Firstly up until this point 2D graphics were my specialty. Secondly the dust had not settled on the standards of 3D. Thirdly no roadmap existed for what we were trying to do. Fourthly we were young guys burning the candle at both ends. I was running a web business (remember this was 1997 and the internet was not what it is now) during the day, paying for new equipment, the rent, the food and the entertainment. I remember that building fondly it was above an old laundromat - the guy who ran it would come yelling upstairs whenever my washing machine overflowed.....Anyway the first manifestations of some of these problems started to manifest once we were modeling the basic houses (AV Jennings style houses - essentially brick block houses with hip roofs). To fill a space up in 2D takes a similar amount of time each time you do it - there are x number of pixels to fill. In 3D, and real-time 3D at that, there is an almost infinite amount of screen space to fill up every time you move the 3D view you essentially have more space to fill up with objects. So on screen we had laughably simple houses that we could look at from all angles - we had very simple lo-resolution textures that would break and distort as we altered the view and we had very empty scenes compared to our 2D versions of houses. We were made constantly aware of the limitations of 4mb of memory in 3D space. As we worked we came up with a number of methods to save memory whilst adding the illusion of detail. We also used photographic techniques to split the scene up into foreground, midground and background to create a pleasing image and to immerse the viewer.The next problem was the user interface. A standard user interface would not suffice. A lot of software has what I would call QWERTY syndrome. Those of you who have read the story will know that the layout of letters on a keyboard are a thowback to the typewriter days and the layout was designed to stop you typing so fast the mechanism would jam. Of course keyboards on a computer have no such mechanical limitations - yet the artificial slow down mechanism is present. On most software packages the interface is so by the book (the Windows Design Document) which is all about consistency between software packages for the sake of user orientation rather than being able to use software efficiently. I wanted to get faster at using the software in a way that actually increased my modeling speed. If you have a look at a lot of software selecting options from menus and windows can actually take as much time as the task at hand. I wanted some form of context to the process and i wanted immediacy. So in that one defining moment the real crux of ViewBuild was born.1. Context Menus. Click the on an object would provide options for that object.2. Remove most of the interface from the screen (how cluttered is you software?) I wanted acres of screen real-estate for the task at hand. So we put all the menu options on a right button click menu (3D Studio MAX uses this mechanism now).3. Modes for control. View Mode to navigate. Draw Mode to edit. Move Mode to move things etc.4. Object focus navigation. Scroll bars do not make a lot of sense in a 3D scene. So we came up with the idea of setting a focus point on an object and using that as relative navigation point.5. Parent Child grouping. In real life you put a table on a floor. A cup on a saucer on the table. We wanted the same sort of real world logic.6. Make the navigatrion dynamic and give it a sense of "weight"7. This is perhaps the simplest idea but perhaps the main reason why our modeling works in a persepctive view. We removed the "Twist" capability that every other 3D software has for altering the view point. With this gone we were able to simplify our control mechanism.8. We reduced our navigation to planes. This allowed us to more easily translate real world mouse controls to 3D maths displayed by a 2D monitor.There were some other innovations to come later.8 Months later we had a tool that worked on Direct X and you could build certain styles of houses.Early in 1998 and very aware of the rent owing my friend and I went seperate ways - he wanted to earn a wage. So without a lead coder I bunkered down and started to seriously use the tool to try modeling some commercial projects. It worked. Over the next 9 months i literally became the biggest user of the tool - I think this may be one thing that many software writers do not have the inclination or the time to do - to actually use the tool and become a critic. I catalogued bugs, designed fixes to existing problems and most importantly started to work out a shorthand for the software. Over this period 3D cards started to become more powerful, cheaper and more standard.The corporate stage.....my time in a tie.
