Translations for our friends around the world.

Click on banner for Tornado website

Click.gif 

Author Topic: An Interview With Digital Integration Ltd (Part 2)  (Read 4524 times)

0 Members and 1 Guest are viewing this topic.

Offline Frankie

  • Moderator
  • *****
  • Posts: 1506
  • From Dogs of War to Pipes Of Peace
    • Frankie's Moodurian
An Interview With Digital Integration Ltd (Part 2)
« on: May 22, 2016, 12:33:32 PM »
AN INTERVIEW WITH DIGITAL INTEGRATION LTD. (PART II)
GameBytes Magazine (1994) Volume 18
by Andrew Stevens


As regular readers of Game Bytes will know, Digital Integration are the UK software house responsible for the TORNADO flight simulator (Game Bytes #13). The successful launch of this product was followed-up by the launch of a version for the Amiga, along with an add-in DESERT STORM scenario disk. Both were released just in time for the Christmas season. Since the folks at DI are always very happy to share a pint of beer, your intrepid interviewer decided to do a short follow-up to find out what happens at a Software house *after* a big release has been completed.

GB: The last time we talked you folk had just started work on the add-in disk. At that time it seemed you were still considering whether or not to go for a more ambitious product incorporating a campaign editor. What persuaded you to settle for the more modest goal of adding `just' a new campaign area with new graphics.

DI: Obviously, an important goal was to be able to get the product finished in good time for the Christmas season. The problem with a campaign editor was that the campaign engine has a lot of pretty subtle code in it that we did quite a way back in the original Tornado project. Basically, even Robin (who originally wrote it) wasn't at all sure what a lot of it did anymore so we were very reluctant to tinker with it. It would simply have been much too risky in terms of wasted man-power. Then there was also the work we needed to do to get the Amiga version completed on time. There was a time when just about everyone was working flat-out on that in an all-out effort to get it finished.

GB: Now that things are a little calmer, can we expect more Tornado add-ons?

DI: No that definitely won't happen. The Tornado code really isn't up to much further development. It is all 16-bit real-mode code and it would require a huge amount of effort to squeeze in extra stuff without breaking all the existing code. Given that the Tornado visuals are beginning to look a bit dated it just wouldn't be cost effective. Also, a good part of why we're all working in this area is the chance to do something a bit new and interesting. Tornado has got to the stage where we're pretty much fed up with messing around with it. It really has got to the point where I at just can't look at the stuff anymore. Since the beginning of this year we've all switched over to working on the new game.

GB: The last I heard, you were still considering a kind of dual launch - using the same graphics engine to develop two simulations in parallel. How far has development progressed?

DI: Well, you could say cooler heads in the company have prevailed as far as the parallel launches plan is concerned. The current plan is rather different. What we're aiming to do is bring out two games spaced by about a year. The initial game will still have the very high-quality visuals and so on but will be a lot less ambitious in terms of things like authentic campaign play, weapons and avionics modelling. A simulation rather than a simulator.

GB: Presumably modelling weapons and so forth takes a lot of manpower?

DI: Yes definitely, as we found out to our cost with Tornado. We were actually quite badly bitten by this - the 3D engine was ready almost 2 years before rest of the game! Those 2 years were basically spent sorting out the campaign system and the flight, weapons and avionics models. So, instead of being able to re-use the graphics engine for another project, it was already a good way to being out-moded when Tornado was released. That is not a mistake we want to repeat.

GB: Was it really that much of a problem.? On the Internet, at least, quite a number of people have said very favourable things about the Tornado graphics and how well they work for low-level flight.

DI: Yes, but that's really more due to some of the little visual tricks we played - shading sub-triangles of the ground, little clusters of trees - rather than the underlying graphics engine. You could do things like that *and* have a much better overall effect. Shading for smooth curves, texture mapping and so on. That is what we are aiming for in the new game.

GB: Have you settled on the type of aircraft that you want to simulate?

DI: The first game will almost certainly be a helicopter. There are several reasons for this. Firstly, we think no-one has really done a helo simulation with a really serious flight model, and it's something we'd quite like to do. We've already gathered some experience in developing the EH101 Merlin mini-simulator for the Royal Navy. Secondly, in a helicopter simulation you can do a first-class job without the endless weapons and campaign code you'd need to simulate a fixed wing aircraft. Helicopters simply have less different weapons and sensors and can be realistically simulated without a very complex mission planning system. The second game will probably be a fixed wing aircraft

GB: How far has work progressed? Has any code been cut yet?

DI: Well currently most of the work has been going into the new 32-bit protected mode 3D engine and graphics drivers. Currently we're working on the basis of being able to support 3 different display modes: 320x240, 640x480, and 800x600. The reason we're using the non-standard 320x240 mode rather than the usual 320x200 mode is that it has square pixels. We had endless troubles getting stuff like the cockpit dials and so on to work properly in Tornado because the pixels aren't square at 320x200. We think we can save a significant amount of time by avoiding that kind of difficulty.

GB: What about performance? Won't 800x600, at least, be rather too slow for a flight simulation.

