Logo by mauxuam - Contribute your own Logo!
News: Visit the official fractalforums.com Youtube Channel
 
*
Welcome, Guest. Please login or register. July 25, 2017, 06:41:37 AM


Login with username, password and session length



Pages: [1]   Go Down
  Print  
Share this topic on DiggShare this topic on FacebookShare this topic on GoogleShare this topic on RedditShare this topic on StumbleUponShare this topic on Twitter
Author Topic: MandelBulb deep journey  (Read 156 times)
0 Members and 1 Guest are viewing this topic.
gurroa
Forums Newbie
*
Posts: 9


« on: November 11, 2016, 09:44:46 AM »

This is more like a technology preview video from my software FractaloScop, not an art showcase.
So I am posting in this category.

Zoom is 3626x and MandelBulb geometry has 100 iterations.

If we call the lowest computable resolution of 0.0000001 as a 1 centimeter, 1 than represents theoretical 100km, and the MandelBulb itself is now an object of 2.2 milion kilometers in a diameter (that is 1/68 AU or 7 light seconds).
If you pause the video at the closest distance or when flying through the structure border you can catch the pixelation - squared objects - that is the 1 cm.

Disapearing and waving geometry is when the structural thickness drops below half of a meter.
Which is current camera structural resolution. And from some angles the algorithm catch it and from others do not.

Game of shadows and light is an error or you can call it a showcase of the limits of global illumination eye trick.
But the same error is responsible for shadows that are making structures look like solid matter.
And I assume that those errors are shadows "projected" from structures insanly far above and behind the camera.
Need further rebalancing for deep zooming.

This is my first try to fly in this insane deepnest. Previously I was flying only around the zoom 900x. New HW and parameters tweaking.
My previous video at this forum was mixing two ways of zooming (density shifting and growing), that is why I called it 25*900 aka 22.500x zoom.

But this way it is still fast enough for live flying - 25+ fps at the full-hd resolution. Of course without any antialiasing, motion blur and half the horizont distance seen in the rendered video (on GTX760) - 100 iterations are really too heavy load.
Average render time for this video (2560x1440) with 4xAA, motion blur and full horizont was 0.8 frame per second.
Video was post-processed on youtube - rebalanced light.

I have also tried to switch all computation and eye tricks off.
And after the restart of my computer I won't be trying it again anytime soon.

<a href="https://www.youtube.com/v/6uS-yfpnI_k&rel=1&fs=1&hd=1" target="_blank">https://www.youtube.com/v/6uS-yfpnI_k&rel=1&fs=1&hd=1</a>
Logged
gurroa
Forums Newbie
*
Posts: 9


« Reply #1 on: November 11, 2016, 11:22:05 AM »

This render represents what I call a density shifting. This time I have tried only on X and negative Z axis.
MandelBulb is suddenly really eye catchy with different structures.

But this kind of false "zooming" was really not suitable for real-time rendering so I went a different way and I am now combining real growing with just a little bit of density shifting.

edit: image is a 6MB jpeg (37Mpix)
http://www.gurroa.com/Fraktal/MandelBulbView2Max.jpg
Logged
Pages: [1]   Go Down
  Print  
 
Jump to:  

Related Topics
Subject Started by Replies Views Last post
Mandelbulb deep deep dive Videos bib 14 1776 Last post January 08, 2010, 10:27:07 PM
by David Makin
A Journey Through Fantasy Movie Madman 0 649 Last post April 27, 2011, 10:21:52 PM
by Madman
A Strange Journey Film khlchau 0 730 Last post May 06, 2012, 10:58:26 AM
by khlchau
The Stationary Journey Film kameelian 0 668 Last post May 29, 2012, 01:42:41 AM
by kameelian
Journey to the Cathedral Movies Showcase (Rate My Movie) Don Whitaker 7 525 Last post March 28, 2013, 09:17:58 PM
by trumanbrown

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines

Valid XHTML 1.0! Valid CSS! Dilber MC Theme by HarzeM
Page created in 0.146 seconds with 28 queries. (Pretty URLs adds 0.009s, 2q)