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: stuff in general
Message-ID: <895@cbmvax.cbmvax.cbm.UUCP>
Date: Tue, 14-Oct-86 13:30:26 EDT
Article-I.D.: cbmvax.895
Posted: Tue Oct 14 13:30:26 1986
Date-Received: Fri, 17-Oct-86 20:04:36 EDT
References: <9800034@uiucuxa>
Organization: Commodore Technology, West Chester, PA
Lines: 26

> I don't think a 6502 has a 4-1 advantage over a Z80. A 6502 has an 
> advantage because it does a full read in a cycle, unlike a Z80 
> which puts the address on the bus in one cycle and reads the data
> in the next. But that's only a two to one advantage. And all the 
> benchmarks I read seem to say about a 2x advantage, which is what you
> would think from my remarks.
> 
>    I run a 6502 at 2Mhz in my Commodo 128. It seem to be about 20% slower
> than an IBM PC.

Its between 3x and 4x from a purely hardware point of view.  When the
Z-80 fetches an op-code, it takes 3 clock cycles, all other memory or
I/O cycles take 4 of these clock ticks.  The Z-80 makes up for this
disadvantage by offering better software architecture for many things;
it has multiple registers and 16 bit operations, which can allow it to 
use more internally based operations, which is fewer memory accesses,
and thus achieve faster overall performance.  A Z-80 limited to the same
instructions that a 6502 has would be 3x-4x slower.

-- 
============================================================================
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.