Fractal Domains

Earlier this year I replaced my old computer, a 2006 Mac Pro, with the new Mac Pro model. That was probably the longest I ever went before replacing a computer.

I ordered it on the first day the new model became available: December 19, 2013. However, since I ordered late in the day. I didn’t actually receive the computer until February 26, 2014. That’s how large the backlog was.

I didn’t post any pictures until now because that is how long it took me to get the initiative to straighten my desk sufficiently to be presentable enough for a photo.

Note that I purchased the desk last year in anticipation of getting a new system. I previously had a set of three 23″ LG monitors which I replaced this year with three 27″ HP monitors.

The Mac Pro is a six core model with 32GB of memory and a 1TB solid state drive.

The first photo was take with a wide rectilinear lens. The other two are crops of photos taken with a fish eye lens.

You can click on the photos below to look at a full size version.

Well, the ability to read the old legacy files was added over the course of the last few months and seems to be working pretty well. A lot of the parameter editing is in there, and now I have the ability to save files and export images. As an example you can see the latest entry in the gallery, Weave.

All that remains is the color map editor and the formula entry editor and the new program will be about up to speed with the old Fractal Domains (except with a much nicer interface). The next step will be to add new features and fractal types.

Here’s a screenshot of the program in its current state.

Update: Our project was funded! Thanks to all those who participated.

I’d like to take this opportunity to talk about a subject that is somewhat off topic for the Fractal Domains site. (However, I will have a comment that will bring it back to relevancy at the end of this post!)

Last year my wife and I, along with some other partners, embarked on a project to develop the prototype of an educational game and to test it with a group of subjects. The game has the dual goal of teaching mathematics and teaching cultural history to Native American students at the fourth, fifth and sixth grade levels.

This year we are embarking on the second phase of development of this game. Part of that is a Kickstarter program to help raise funding for this phase of development. I invite you, if interested, to go to our Kickstarter page and consider becoming a backer.

Of course, the entire area of educational games is pretty large. Instead of going into detail here about what makes this particular game unique, I will refer you to our web site 7 Generation Games which contains the pertinent details.

Also, I invite you to read my wife’s latest blog post on the subject of the start-up. She is a much more interesting writer than I am!

Last year, when I retired, I was looking forward to having plenty of time to work on Fractal Domains, which had been in a state of neglect for a few years. However, when the opportunity to work on this game came up, Fractal Domains once again was pushed to the margins.

I did learn a lot in the process of working on this game. My professional career was writing software for engineering purposes (mostly embedded and simulation software using Fortran, Ada, C and C++). The game required me to get up to speed on a whole panoply of languages and technologies: C# and Javascript (for game scripting), 3D modeling, HTML5, interfacing with MySQL databases, and more. It was a great opportunity, and I am looking forwards to extending my experience and abilities further this year.

However, I did find some time in the past few months to make some progress on the successor to Fractal Domains (see this post for an update). I hope to find time this year to publish what I hope will be the first of a long succession of Fractal Domains enhancements. I daresay that if this Kickstarter campaign is successful, we may have some money to hire people more experienced that I in some of the aforementioned technologies, which may help us complete some of our goals more quickly — and may free up some of my time to do more work on Fractal Domains!

It’s been over a year since my last post. It’s been a big year in my life. On the plus side for Fractal Domains development, I retired from my day job at the end of 2011. On the minus side (for Fractal Domains development), my wife quickly came up with a project that required some programming skills and was intriguing and allowed me to learn some new things (and also paid money). That, unfortunately, brought the development of Fractal Domains X to a virtual halt for most of 2012.

However, since November I’ve had an opportunity to put a lot more time into the plan I described in my last post. I’ll go into it more in some future posts, but a brief summary would be: I now have a Cocoa application that can read all of the Fractal Domains files I have on hand and can display them accurately.

A lot more has to be done on the interface. I have a parameter editor that is pretty much a clone of the old one; I have yet to implement an interface to enter polynomials for rational fractals. and I haven’t implemented a color map editor. In both cases, I don’t want to clone what I did before, but instead introduce interfaces that will be an advance on the previous versions.

When I get these interfaces done I will have an alpha version available so that I can find out if there are any flaws in the ability to read legacy files. I would like to fix all such bugs before I proceed further.

Here’s a screen shot:

I’d like to discuss the current state of Fractal Domains development.

