loads of optimisations/improvements and adding them to the current version in a backwardly compatible manner would be too messy Note that there are still one or two things that don't work properly and some hangovers from the fact that it was started from my mmf4.ufm:3D IFS formula (such as colour by genetics). Note that for best results set the max iter to a high value (e.g. 400) and to get more detail reduce the "solid threshold" value. Here "Note that for best results set the max iter to a high value (e.g. 400) and to get more detail reduce the "solid threshold" value." I was referring to the max iterations parameter in the formula rather than UF's standard max iterations formula parameter - UF's standard formula max iter should be set to either 1 (for plain lighting) or 4 (for lighting+colouring) or 8 (for lighting+colouring with advanced lighting enabled). Also if colouring by just "Lighting" alone then you should use mmf.ucl:MMF 3D colouring rather than the colouring formula from mmf4.ucl
|
|
« Last Edit: February 05, 2010, 09:42:45 PM by David Makin »
|
Logged
|
|
|
|
fractalrebel
|
|
« Reply #7 on: December 11, 2009, 09:27:00 PM » |
|
Actually, the iteration depth is the same as for ANY standard Ultrafractal ufm, and can be set to anything from 0 to very, very large. Look on the Formula tab. There is an entry box for maximum iterations. You set the value there. This is assuming you are using the correct formula, which is found in reb5.ufm, and is titled "3D Fractal Raytrace". For best results this formula should always be used with the the coloring formula "3D Fractal Coloring Direct", which is in reb5.ucl. I have not updated my current version of 3D Fractal Raytrace to the UF database in a while. There are a number of changes that I am still beta testing. I will post a message to FractalForums when I am finished beta testing.
|
|
|
Logged
|
|
|
|
fractalrebel
|
|
« Reply #8 on: December 11, 2009, 09:30:12 PM » |
|
After I update my formulas to the UF database I will post some sample Ultrafractal uprs here as demos.
|
|
|
Logged
|
|
|
|
fractalrebel
|
|
« Reply #9 on: December 11, 2009, 11:42:12 PM » |
|
The UF database has been updated with my most recent version of 3D Fractal Raytrace. I will be posting a short tutorial and some examples soon. You will need Ultrafractal 5 to access the formulas.
|
|
« Last Edit: December 11, 2009, 11:44:15 PM by fractalrebel »
|
Logged
|
|
|
|
cKleinhuis
|
|
« Reply #10 on: December 12, 2009, 05:06:46 AM » |
|
@fractalrebel
you can add them as demos to the ultrafractalwiki database
or, just publish the uf files, i will include them on the wiki page
|
|
|
Logged
|
---
divide and conquer - iterate and rule - chaos is No random!
|
|
|
fractalrebel
|
|
« Reply #11 on: December 12, 2009, 05:58:39 PM » |
|
@fractalrebel
you can add them as demos to the ultrafractalwiki database
or, just publish the uf files, i will include them on the wiki page
Trifox, you have the wrong link in ultrafractalwiki for my Ultrafractal tutorials. The link should be: http://www.hiddendimension.com/Tutorials/Tutorials_Main.html I will add a full tutorial to my website. The main code is in reb.ulb, which changes frequently, as it is my main object library file. It is also large (currently about 2 Meg). Objects from common.ulb and dmj.ulb is also used. Consequently, I don't think putting the files on the wiki page is a good idea. They are all in the Ultrafractal database, and anyone who owns Ultrafractal has access to them.
|
|
« Last Edit: December 12, 2009, 06:09:55 PM by fractalrebel »
|
Logged
|
|
|
|
David Makin
|
|
« Reply #12 on: December 12, 2009, 08:51:33 PM » |
|
@fractalrebel
you can add them as demos to the ultrafractalwiki database
or, just publish the uf files, i will include them on the wiki page
Trifox, you have the wrong link in ultrafractalwiki for my Ultrafractal tutorials. The link should be: http://www.hiddendimension.com/Tutorials/Tutorials_Main.html I will add a full tutorial to my website. The main code is in reb.ulb, which changes frequently, as it is my main object library file. It is also large (currently about 2 Meg). Objects from common.ulb and dmj.ulb is also used. Consequently, I don't think putting the files on the wiki page is a good idea. They are all in the Ultrafractal database, and anyone who owns Ultrafractal has access to them. Hi Ron, I think Trifox meant example parameter files
|
|
|
Logged
|
|
|
|
cKleinhuis
|
|
« Reply #13 on: December 13, 2009, 12:20:09 AM » |
|
@rebel yes, i meant the parameter files, some example copy&paste content for the wiki as a starter for beginners i will correct the link to your page, you have a direct link for the page with the wrong link ?!
|
|
|
Logged
|
---
divide and conquer - iterate and rule - chaos is No random!
|
|
|
fractalrebel
|
|
« Reply #14 on: December 13, 2009, 01:31:55 AM » |
|
@rebel yes, i meant the parameter files, some example copy&paste content for the wiki as a starter for beginners i will correct the link to your page, you have a direct link for the page with the wrong link ?! FractalWiki already has a direct link to my Ultrafractal Tutorials, but the link is wrong. It should be: http://www.hiddendimension.com/Tutorials/Tutorials_Main.html It is in the section on tutorials: Online tutorials Fractal Marbles Ron Barnett's Fractal Formula tutorials - this has the wrong link Ron Barnett's UF5 Challenges Janet Parke's tutorials Zoorekas tutorials My tutorials already have parameter files, as will the new one for 3D Fractal Raytrace. Do you still want a separate set of parameters?
|
|
« Last Edit: December 13, 2009, 01:38:08 AM by fractalrebel »
|
Logged
|
|
|
|
|