Path: utzoo!attcan!uunet!cbmvax!daveh From: daveh@cbmvax.UUCP (Dave Haynie) Newsgroups: comp.sys.amiga Subject: Re: 68020/68881 board Message-ID: <4447@cbmvax.UUCP> Date: 8 Aug 88 21:13:09 GMT References: Organization: Commodore Technology, West Chester, PA Lines: 33 in article , melnik@topaz.rutgers.edu (Ofer Melnik) says: > In article <7797@cup.portal.com> Gary_D_Walborn@cup.portal.com writes: > !M.A.S.T. (Memory and Storage Technology) announced a new accelerator board > !for any Amiga that plugs into the 68000 slot. The board features a 14 MHz > !68000 and a socket for a 68881. ... > Sounds very interesting... BUT, To venture a guess, based on the assumption they did it right... > 1) How compatible is it with current software? Very, like probably 99.9%. What's likely to die? Only thing that occurs to me is the R-C type dongle, as used with X-CAD. These things are very sensitive to CPU speed, generally depending on CPU dependent timing loops. Under V1.3, the 68881 interface will be supported if they did it right. > 2) How does it compare to a 14 Mhz 68020? w/68881? The 68020's 68881 interface is much faster, but the internal operations should be about the same. In a 68020 system without 32 bit wide memory, a fast 68000 will sometimes do a little better, sometimes a little worse. The trick is that the 68020 always does 32 bit pre-fetches, even if it only needs 16 bits. However, it has an instruction cache which usually makes up for this, and in some cases goes far beyond. If you've got fast, 32 bit wide memory in the system, the 68020 will show up way out in front of a fast 68000. > Ofer Melnik Rutgers University -- Dave Haynie "The 32 Bit Guy" Commodore-Amiga "The Crew That Never Rests" {ihnp4|uunet|rutgers}!cbmvax!daveh PLINK: D-DAVE H BIX: hazy "I can't relax, 'cause I'm a Boinger!"