- Fractdev - mailman.xmission.com

RE: [Fractdev] Request for reflections on a .Net port
by Burnett, David (David) ** CTR ** 21 Jun '04

21 Jun '04

I find that my ability varies a lot, especially early in the morning after a late night ..... Of course, the advantages in porting fractint are that you have a working model to compare with, which is enormously useful (not to mention confusing). What are you up to in Alaska, holidaying, or is it a midlife career change ;-) David -----Original Message----- From: Timothy Wegner Heh heh heh heh!!! >From one who knows how much time programming fractint takes ... The only caveat is programmer ability does vary enormously - my 2000 hour estimate assumed very good skills but not world class genius. I would say that the original fractint has more like 4000 hours. Writing this from Park Service computer in Wrangell mountains, Alaska ... Tim "Burnett, David (David) ** CTR **" <dburnett(a)avaya.com> wrote: that should have read "more than 2000 hours" .... but probably on reflection, it wouldn't be that much yet, 2000 is a lot of time... David

1 0

RE: [Fractdev] Request for reflections on a .Net port
by Burnett, David (David) ** CTR ** 21 Jun '04

21 Jun '04

that should have read "more than 2000 hours" .... but probably on reflection, it wouldn't be that much yet, 2000 is a lot of time... David

2 1

RE: [Fractdev] Request for reflections on a .Net port
by Burnett, David (David) ** CTR ** 20 Jun '04

20 Jun '04

Which probably comes back to Rana's "writing owns own fractal app is an obligatory action for many programmers" Seriously though, any of us who are writing ports of fractint (or just their own fractal generators) could share analysis & design ideas and get some effort reuse and idea checking out of it. It certainly does represent a large effort, and there has to be a good reason to write yet another fractal app (YAFA) when there are already so many good apps out there. Or is the answer "because it's there" ;-) David -----Original Message----- From: Rich All I have to say is that if you are serious about putting in the amount of necessary work (I think Tim is underestimating it, frankly), then do what you think is best and other people's ideas of what you should do be damned. -- "The Direct3D Graphics Pipeline"-- code samples, sample chapter, FAQ: <http://www.xmission.com/~legalize/book/> Pilgrimage: Utah's annual demoparty <http://pilgrimage.scene.org>

1 0

Request for reflections on a .Net port
by Rana Ian 20 Jun '04

20 Jun '04

Hi, I'm looking at how to bring forth an up-to-date, open source fractal generating application into creation. The vision is to have a more robust Fractint-like application that offers a modern user interface, GPU (video card) calculations, and a more robust feature set. Fractint users have offered many wonderful suggestions on Robin Bussell's wish list that I would like to see realized. A survey of the field of fractal applications shows ~80 fractal generation applications with varying feature sets. Only a handful of the current fractal generating applications seem to have on-going maintenance and development. Fractint seems to have the most features of fractal generating applications at the moment. One of the developers of the fractal apps was joking that writing owns own fractal app is an obligatory action for many programmers. The existence of so many fractal apps offers some evidence to that truth. I think one of the strengths Fractint has is that instead of ~80 developers writing their own application ~80 developers contributed to an open source project which became greater than any application made by one or two developers. I abandoned the idea of writing my own application. Re-writing the wheel is not really appealing. What I want is a richly featured modern fractal generating app. What I want is basically Fractint with up-to-date features. So, how will this up-to-date fractal generating application manifest? .Net, C#, and DirectX offer a modern user interface, real-time fractal calculation using GPU pixel shaders, and up-to-date technologies to develop new features going forward. I propose to rewrite Fractint in c# for the .Net platform. Windows would be supported. Unix would be supported to the extent .Net is implemented on Unix systems by the Mono project. As we know, there are many interesting things that can be done. Some ideas Fractint users have suggested to the wish list: Video Card acceleration, Modern user interface Distributed Calculation across CPUs Key Frame sequencing for deep zoom animation Animation response to audio patterns Fractal Audio Generation Photoshop Plug-ins Winamp Plug-Ins More Coloring Options Localization Video Animation Screensavers Internal formula editor Some other features I would like to see are: Peer-To-Peer File swapping and Gallery Item Rating System In-app bug reporting In-app feature requesting Variable Canvas sizes KPT-like mutation lab (Randomness suggested form wish list) Gradinet Designer (like KPT) XML File Formats File compatability with competing fractal applications Inter-operability with Adobe technologies - Palettes - Layers Multiple Image export formats A rewrite would go a long way to enabling an integrated engine to address all of the would-b-nice requests to go forward. The intent is to build upon what has already been done. The intent is to maintain open source ideas. The intent is to continue developing the best fractal exploration application. What do you think? Is there any support, criticism, or suggestions you would like to offer towards a Fractint.Net port? Thanks, Rana Ian Tucson, Arizona, USA Fractint Wish List http://web.ukonline.co.uk/members/robin.b2/olig/fracwish.htm List of Fractal Applications http://home.att.net/~Paul.N.Lee/Fractal_Software.html GPU Shader Development http://developer.nvidia.com/object/FX_Composer_Shaders.html .Net Development http://msdn.microsoft.com/netframework/ The Mono Project (.Net on Unix) http://www.go-mono.com/