DI: Well, the 800x600 mode may only be used by the front-end. One of the main design goals is to provide a clean interface to the graphics drivers so that they can be shared by both the simulation and the front-end. That was another lesson from Tornado: we ended up doing a lot of extra work because all the low-level graphics stuff - line drawing bit-block transfers and so on - all had to be done twice. Even so, we haven't by any means ruled out supporting 800x600 for simulation on fast machines with fast graphics cards. I've been doing quite a bit of experimentation as the graphics drivers develop and I'm a lot more optimistic about performance than a little while ago. Basically although the SVGA mode uses four times as much display memory it isn't turning out to be four times as slow. The kind of results I'm getting suggest a frame-rate for Tornado level graphics of about 65 frames per second at 320x240. That is with a mid- range 486 with a DRAM based localbus graphics card using the S3 chip-set. Moving to 640x480 slows things down to about 30 frames per second whilst going to 800x600 results in about 15 frames per second. If you used an ET4000W32 based video card you'd probably get things to go at least 10 per cent faster. Using high end cards with VRAM and so on would probably still a better performance at the higher resolutions.

We've also been working on how best to implement a proper `virtual cockpit' view where you can look around freely with bits of cockpit or aeroplane coming into view depending on the direction you're looking. In current games that offer this kind of thing you still have to switch to the ordinary forward view if you want to read cockpit instruments. We reckon we've figured out a way to go one better than this and have an `active' texture-mapped cockpit: one where you can still glance at the radar and dials and so on.

GB: This sounds very juicy but also quite ambitious. What kind of release date are you aiming for?

DI: We're hoping to complete the first game in around a year. The basic strategy is to work on designing objects and landscapes in parallel with the development of the software. Designing objects and maps is a lot of work - something you can easily underestimate. The exact release date will probably depend on whether we can hire some extra people to work on that side of game. The problem is that it is actually pretty hard to get the right kind of people. You need people who are adaptable and who have decent practical skills. Unfortunately a lot of the people you get from University just don't have those kinds of skills. For example, we had a guy apply who'd already done quite a lot of graphics work as a student. He showed us pictures of images he'd programmed - absolutely gorgeous. The problem was that they'd taken days to generate on a fancy workstation, and he didn't have a clue how to work at the low-level to get good images fast on modest hardware. To succeed in developing games you absolutely have a good idea how things work at the bit-twiddling level.

The same kind of thing applies in designing objects and maps and so on. You have to have an idea about what is going on under the hood and have to know how to get things done fast. You have to be able to work with pretty crude software that has been knocked up to do a specific job, and you have to understands the short-cuts that can use to speed up the design of an object.

GB: Presumably moving to 32-bit protected mode will speed up development somewhat?

DI: Yes. In Tornado the simulation engine is written almost entirely in assembly language. For the new game we're using the Watcom C compiler for the main 3D routines with only the low-level graphics engine and other really time-critical stuff done in assembler. We're planning on writing the user-interface and front-end in C++.

GB: Are there any plans to produce versions of the game(s) for platforms other than the PC? What about the new 32-bit consoles, for example?

DI: No, we haven't got any console distribution plans at present. The newer designs that are still under wraps might be suitable, technically. For example, we've heard some very interesting things about the console Sony are developing. This, apparently, has an extremely fast RISC CPU (Mips R4200) and some very useful graphics facilities. However, that system and the others like it are still very much under wraps and things could still change a lot. The machines that are available now, CD-I and the Amiga CD-32 for example, are simply too slow for what we have in mind. Even if the machines themselves were fast enough, you still have the problem of running the game directly off a CD-ROM. Live access on a CD-ROM is simply far too slow for the kind of thing we're interested in.

That said, we can't say that console versions of our games *won't* appear, but we probably wouldn't be the people responsible. To give you some idea: we sold the console rights to Tornado quite some while ago. When we asked if they wanted the source code or stuff like that they basically weren't interested. All they wanted was the name plus the sound-effects and the cockpit graphics!

GB: What about CD-ROM for the PC?

DI: Well , we are definitely intending to use CD-ROM as the primary distribution medium. However, that really will only be for distribution purposes. You almost certainly still have to install on to hard disc to get sensible performance.

This interview is Copyright (C) 1994 by Andrew Stevens for Game Bytes magazine. All rights reserved.
« Last Edit: May 22, 2016, 12:40:31 PM by Frankie »
funny
0
informative
0
Thanks
0
No reactions
No reactions
No reactions

After the Dogs of War are let slip, let us smoke the Pipes Of Peace.

Offline Asid

  • HAVOC
  • *
  • Posts: 26361
Re: An Interview With Digital Integration Ltd (Part 2)
« Reply #1 on: May 22, 2016, 03:54:52 PM »
Thanks for part 2

Very interesting.

Thanks for posting :)
funny
0
informative
0
Thanks
0
No reactions
No reactions
No reactions

I stand against Racism, Bigotry and Bullying

Tags:
     

    Flying the Tornado GR1 | John Nichol (Full July 2021 interview

    Started by Frankie

    Replies: 0
    Views: 1568
    Last post January 05, 2022, 03:16:31 PM
    by Frankie
    Actor practiced on "Tornado computer game at home" as part his TV role

    Started by Frankie

    Replies: 1
    Views: 4502
    Last post May 12, 2017, 08:21:03 AM
    by Lusik
    Italian review of Digital Integration's 1993 Tornado. Blast from the past.

    Started by Frankie

    Replies: 14
    Views: 15182
    Last post February 23, 2016, 02:02:47 PM
    by Frankie
    Recent articles on Digital Integration's Tornado

    Started by Frankie

    Replies: 2
    Views: 3137
    Last post March 08, 2019, 03:18:11 PM
    by Rinix
    In Memoriam of David Keith Marshall Co-Founder of Digital Integration

    Started by Frankie

    Replies: 3
    Views: 5252
    Last post February 01, 2019, 03:28:55 AM
    by Speedwagon