Quick Links
External links, these open in a new tab (Firefox), or window (IE)
RealiMation before November 1997 Datapath
13th December 2008
All of this information is necessarily second hand as I wasn't there - I joined the company in November 1997. I learnt most of this from general chat and discussions as it was either knowledge I required to be able to fix a bug or enhance a system. However, I did have to check some of the facts with Mike Pridmore, so thanks Mike.
The software that was to become RealiMation was originally developed to promote Datapath's 3D graphics cards. Work started in November 1993 by Greg Whitfield and Mike Pridmore, with Joseph Steel joining in January 1994. At that time it was called RealiTools hence the RT prefix for the library functions. The product was launched in 1995 as version 4 (for marketing and psychological reasons).
Version 4.1 followed in November/December 1996 and version 4.2 in September/October 1997. Each version added more functionality to the libraries and the STE, or Space Time Editor, as it was called then. This was an application, built on top of the libraries, used by customers to create their virtual worlds. The key extensions were the import filters and the various versions of RealiStorm. These enabled models from other systems to be imported into the STE. The import filters read other file formats such as 3DS, DXF and VRML (1 and 2), while each version of RealiStorm was an extension to a particular CAD or modeling package such as AutoCAD, 3D Studio MAX or MicroStation.
The air traffic control simulator (Reflectone) was developed during this period. This was a multi channel system in fact the first one that used depth cued photos of real airports for view backgrounds. Generating these images was a labour intensive activity as elements in the image such as hangars, trees, hills etc. had to be traced around and assigned a depth value equivalent to the distance from the camera. While it's not a problem in this case, as the control tower at an airport is static (well one would hope so!), it does mean that the camera used by the simulation has to be completely stationary. It was important that the image was that of a real airport the training of the air traffic control staff depended on it. This was an upgrade of an existing system which used Datapath graphics cards and software from a company called SDI. The system as originally installed had three screens giving a field of view of 180°, but later expanded to four or even five screens.
The SWIG trainer, again using depth cued photos for view backgrounds, was developed for Jasmin a defence contractor based in Bulwell, Nottingham. This was an anti-tank missile training system that went into service with the UK military. I don't want to say much more than this as I don't know whether it's still covered by the Official Secrets Act or not. In fact I've probably said too much already :)
One of the big selling points for RealiMation for industrial and defence simulation was the fact that it was graphics driver independent and could support many of the specialist 3D graphics cards that were available at that time. This wasn't just the ones developed by Datapath, but included 3Dfx cards and others. This, however, was seen as a disadvantage by the games industry and was, perhaps, one of the reasons that RealiMation never supplied the actual game render engine, though the tools were sold to some games developers (Inner Workings and Codemasters for example), and eventually the company to Criterion Software.
There was even a version that ran on the DEC-Alpha processor version of Windows NT.
Another reason for not being able to sell into the game industry was the Not Invented Here syndrome. This was especially true where developers saw their job creating cool graphics effects or as a way of improving their Direct3D or OpenGL skills as opposed to developing a game with excellent game play and reliability. This view was so prevalent that when we wrote a series of technical papers on RealiMation the first addressed this very issue.
Another selling point was the continued backwards compatibility built into the system. Each new version of the application could read databases generated by any previous version. This was very important to the industrial user. They didn't want to have to re-export their models and regenerate their simulations with every release of RealiMation.
This period saw the start of a long association with Lanner an industrial simulation firm based in the West Midlands. They used RealiView (the free viewer) as the basis for a viewer for their simulations. They were a steady source of income for the product over the years.
The tools were also sold (or should that be given?) to Universities in the UK and overseas.
People:
- Brett Butcher (director)
- Greg Whitfield (team lead)
- Mike Pridmore (developer)
- Larry Binks (developer)
- Simon Brighty (graphic artist)
- Phil Sturdy (developer)
- Mark Ferguson (developer, to1997. I was his replacement)
- Joseph Steel (developer 1994 1997, he left sometime before November)
- Jo Rice (marketing 1994 1999)
- Greg Rice (sales 1996 1999)
Jo and Greg Rice stayed with Datapath when RealiMation went its own way (but more of that in future posts).
Awards:
It was during this initial period that the software won most awards. These included:
- Best Byte
- Ziff Davis Best Award 1997
- PC Pro Recommended
- PC Format Award
- CD Action Award
- PC Anwender Award
Next time: November 1997 to December 1999, from when I joined the company to when we went it alone.
(mailto: realimation at forman dot free-online dot co dot uk)