OldComp.cz

... I don't see any systematic approach to the CPU-on-FPGA projects right now. I wanted to introduce a scientific touch to that work by coordinating a joint effort to write a 68k test suite. Obviously, this is a lot of work, but IMHO it's the only way to really *prove* that the newly-developed CPU is doing what the original 68k is doing.

See on a1k.org what that turned into: I was accused of making the project too expensive; any and all reasoning *for* a scientific approach was slapped down. Merely asking for the way that the core is tested has gotten me the role of the bad guy. That's not the way to make a product.

Yes, there is a core that runs a benchmark, and it runs that benchmark really fast. However, just starting the Amiga and run a few programs does not prove that the CPU is 100% working. It's not a product, but a proof of concept. It does have a chance to become a product, though. Lots of time has passed since I've made my comments about a requirement for a test suite. Maybe they did something like that now - I don't know.

Jens

A hele dnešní Jensovo vyjádření k FPGA. Bohužel to má hlavu a patu. Pokud má na něco dát záruku 2 roky, tak to nemůže fungovat napůl
...