[I will try to keep this non-technical, but I will have occasion to mention the two ways Apple provides to write software for OS X. Carbon is the name of the interface a programmer would use that is compatible with the old (pre-OS X) operating system. Cocoa is the name of the “new” interface introduced with OS X. It is the preferred interface for Mac software development.]

For several years now, I have had a “Future Plans” page on this web site. This page contained a wish list of features for a version of Fractal Domains that was basically planned to be a complete rewrite, which a new architecture and the ability to support a much wider variety of coloring schemes and many more fractal types.

Over the years, I actually made a start on this project more then once, on two separate occasions generating quite a bit of code. In each case a lack of time combined with an overambitious design (I’m sure many software developers know what I’m talking about) eventually led to abandonment of these projects.

In the meantime, I patched up the Carbon code base of the original Fractal Domains several times, fixing bugs and adding features in the process (Intel-native code, multiprocessor support).

Let’s back up a bit and take a look at the entire Fractal Domains timeline. Fractal Domains’ pedigree extends to well before Mac OS X was available. Originally the program’s name was FracPPC, and it was written in order to take advantage of the number crunching abilities of the PowerPC processor. The major fractal applications for the Mac at that point were all 68K programs and these programs were not being upgraded to PowerPC native.

Permit me to bore you with an abbreviated timeline of Fractal Domains development:

• March 1994 — first PowerPC-based Macs become available (I bought a 6100 the first day they became available.) Came with OS 7 installed.
• May 17, 1994 — FracPPC 1.0b1 is released. Only generated Mandelbrot zooms and you couldn’t even save as image.
• September 16, 1996 — FracPPC 1.1. You can edit color maps, save images.
• November 1, 1997 — Fractal Domains 1.2. Previous versions were free. This version is shareware and the name is changed to Fractal Domains. At this point most of the orbit trap and fractal formula features are there.
• July 3, 2000 — Fractal Domains 1.3.7. This would be the last version produced for OS 9.
• March 24, 2001 — Apple introduces OS X 10.0 (“Cheetah”).
• September 14, 2001 — Fractal Domains 2.0a1. Alpha version of the “Carbonized” Fractal Domains. Runs native on OS X.
• February 7, 2004 — Fractal Domains 2.0b5. Fractal Domains had been in beta for several years at this point with some feature additions and bug fixes. This is noteworthy for being the last version that runs on OS 9.
• December 31, 2006. — Fractal Domains 2.0. Three major noteworthy aspects:
• First Intel-native version (actually a Universal application, runs on PowerPC and Intel)
• First multi-threaded version (takes advantage of multiple processors)
• No longer runs on OS 9 at all.
• September 4, 2010 — Fractal Domains 2.0.11. The last version of Fractal Domains to be released as of this data.

Oh yes — my marriage and the birth of my daughter occurred somewhere in there.

I don’t know exactly when I first put up the “Future Plans” page, but it was definitely before the 2.0 final release on New Year’s Eve 2006, since some of the features incorporated in that release were in the original “Future Plans.” Possibly the page dated back to when OS X was released in 2001.

From the initial introduction of OS X, I have wanted to do a re-implementation of Fractal Domains in Cocoa. However, I succumbed to the temptation to completely redesign the architecture of the program, and that in turn led to a long delay in actually achieving the rewrite.

Meanwhile, time was marching on and I was obliged to make bug fixes to the current code base. The advent of Intel processors and multi-core systems even prompted me to make some major changes/additions to the program.

At the same time, it was becoming increasingly difficult to make changes to the Fractal Domains code. A large part of the problem was that the program had originally been written with a framework called PowerPlant. This framework, a redoubtable product in its day, was part of the set of developer tools sold by Metrowerks in the PowerPC days. When Apple started essentially giving away their developer tools for OS X, Metrowerks eventually gave up on trying to compete with “free,” and PowerPlant was orphaned.

I was not the only one in this boat (some major vendors of Mac software used this framework at one time). As OS X has developed it has become more and more impractical to make changes to the code in its current state.

Furthermore, the time is approaching when Carbon itself will no longer be supported by Apple. It is still supported as of OS X Lion, but the handwriting has been on the wall for years. My personal guess is that Carbon support will finally be dropped in the next major version of OS X (which if past precedent is any guide should be in about two years.) [Read more…]