The Super FX coprocessor, released with Star Fox in 1993, provided 3D rendering capabilities that were absent from the SNES itself. Given that 3D gaming was quite widespread by the late eighties, it seems unlikely the omission of such capabilities from the base console (apart from the 'mode 7' background affine transformation) was because no one at Nintendo thought of them at the time, more likely because they could not be developed by the 1990 release date, and even more likely because a larger graphics chip would have gone over the chip area budget at that time. Presumably an extra three years of Moore's law made it possible to pack the extra circuits into a smaller, cheaper chip.

How many transistors are in the Super FX, compared to the base Super PPU?

  • 1
    I'm not sure I would agree with the "3D gaming was widespread by the late 80s".. if we consider the arcade to generally be cutting edge in the 80s, I think Atari's "Hard Drivin'" (1989) was one of the first real 3D games that wasn't wire frame (e.g. Battlezone 1980) or fixed perspective (e.g. Zaxxon 1983). 3D wasn't really widespread IMHO until Wolfenstein 3D came out in 1992. I would imagine that given the lead time for console development, it wasn't an oversight but rather just not possible with mid-80s tech. The fact you could add it later with the SFX chip was admirable forethought.
    – bjb
    Commented Mar 12, 2019 at 17:11
  • 1
    I'm not sure raw transistor number is a useful comparison to make. The developers of the chip, Argonaut, joked how the Super NES was just a box to hold the superfx chip. A more useful comparison would be between the features of the two chips.
    – nabulator
    Commented Jan 4, 2021 at 4:43


You must log in to answer this question.

Browse other questions tagged .