Diary of a Game
Uridium 2 Part 5
Will Andrew Braybrook's latest be his greatest? The Amiga sequel to his C64 megablaster is now in its fifth month of development and getting better all the time. Every month in his exclusively-serialised diary, Andrew delivers a blow-by-blow account of how the potentially blockbusting game is coming along. THIS MONTH! The game reaches an important crossroads and there's personal drama to rival any soap opera...
Part Five — September/October 1992
Friday 18th September, 1992
The Uridium 2 program has now got so massive that in order to develop it further I will have to temporarily chop something out. First thing to go is the sound. Now I remember why I don't normally put that in until near the end of the project. That saves about 150K which keeps things to sensible sizes. Jason can still work on the sound with his sound editor and I can get the sound back by making up a 'live' version of the game if I need to. While I was out yesterday he created some new SFX for me, including two really strange effects that aren't really appropriate to the game but are so good that I want to put them in somewhere.
Got the fighter attack waves to be produced near the player wherever he is, as there's no point in going to all the trouble of producing a great wave of fighters when no-one is looking.
Monday 21st September, 1992
In order to squash things down still further to give me as much chip RAM as possible I've separated off the robot sprites as they're not needed most of the time. When inside the ship the robot sprites are swapped in as the enemy ships are swapped out. There's a bit of monkey business going on to keep the assembler happy about which graphics are which and it took most of the day just to get the game working again.
Tuesday 22nd September, 1992
I'm sure we fixed that once. The first dreadnought of the first fleet is now almost complete but for adding destructible bits and other features like bunkers and missile silos.
Spent the rest of the time rearranging the memory map in order to be able to load the game into any 1Mb Amiga, not just ones with 1Mb of chip RAM. This task is nearly complete now. With all the preparations being made for loading files from disk, it seemed appropriate to add in the routine that asks for the right disk should it discover the wrong one, or none at all, in the drive.
At this stage I've nearly worked out completely what data gets loaded in, what is always there but in a packed format, and what is always present unpacked. I want to avoid irritating delays where possible so it'll do just the one load from disk to pull in a whole fleet of dreadnoughts. It should also have room enough to cache the first one in memory all the time.
Wednesday 23rd September, 1992
Put in a feature to darken down the control panel where it says '1 UP' and '2 UP', so it only lights up when the player in question is in control. This also has to take into account the two-player mode and light both up. I hate games that remove the waiting player's score from the screen — I call that player the bench player. You need to know how both are doing all of the time for comparison, and because the only time you really can take your eyes off what you're doing is when the other player is "on".
Thursday 24th September, 1992
Went through the old C64 Uridium flight patterns to familiarise myself with what they do. My new code is much smarter than the C64's so a lot of the patterns are unnecessary. For example, I can vertically reflect a flight plan to get two ships mirroring each other's movement just by specifying upside-down mode. One thing I had missed out though was chaser mode, where an enemy ship tries to line up with the player before squirting Bonios of Doom at him. I want this to be a bit smoother so I'm working on an algorithm to line up behind the player more intelligently. This is not going well at the moment.
Friday 25th September, 1992
Cured the last long-term niggle which involved the homing missiles. They have a minimum turning circle and if fired very close to their target they can circle it until they time out. What they need is a contingency plan. First they have to recognise that they are in that situation. This can be identified by keeping track of consecutive moves at full turn. Any turn less than full means that they have successfully lined up with their target. After turning 270° on full lock they fly straight for about six moves before resuming targeting. By this time they are far enough away that they have room to turn and hit the target every time.
Tuesday 29th September, 1992
Did some end-game-screen designing. Obviously I'm not at liberty to tell you precisely what I've been doing, so I'll just say that I've come up with a new golden palette for the title screen, rather than plain yellow.
Wednesday 30th September, 1992
More meanies are the order of the day. Cargo ships that take off from runways, concealed laser towers that spring up and spray doom and disaster before hiding away again and such-like. I'm constructing meany algorithms in our Alien Manoeuvre Program which allows me to make things happen more quickly.
Thursday 1st October, 1992
Test-pilots Robert Orchard (the inventor of the name 'Uridium') and Richard Harvey came in to give the game the once-over. Within seconds they had the Manta flying backwards, something that hasn't occurred for months, and I can't reproduce the effect. Users! Bah!
Friday 2nd October, 1992
I want to get some shadows on the enemy ships. That's a bit of a hefty overhead but I've thought of a way of persuading the hardware sprite multiplexor to do the job. Multiplexor's a fancy name for a routine which sorts the objects to be displayed into vertical position sequence and re-uses the hardware sprites to show them, by the way, Had another juggle about with the robot speeds to make it feel more controllable. Just about anything that can he changed is defined as a variable so that I can change them easily and immediately see the results. It's getting tricky to tune the game now because I'm so used to it. That's why it's good to get outside help from time to time. A stranger to the game can be a lot less subjective about things.
Monday 5th October, 1992
It's one of those days where you start thinking "What if...?". What if I did the game in dual-playfield mode? How long will it take to find out? Never tried dual-playfield before. How difficult can it be? Only had a couple of hours at it but I got the bare bones of a system up and running. I copied all the game files to a new directory so I don't wreck any code, and started converting it.
