Two-Bit Astrophotography

The Game Boy Camera is a 128×112 pixel sensor from 1998 that was probably the first digital camera in many, many homes. There’s not much you can do with it now, besides replicate old Neil Young album covers and attempting and failing to impress anyone born after the year 1995. Nevertheless, screwing around with old digital cameras is cool, so [Alex] strapped one fo these Game Boy Cameras to an old telescope.

For any astrophotography endeavor, the choice of telescope is important. For this little experiment, [Alex] used a 6” Fraunhofer telescope built in 1838 at the Old Observatory of Leiden. The Game Boy with Camera was attached to the scope using a universal cell phone adapter. Apparently the ‘universal’ in this universal cell phone adapter is accurate – the setup was easy and [Alex] quickly got an image of a clocktower on his Game Boy.

Turning to the heavens, [Alex] took a look at the most interesting objects you can see with a 6-inch telescope. Images of the moon turned out rather well, with beautiful 2-bit dithering along the terminator. Jupiter was a bright white spot in a sea of noise, but [Alex] could see four slightly brighter pixels orbiting where Stellarium predicted the Galilean moons would be.

Was this experiment a success? Between cloudy nights and a relatively small telescope, we’re saying yes. These are pretty impressive results for such a terrible digital camera.

12 thoughts on “Two-Bit Astrophotography

    1. The camera itself actually has an analogue output and a clock input. Every clock, you get a voltage on the pin per pixel. Read all the pixels and assemble a greyscale image after. 4 colors is what the gameboy does with it because it doesn’t have more colors to work with on it’s display.

  1. Wasn’t the imaging chip used in this camera pretty unusual? I remember reading about it when it came out and despairing at my lack of coding ability to take advantage of the fact that it had a simple vision-processing microcontroller built into the chip that would allow you to do things like edge detection and scene-change detection using the imager itself.

    It’s actually been rather unfortunate that the hacking possibilities of this architecture were never fully explored.

    1. Not particularly. The image itself (M64282FP) appears to be a “normal” CMOS sensor, complete with analog output. The ADC inside the cart (MAC-GBD) does have some interesting features for converting that analog output into something that can be displayed on the DMG, including ordered dithering as part of the ADC and simple convolution matrices. (see )

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.