nfg.forums

NFG Forums => SIG X68000 => Topic started by: TheRogue on December 01, 2009, 08:59:15 AM

Title: Using a 68882 with an X68000 XVI
Post by: TheRogue on December 01, 2009, 08:59:15 AM
I know that the X68000 XVI has a socket for a 68 pin PGA 68881 FPU, and I'm wondering if anyone has had success using a 68882 in that socket.  The 68882 is supposed to be pin and opcode compatible with the 68881, however I have only seen the 68881 mentioned in conjunction with the XVI.  All the information I can find seems to point to the 68882 being only compatible with the 68020 and 030. Buying either a '81 or '82 in 68 pin PGA form factor is not cheap, so I don't want to get a '82 if it won't work, however I don't want to be stuck with the inferior '81 if the '82 is compatible.  Anyone have any experience in this area?
Title: Re: Using a 68882 with an X68000 XVI
Post by: Magic Knight on December 01, 2009, 07:22:07 PM
I can't really see any point in looking for a 68882, you'd have to get one that runs at 16MHz for a start, and I don't know what benefits you'd get from the 68882. I have a 68881 (CZ-6BP2) and I've never used it for anything, as far as I know. Incidentally, you need to use \SYS\FLOAT3.X to make use of the 68881.

What kind of programming do you intend to do with it, as a matter of interest? I've got GCC and HAS installed on mine.
Title: Re: Using a 68882 with an X68000 XVI
Post by: aramokona3 on December 02, 2009, 12:08:10 PM
The floating point coprocessor MC68881 and MC68882 were magic items to improve science programs dramatically at that time. However, most games do not use a coprocessor even if it exists.

MC68882 may work with X68000 XVI, but it is not recommended for X68000 XVI users. Because, the floating point coprocessor driver FLOAT3.X can perform one floating point instruction at a time. It means that the driver negates the single most important advantage, that is overlap execution of instructions, of MC68882 over MC68881.