From scratch to cartridge

TranslationShow original
How Atari 2600 games were made in the 1970s and 1980s
TranslationShow original

From scratch to cartridge: How Atari 2600 games were made in the 1970s and 1980s

A journey through early game development, from sketches on paper to a finished game inside a plastic cartridge.

Making games for the Atari 2600 in the early days of video games was very different from working with today's game engines. Often, a single person handled the programming, graphics and game design. Instead of a convenient editor, they relied on notes, custom tools and plenty of patience.

1. Paper, pencil and a plan for the screen

Graph paper was a useful companion when designing a game.
  • Designing sprites: Characters and other objects could first be drawn dot by dot. The rows of the drawing were then converted into numerical values, often written in hexadecimal, and included in the program as graphics data.
  • Planning the timing: The console had no memory buffer for a complete screen image. The programmer therefore had to time updates to the TIA graphics chip precisely as the picture was being drawn. This part of the program is called the kernel, and every processor cycle counts.
For a closer look at why timing was so demanding and how developers worked around the graphics limitations, see The extreme challenges of developing for the Atari 2600.

2. Writing the code in assembly language

Games of the period were usually written in assembly language. It gave programmers direct control over what the processor did and how long it took.
  • The MOS 6507 processor: It used the instruction set of the 6502 family. Mnemonics such as LDA, STA and JMP stood for individual instructions. Programmers had to conserve both the ROM space for their code and the console's mere 128 bytes of built-in RAM for working data.
  • Development computers: Source code was written on another computer, where an assembler converted it into machine code. That code was then transferred to development hardware connected to the console. The equipment varied from company to company.
To find out what LDA, STA and JMP mean, see Atari 2600 instructions, a guide to all 56 official instructions.

3. Debugging on real hardware

A timing mistake could shift an object or corrupt the picture; incorrect synchronisation could even make the image roll. Fixing a program sometimes meant investigating what was happening inside the console, too.
  • Test equipment: A logic analyser or an oscilloscope could help trace signals and reveal a problem that was not obvious from looking at the television.
  • ROM emulators and EPROMs: A ROM emulator replaced cartridge memory with rewritable RAM, allowing a new version to be downloaded quickly from the computer. Another option was a test cartridge containing an EPROM. This had to be erased with ultraviolet light before reprogramming.

4. Turning the finished program into a cartridge

Once the game was finished and tested, it was time to manufacture the cartridges.
  • ROM chips: Large production runs commonly used mask ROM chips, with their contents built in during manufacture. Finished cartridges were not individually “burned”. The chip was mounted on a circuit board inside the plastic shell.
  • The journey to the shop: Each cartridge received a label with the game's name and artwork, was packaged and shipped to retailers. At home, you simply inserted it into the console with the power off, then switched it on. The processor read the program directly from the cartridge, and the game could begin.
Share
Share Facebook WhatsApp X Telegram Reddit E-mail

Sign in to add a post.