4 3

RE: [Fractdev] Request for reflections on a .Net port
by Burnett, David (David) ** CTR ** 20 Jun '04

20 Jun '04

Being completely unbiased ;-) I would go for Java, and reap the platform independance that goes with it. Nothing is to be gained by C# (the evil empires java clone), except platform lockin (and you still end up with an interpreted language). C++ would be faster, but very much harder to get working on multiple platforms. I think the best option is to write a pure java version and add platform specific C++ (or even assembler with C++ shell) native calls for specific OSs to speedup the actual calcs. Best of both worlds then, all the advantages of java maintainability and ease of development, with speed when needed. I have spent more than 200 hours putting together a java fractal app (FractalCanvas - see qlt.net.au/canvas - but not for a few days). I didn't intend on making a fractint clone, but have come closer than I thought I would. Because I decided to use a collection of fractint pars to test my formula parser, I ended up implementing a lot of fractint functionality in the same mode. I currently get about a 60% hit rate for my test set of 1500 pars/images, that is about 60% of them generate images that look the same as fractint generates, about another 20% get similar results, and the rest do not do much at all. The reasons for a lot of that is the post-processing options (such as logmap, continuous potential, periodicity etc) I have implemented are not complete, or simply do not work correctly yet. Added to which I haven't implemented any of the backwards compatibility features from fractint at all (yet). I will probably ask some precific questions about these algorithms (when I have time to figure out what it is I don't understand). I will put a version on the net soon, the current net version hasn't been updated for a while and doesn't have the formula parsing in it. I just have to write a UI for editing a formula (and a few hundred more things no doubt ;-)..... David -----Original Message----- From: fractdev-bounces+dburnett=avaya.com(a)mailman.xmission.com [mailto:fractdev-bounces+dburnett=avaya.com@mailman.xmission.com]On Behalf Of Tim Wegner Sent: Sunday, 20 June 2004 1:05 PM To: Fractint developer's list Subject: Re: [Fractdev] Request for reflections on a .Net port Greg wrote: > I think it's great that you want to modernize > Fractint, and I would like to definitely help with > this, but .NET is the devil. Microsoft and open-source > really shouldn't go in the same sentance. > Think Java or C++. C++ would be very nice because it > would be easier to port the existing code, as well as > performance considerations. You may be right, but discussions about platform mean nothing without skill, time, and commitment. We're talking 2000 to 4000 hours. Tim _______________________________________________ Fractdev mailing list Fractdev(a)mailman.xmission.com http://mailman.xmission.com/cgi-bin/mailman/listinfo/fractdev

1 0

Xfractint subdirectory display in menus
by Jonathan Osuch 14 Jun '04

14 Jun '04

Tim, I've made some progress getting subdirectories to display in Xfractint. It appears to work for most items, but not color maps. For some reason a slash is getting added to the color map name (Xfractint thinks it's a directory). So, color maps won't load. Jonathan

3 6

Question on acos function in Fractint
by Edwin 06 May '04

06 May '04

Hi, I'm comparing the results my fractal program produces against Fractint's. In general I get the same results, but I'm getting dramatically different images for some fractals involving the acos function. Using the formula: mandelfn { z = 0: z = fn1(z*z) + pixel |z| < 4.0 } With fn1 = acos and default parameters, Fractint produces : http://www.speakeasy.org/~edyoung/fract001.gif And I get: http://www.speakeasy.org/~edyoung/acos_image.png I think this stems from different functions we use to compute acos: Fractint uses rz=Arccos(z)=-i*Log{z+sqrt(z*z-1)} Whereas my code uses: acos(z) = pi/2 - asin(z) According to MathWorld, ( see http://functions.wolfram.com/ElementaryFunctions/ArcCos/27/02/01/) the formula Fractint uses is only valid for certain domains, so I'm suspicious that Fractint's producing the wrong output. Any comments? Regards, -- Edwin

2 3

Reading deep zooms from PARs
by Tim Wegner 02 May '04

02 May '04

A user sent me an example of a deepzoom PAR that won't reload into Fractint. I looked into it, and it wasn't hard to find the problem. I have a fix of sorts which I'll turn into a patch as soon as I get it cleaned up. Only cmdfiles.c is involved. I expect this is a known problem. Tim

2 9

CVS, again
by Jonathan Osuch 08 Mar '04

08 Mar '04

Tim, I've got a CVS repository set up on a Syquest disk. I'm in the process of modifying the source such that Fractint and Xfractint can reside in the same repository. I should probably put together a patch, but I found out today that my last day here at Clinton will be March 19th. Subject to change, of course. But, that means I need to move a bunch of stuff back to Iowa and also find a job. This was supposed to happen in the middle of July, so I wasn't totally unprepared for this. A patch isn't out of the question, I'm just a bit stressed. Jonathan

1 0

Xfractint and reading subdirectories
by Jonathan Osuch 03 Mar '04

03 Mar '04

Tim, Has Xfractint always had a problem with reading subdirectories, or is this a recent occurrance? For example, if you want to change PAR or FRM files, no subdirectories are displayed. Jonathan

2 1