Category: Downland

This article is part of a series exploring the reverse engineered inner workings of Downland, a game for the Tandy Color Computer, released in 1983, written by Michael Aichlmayr.

If you look at the game’s video memory, internally Downland is actually a 1 bit game. The graphics are effectively black and white.

But the game actually has color. How does that work? Notice the spacing between the pixels on the floor and the diamond. The game draws its graphics in such a way to take advantage of CRT artifacting effects. These effects helps the game simulate a limited set of colors. This was a common technique used at the time, as seen on other computers like the Apple II.

Colors

Because of how the NTSC television signal works, different pixel patterns determine which color will be displayed and how it blends with other colors.

On the CoCo’s 256×192 graphics mode, each pixel is represented by a single bit. The artifacting in this mode interprets pairs of bits like so:

• 00 will be black
• 01 will appear blue
• 10 will appear orange
• 11 will be white
• A blue pixel next to an orange pixel (01 10) will also be white

The Tandy Color Computer also has a quirk where the blue and orange colors can be swapped when it powers on.

(To me, the blue version is the canonical version.)

The above is a very horribly simplified explanation of the Coco’s CRT artifacting. A much deeper dive on how all this works can be found in this great Coco Town YouTube video:

But in practical terms, the graphics mode effectively gives a 4 color 128×192 screen.

This “effective” resolution explains what I was saying about drop placement in the previous Downland Unearthed article. The game logic works at 128×192 while the graphics are drawn at twice the horizontal resolution. This counts for both the background and sprite drawing.

Another way of saying it is sprites need to move horizontally every two pixels, to maintain color stability when going across the screen. Otherwise the pixel colors would cycle like a (really limited) rainbow.

Here’s a capture of a hacked version of Downland_C simulating the player and the ball being placed every pixel instead of every two:

One-Bit Shapes

What I find interesting is the way the sprites have been drawn to take advantage of the CRT artifact effect. Here’s are side-by-side comparisons of the 1-bit graphics and the color graphics. The color versions were produced using the XRoar emulator in the the 5-bit LUT Composite rendering mode.

If you stare at the player sprite too long, you start to notice that he’s got blue AND orange hair and you can’t tell what the orange and blue parts of his face are supposed to be. Is that a blue eye at the lower half of his face? He has identically sized mustache and eyebrows? Also, he has ridiculously long arms and his shirt either has a single button or he’s just showing off his nipples. It’s hard to tell as this resolution. And who knows what the white pixel under his hand is supposed to represent!

Without color, Downland’s iconic bouncing enemy of doom is finally revealed to be just a killer walnut.

For a diamond, it’s drawn pretty plain, isn’t it? Still, this orange upside-down triangle gives 400 points! (more or less…)

M is for money, obviously. I like how the stray pixels add a kind of fringe around the top of the money bag.

Dithering makes straight lines? *mind blown*

I never know what the shapes on the doors are supposed to represent. A porthole and a wheel? Are they actually doors from a submarine? Is Downland actually set underwater?

Rom Storage

All the sprites in the game are stored with these patterns, as seen here using a raw pixel viewer to look at the rom’s contents:

(Notice how some of the sprites like standing left and right are perfectly mirrored but some, like the jump, aren’t.)

The one exception to this is the game’s font. It’s stored without effects. I couldn’t get the viewer to align the characters perfectly, but you can see that they’re made up of solid pixels.

Each character is 8 bits wide and 7 rows tall. The character drawing function draws each row while applying (and’ing) a bitmask (01010101) on top to turn off every other bit. This causes the character to appear blue.

I wonder why the mask wasn’t pre-applied. The game could’ve saved a few cycles when updating the timer and score to the screen. Maybe the author wanted to keep the font pristine for easy edits, not minding the performance hit. Or maybe he started with white text and then changed his mind.

Half The Battle

So that’s how Downland takes advantage of CRT artifacts to generate its graphics. One obvious unanswered question is how the graphics are actually drawn on screen. The graphics are 1-bit per pixel but writing to memory is done eight bits at a time as a byte. I hope to explain how Downland draws it sprites in the future.

This article is part of a series exploring the reverse engineered inner workings of Downland, a game for the Tandy Color Computer, released in 1983, written by Michael Aichlmayr.

Acid drops. If you die in Downland, chances are it’s because of one of these falling bastards.

Never mind the ball with its repetitive and predictable path across the screen. These white liquid drops of death are random, fast, and relentless.

Drop in a Bucket Chamber

For each chamber, the game data defines a number of drop spawn areas. Each drop spawn area is defined by an X and Y position, plus the number of drop points it has across the area. In the first chamber seen in the image below, there are seven drop spawn areas (boxes in red):

The first drop spawn area has thirteen points across it to spawn points from (marked in green).

A curious note is that in the game data the number of spawn areas and the number of drops to spawn per area is one less than it should be. So for the first chamber, the number of spawn areas is stored as 6 (not 7) and the number of drops in the first spawn area is stored as 12 (not 13). I don’t know what the advantage of this would be, if any.

Placing Drops

Some important background: The game’s world size is 128 by 192 points. The horizontal coordinate system goes from 0 to 127. The graphics mode that the game uses is PMODE4, which is 256 by 192 pixels. That means a point on the horizontal axis corresponds to two pixels on screen. Placing an object at the horizontal position 64 means they’ll appear at pixel 128 in the middle of the screen.

When the game wants to spawn a drop, it first picks randomly a drop spawn area. Then it picks a number between 0 and the number of drop points for that area. The drop is given a location that is starts from the drop spawn area’s XY position and then offset to the right by the the drop point chosen multiplied by eight.