Tuesday 6th October, 1992
Carried on with Operation Dual-Playfield. Made an astounding discovery — you don't have to have two playfields of three bitplanes each. You can have one of three bitplanes and one of two. This cuts down on the CPU welly that you lose as you're only displaying five bitplanes, not six. Since I'm already using five bitplanes in 32-colour mode, I won't lose any speed. Any loss I make on keeping two playfields running I make up for by only plotting three bitplanes. So now the question is, which version do I carry on with?
The actual gameplay will be largely unaffected by the choice but the effect on the game is rather drastic, to say the least. The appearance will be completely changed. Anybody who suggests running both methods is likely to be severely damaged as all the graphics would need to be stored in two different formats.
Wednesday 7th October, 1992
Graeme from Renegade paid us a visit so I showed him the two versions of the game. When he saw the dual-playfield version I think he was quite impressed as he uttered a number of unprintables! So, how can I remove some of the cons of this mode? Well, following on from the statement running both 32 colours and dual playfield in the same game, how about if all the titles and presentation is in 32 colours and the game is in dual-playfield? That way I keep my plasma and my lovely font.
It took an afternoon of machine crashes to do it, but it's done.
Thursday 8th October, 1992
More work getting the dual-playfield version running. It's now all working so all we have to do is decide which version to go with. The CPU use on both versions is about the same, so it doesn't make much difference to me. But can anyone draw sensible graphics in seven colours, of which one must be black?
Friday 9th October, 1992
Walking home last night I'd just about talked myself out of doing the dual-playfield version. How about saving that for Uridium 3? I hadn't seen a satisfactory palette to do the job. Two sets of three colours just doesn't give the nice shading and one set of six colours looks like the TV isn't working properly. Just to add a bit of excitement, Phillip the graphics artist has resigned, which should cut down on the arguments a bit, but it leaves a shortfall in the graphics department. Over to Mark Bentley, one of the new recruits with a last-ditch first task — find me a palette I can live with!
I now have a palette of near-grey shades with a red and an orange for explosions. The greys are neutral enough that they can be mixed with the red and orange, and there are enough similar colours for good shading effects. The backgrounds need massaging but all the sprites converted well with little touching up. One version has to go soon as I can't maintain both forever. It's like juggling three chainsaws at once.
Monday 12th October, 1992
Added a new bonus feature, a score-doubler. Needed a new icon for that so I set about drawing a new icon block to put all the bonuses in. The old ones didn't stand out too well.
Tuesday 13th October, 1992
Swaying towards dual-playfield mode. Only worked on that today. I've built in some variations in the display, like being able to swap the playfields over so that the three-colour one is on top. I can also free up the second playfield so it doesn't parallax at all. I now have the capability of running a giant meany, almost for free, although it only appears in three colours. How useful that will be remains to be seen. Also worked on an end-of-dreadnought sequence, where the Manta(e) speed away from the doomed dreadnought before it blows up. Mark has been working on the backgrounds for the dreadnought proper; he tells me he's nearly got enough to go for a test map.
Thursday 16th October, 1992
Mark remapped the 32-colour backgrounds into seven colours to create a palette in sympathy with the backgrounds. From a distance you wouldn't spot the difference, and that's without any touching up. The sprites needed reworking to get them looking good in the new palette, which has three red/oranges and only two blues. The explosions should now look even better. Definitely swaying 65%/35% in favour of the dual-playfield version! See what you think from the pictures here.
What will Andrew do next? Is Dex really Tinsel's illegitimate lovechiid? Who shot Brad? All this and much, much more, only in your Number One One.
CALLOUT: "It's getting tricky to tune the game now because I'm so used to it. That's why it's good to get outside help from time to time. A stranger to the game can be a lot less subjective about things."
(Above) This is the new dual-playfield version of Uridium 2, which has two layers of background graphics scrolling independently in parallax. The orange plate is the top layer, with the less colourful green plate on the bottom. You can see how the graphics are less colourful, but where this version shines is in the scrolling — duller it may be, but it's jaw-droppingly impressive and console-y once you see it move.
The latest version of the 32-colour Uridium 2 that Andrew has been working on up until now is likely to be scrapped in favour of the new dual-playfield format. Which one do you prefer — super-sexy scrolling or the multicoloured swapshop on offer here?
(Above) Flying the Manta over the vertical strip of flashing lights triggers the landing procedure. Here the ship is caught halfway through its process of transforming into a robot.
(left) the deadly "toothpaste" lasers in action in the controversial dual-playfield mode.
Andrew Braybrook Amiga Softography
Graphics: John Cumming
Music: Jason Page
O.O.P.S Kernel.: Dominic Robinson
Graphics: Michael A. Field,
John W. Lilley
Music: Jason Page
Graphics: John Cumming
Music: Jason Page
Sound: Steve Turner
O.O.P.S. Kernel: Dominic Robinson
Graphics: John W. Lilley,
Music: Jason Page
Graphics: Colin Seaman,
Music: Jason Page