The Nintendo 64 was released in 1996. Its "Controller Pak"s, which was the name of the memory card that you put inside the controller to save the progress in certain games, require a battery for them to retain the data. It can be seen in this video: https://www.youtube.com/watch?v=Su1Yr6rzmxE
If and when the battery runs out of power, or breaks down, or is removed, the save files are gone forever. In other words, it uses the kind of primitive "battery memory" that was also used in game cartridges starting with Zelda for NES in the 1980s.
However, the Sony PlayStation, released in 1994 in Japan, also had memory cards which had no batteries inside them, as proved by this video: https://www.youtube.com/watch?v=Yf5u1Jvvrkc
So, if Sony was able to do this two years earlier, why did Nintendo still opt for this archaic hardware design? Clearly, "persistent flash memory" (as I believe it's called) was quite invented and available. I actually always assumed that there were no batteries in the Nintendo 64 "Controller Pak" memory cards.
The price seemed to be roughly the same for both.