Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

AMD was already in the CPU market with bit-slice LSI chips, the Am2900 set of chips: https://en.wikipedia.org/wiki/AMD_Am2900

Those worked in 4-bit slices, and you could use them as LEGO blocks to build your own design (e.g. 8, 12 ou 16 bits) with much fewer parts than using standard TTL gates (or ECL NANDs, if you were Seymour Cray).

The 1980 Mick & Brick book Bit-slice Microprocessor Design later gathered together some "application notes" - the cookbooks/crib sheets that semiconductor companies wrote and provided to get buyers/engineers started after the spec sheets.



Intel has launched in 1974 both the NMOS Intel 8080 and a bipolar bit-slice processor family (Intel 3000).

AMD has introduced in 1975 both its NMOS 8080 clone and the bipolar bit-slice 2900 family.

I do not know which of these 2 AMD products was launched earlier, but in any case there was only a few months difference between them at most, so it cannot be said that AMD "was already in the CPU market". The launch of both products has been prepared at a time when AMD was not yet in the CPU market and Intel had been earlier than AMD both in the NMOS CPU market and in the market for sets of bipolar bit-slice components.

While Intel 8080 was copied by AMD, the AMD 2900 family was much better than the Intel 3000 family, so it has been used in a lot of PDP-11 clones or competitors.

For example, the registers+ALU component of Intel 3000 implemented only a 2-bit slice and few ALU operations, while the registers+ALU component of AMD 2900 implemented a 4-bit slice and also many more ALU operations.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: