Title: Mandelbulber 0.90 - with Hybrid Formula Post by: Buddhi on August 02, 2010, 10:16:30 PM Mandelbulber 0.90 - for Linux x86, Linux x64 and also for Win32
Open source program (GNU GPL) for rendering photo-realistic 3D fractals Free download: http://sourceforge.net/projects/mandelbulber/ New features in version 0.90: - added hybrid formula - all formulas can be used in sequence - please try included example parameter files - Menger Sponge formula was changed to IFS Kaleidoscopic version (a little slower but compatible with other formulas) - all colour palette is stored in settings file (not only random seed) - colour palette can be grabbed from selected image - added "Absolute distance mode" option in 3D Navigator - actual distance between camera and viewpoint in displayed in 3D Navigator - program remembers last selected files in file choosers - on title bar is displayed name of last saved/loaded settings file - added logfile - all important program step are recorded in logfile - fixed bug: fixed problem with compiling program on Crux Linux - source files: fractal formulas was cleaned up Program should run also on computers with older CPUs. I have changed a little hardware profile in GCC. I know that version 0.85 worked not so good on AMD CPUs and on old Intels. Now should be better. Before first run please use install script My image gallery with images created using Mandelbulber: http://krzysztofmarczak.deviantart.com/gallery/ Title: Re: Mandelbulber 0.90 - with Hybrid Formula Post by: blob on August 02, 2010, 11:13:02 PM 0.85 was crashing on rendering most formulas on my Athlon 64. All good again with this one. ;D
Cheers. Title: Re: Mandelbulber 0.90 - with Hybrid Formula Post by: Rathinagiri on August 03, 2010, 05:59:23 AM Thanks for the new version!
Title: Re: Mandelbulber 0.90 - with Hybrid Formula Post by: Loadus on August 04, 2010, 01:46:18 AM This is now the best software on the internet. Good job!
Doing test renders right now ... Title: Re: Mandelbulber 0.90 - with Hybrid Formula Post by: cKleinhuis on August 04, 2010, 11:09:54 AM amazing!
but i have one question: how does saving of parameters work ? i mean how do i set the fractal parameters for a certain fractal in the alternation loop ? Title: Re: Mandelbulber 0.90 - with Hybrid Formula Post by: Buddhi on August 04, 2010, 12:20:52 PM amazing! but i have one question: how does saving of parameters work ? i mean how do i set the fractal parameters for a certain fractal in the alternation loop ? I'm not sure for what are you asking. If you want to render several fractals in queue you can create some script (.bat, bash, python) and run program with name of the settings file as a command argument. If you want to create series of parameters file, it should be also not so difficult. Parameters are stored in text files with very simple structure: Code: parameter_name value; Title: Re: Mandelbulber 0.90 - with Hybrid Formula Post by: Direct2Brain on August 14, 2010, 10:38:37 PM I have tested this version on my notebook with windows 7 premium and seems stable but on xp crash during the rendering. :sad1:
Title: Re: Mandelbulber 0.90 - with Hybrid Formula Post by: XICO2KX on August 26, 2010, 08:30:56 PM Realy awesome work, Buddhi! ;D
Just a quick question! :tongue1: Is the engine polygon-based or voxel-based? ;) Title: Re: Mandelbulber 0.90 - with Hybrid Formula Post by: Buddhi on August 26, 2010, 08:47:11 PM Is the engine polygon-based or voxel-based? ;) Neither of both :) Every pixel of surface is calculated directly from fractal formula. There is no any representation of object in memory like polygons or voxels. It based on "distance fields" technique. The surface of the object is calculated "on the fly" and most of shaders are based of analysis of estimated distance to the fractal surface. Title: Re: Mandelbulber 0.90 - with Hybrid Formula Post by: XICO2KX on August 26, 2010, 09:30:53 PM Oh, ok! I understand! Thanks! ;) That means that each new frame is independent from the previous ones, and that no information is re-used from frame to frame, right? If you could get a way to detect the pixels whose distance was the same that in the previous frame, they would only have to be (possibly) shifted, and you only needed to compute the pixels with a new distance! :tongue1: If that is feasible, I think it would improve your rendering speed a lot! ;) |