So like this:

dropSpawnArea = pickRandomDropArea();
randomDropPoint = rand() % (dropSpawnArea->dropSpawnPointsCount + 1)
newDrop->x = dropSpawnArea->X + (randomDropPoint * 8)
newDrop->y = dropSpawnArea->Y

That’s generally the way it works, but there are two exceptions.

The first exception is about giving room to the player. Look at the left side of the top most drop spawn area in the image above. Notice that the drop is actually offset to the left of the box where its supposed to spawn?

Enhance!

That’s because the game tries to detect whether there’s a rope nearby. It checks four points to the right and six points down. If a rope is detected, the drop is offset to the left. This gives enough room around the rope to let the player climb up without being burned to death.

The second exception is about removing that room from the player and increasing the chances of them climbing up and getting burned to death. What Downland giveth, Downland taketh away. But why?

In the drop spawning code there is a part that says that if ever the player has completed the game three times in a row (which we must all admit is some kind of superhuman feat) the X position of the drop is made even. As the positions of the drop spawn areas are on odd positions, it effectively moves all the drops one position to the left, which corresponds to two pixels in that direction. In the image below, the small bright red boxes are the ones that have been affected by that rule.

This makes it so that the drops on the right side of the vine will always touch and kill the player, forcing them do perform rope acrobatics when climbing. This makes the game pretty much impossible to play, because drops fall on both sides of the rope simultaneously very often. Well, maybe it’s not impossible but the chances are definitely not in your favor if ever you manage to get to that point.

Wiggle, Wiggle, Wiggle

When a drop is finally placed and spawned, it hangs on the ceiling for 40 frames. As it does so, the game flips its vertical speed up and down every frame so that it wiggles. There were rumors that the wiggle time was random, but it’s not. Because of how drops can be spawned over other drops it might give an effect that it’s wiggling longer.

Gravity? What Gravity?

Drops fall at a steady rate. They go down two points per every update (but not every frame! see below). No fancy physics calculations here!

Yeah, But… How Many?

The maximum number of active drops is ten. But the actual number of drops spawned depends on the chamber number and whether you’ve completed the game at least once.

The rules that determines the number of active drops in a chamber are:

• If the chamber number is 5 or less, then the maximum number of drops is 6.
• If the chamber number is greater than 5, then the maximum number of drops is chamber number plus one.

• After completing the game once, the maximum number of ten drops is used for all chambers.

The maximum number of drops is also used for the title screen and the “get ready” screen. On those screens, the game lies to the drops system that the game has been completed so that it spawns all the drops. Also on those screens the game doesn’t actually wait for vblank so it just draws the drops as fast as possible.

Half The Work In Half The Time

Not all the drops are updated on every frame. The drops system only updates five drops at a time, alternating between the odd numbered and even numbered drops. This partial update means there’s more time left over for other parts of the game’s logic to be processed.

Waitaminute, You Forgot a Section!

So that’s the drops system in a nutshell. What I haven’t talked about is how the game handles the drop collisions with terrain and with the player. And also the drawing routines. I want to cover those topics more in-depth in later posts.

This article is part of a series exploring the reverse engineered inner workings of Downland, a game for the Tandy Color Computer, released in 1983, written by Michael Aichlmayr.

Late last year I started one of those projects I always wanted to do but could never find the time or couldn’t start it successfully. I reverse engineered Downland for the Tandy Color Computer, released in 1983.

Downland was one of those formative games for me. My best friend had a CoCo 2 with a copy of the game and we spent hours trying to beat it. But man, that game is hard.

The player jump physics are game’s best and aggravating aspect. They were the most fluid I had ever seen. But man, no air control whatsoever. When you leaped, you were committed all the way, like in Castlevania for the NES. The jump was floaty enough that there was enough time for a drop to appear ahead and fall on top of you just when you landed. Those drops are the worst.

It was my third or fourth attempt at RE’ing the game. I forget. For these first attempts I had used MAME’s debugger to run through the code and memory variables. I’d try to track memory changes and figure out what they did. And of course I had to learn how to read 6809 assembly code. (hint: I am not an assembly code guy)

At the time I could figure out some of the player state and the section of memory that tracked the drop state. But this method was too slow going for me. The MAME debugger was too cumbersome to use. I’d give up after a week or two. With just the MAME debugger and a text file, reverse engineering this way was like trying to figure out the layout of a blacked out mansion while walking around holding a match.

This time was successful because I decided to learn Ghidra (https://github.com/NationalSecurityAgency/ghidra). It helped a lot with making sense out the code and rom layout, and figuring out the data formats. I also used the trs80gp emulator (http://48k.ca/trs80gp.html) which has great debugging capabilities. The pair of tools really opened everything up and in very little time with these new found powers I managed to eclipse the little knowledge I had gained in the past doing things manually. I soon started learning things about the game that only the original author had ever seen. Finally being able to unlock secrets of this 40 year old game gave me such a wonderful feeling.

I forget the timeline but I think it took me over two months, maybe three, to get the disassembly in a state where I could make sense of how most of everything worked. Enough to be able to start figuring out how to convert it to the C programming language. At the same time I created a tool to convert the Ghidra listing to a buildable assembly file for LWTools (https://www.lwtools.ca/), with which I could build a byte-for-byte reproduction of the original Version 1.1 rom.

After the initial two months I kept updating the disassembly during the year the more I had to get into the details during the conversion to C. Some corners of the disassembly could use a bit more documentation but what’s there I believe is very usable.

The work can be found on GitHub at https://github.com/pw32x/Downland_RE

Maybe in the future if I’m not too lazy I’ll talk about more in detail.