53 karma · joined May 4, 2022
wrote my first 3d games on the bbc computer (1982-1983) and then got a Mac in 1984.. and did more 3d on the Mac. I started StarGlider on the Mac, written in 68000 code.. and then ported it to the Amiga and Atari ST when they came out. I had the first Amiga outside the USA.
https://www.amazon.co.uk/Principles-Interactive-Computer-Gra...
https://www.amazon.com/Applied-Concepts-Microcomputer-Graphi...
And Fuzzz (aka James Hakewill, now at Tesla AI) was designing silicon including The ARC.
its a pretty well known story as to how we got involved with Nintendo.
I went up to a senior Nintendo guy at a CES show and showed him some stuff that blew him away. I was on a flight to meet Mr Yamauchi in Japan a few days later.
We had Argonaut team members stationed inside Nintendo working directly for Mr Miyamoto, et al
our team back in UK did the tech stuff, and the team inside Nintendo built the games.
It was only the second time that Nintendo had worked with a company outside Japan (the first was Rare ltd)
we thought that if we built a trustless casino technology that used smart contracts to power the games and generate provably fair random numbers so that no customer would ever have to play an unfair game.. and that no funds had to be held in the custody of the casino... that there'd be a market for that. we wuz wrong. the market doesn't seem to care about something that's non-custodial or guaranteed fair. we shuttered it about 18 months ago.
so we've pivoted. we're experimenting with NFTs, and driving games. And investing in other worthy projects.
we used the opposite approach to design the Super FX chip (we called it the Mario chip - likely the first ever gpu)
Carl & Pete would write code to do 3d math and render polygons... and write and re-write the code hundreds of times, using an imaginary assembler language. they would invent new instructions on the fly to let their code get tighter. they would keep coming up with new instructions, new registers, new ways of forming instructions and referencing registers etc... until the code was the tightest... and THEN, we designed a microprocessor from scratch to execute those instructions.
Also, likely the opposite way that microprocessors were designed... because (in the past) the designers had no idea what their cpus were going to be used for, whereas we knew what we wanted to achieve, so got to use software to drive the hardware design.
this was the first time that a cpu was able to be designed using programmable gate arrays. literally Ben & Rob could make a change to the hardware design - the netlist, and email it off to the usa, and actel would Fedex back the next day a working chip for us to try out. this was before you could program your own fpga's yourself (and Xilinx at the time wasn't large enough to fit a whole microprocessor). this iterative approach allowed us to design the optimum 'gpu' for the task.
btw, the arc core got used in lots of things that you never hear about... like in Sandisk memory cards (no relation), and various set top box chips (mpeg etc), and wifi chips (at the time, most intel wifi chips used arc)