Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!watmath!clyde!caip!cbmvax!daveh From: daveh@cbmvax.cbm.UUCP (Dave Haynie) Newsgroups: net.micro.apple Subject: Re: Re: Re: InCider's comments on the new //GS Message-ID: <885@cbmvax.cbmvax.cbm.UUCP> Date: Mon, 13-Oct-86 15:03:18 EDT Article-I.D.: cbmvax.885 Posted: Mon Oct 13 15:03:18 1986 Date-Received: Tue, 14-Oct-86 07:02:40 EDT References: <2116@ecsvax.UUCP> Organization: Commodore Technology, West Chester, PA Lines: 27 > CLOSE to RISC. The RISC-like features are: hardwired for speed, lots > of piplining and many 1 cycle operations. Although the 68000 can work > on 16 bits at a time it takes 4 cycles to do it. The '816 can load or > store 16 bits in 2 cycles. Not to put down the 68000 however because > it is a clearly superior chip. Comparisons to the 8088 are more apt > (and more fun!), the 65816 is a clear winner here, but there is all > of that 8088 software out there..... Rick, I agree with the 8088 comparison. But you've got more bus above that! The 68000 loads 16 bits in 1 (one!) MEMORY cycle, the '816 and the 8088 each load 16 bits in 2 (two) MEMORY cycles. The '816 uses a straight processor clock, the 8088 and 68000 essentially divide their input clock by 4 for their memory cycles. Its the memory cycle that counts, that's where the cost of memory is figured in, and its a much better meter of performance than the clock speed (which is why your 2.8MHz '816 will probably kill an 8088 at 4.77 MHz in most operations). The advantage of the 4 cycle input clock is that fewer internal operations are left to chance, since they have definate clock edges from which to be referenced. It makes for a much less asynchronous design. -- ============================================================================ Dave Haynie {caip,ihnp4,allegra,seismo}!cbmvax!daveh "Techno-Hippie, heathen, designing evil computers" These opinions are my own, though if you try them out, and decide that you really like them, a small donation would be appreciated.