Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!seismo!rutgers!mit-eddie!uw-beaver!tektronix!tekgen!tekigm!tekigm2!phils From: phils@tekigm2.TEK.COM (Philip E Staub) Newsgroups: comp.sys.amiga Subject: Re: Some thoughts and questions about the Ranger Message-ID: <1542@tekigm2.TEK.COM> Date: Sun, 8-Mar-87 17:28:19 EST Article-I.D.: tekigm2.1542 Posted: Sun Mar 8 17:28:19 1987 Date-Received: Mon, 9-Mar-87 18:59:07 EST References: <140@tahoma.UUCP> <158@rocky.STANFORD.EDU> <1513@cbmvax.cbmvax.cbm.UUCP> <2725@well.UUCP> Reply-To: phils@tekigm2.UUCP (Philip E Staub) Organization: Tektronix, Inc., Beaverton, OR. Lines: 44 Keywords: C64 In article <2725@well.UUCP> ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) writes: >>[ a discussion to which Leo is responding containing comments that a 6502 is a pretty quick machine] > Really? You think so? The fastest I've ever seen those things >commercially is 2 MHz. The C-64 runs at (slightly less than) 1 MHz. After >moving to the 68K, the very THOUGHT of going back to the 6502 makes me gag. > Careful here. I've been sucked in by that trap. Notice that the clock rate does *not* indicate how long it takes to execute one instruction. The 6502 requires *at most* 2 clocks to execute each instruction, where the 68000 requires no fewer than 4. Granted the instructions are generally more primitive, but there still will be no way to emulate more than one 6502 instruction per 68000 instruction (in fact it will take several). Let's make some rash assumptions here: First, let's assume you *can* emulate any 6502 instruction with a single 68000 instruction. Next, let's assume that an "average" 68000 instruction requires 6 clocks per instruction, while an "average" 6502 instruction requires 1.5. The way I figure it, that means that a 2 MHz 6502 would require an 8 MHz 68000 to emulate it in "real time" even if no other overhead (read: software emulation of whatever other hardware had to be done) were required. (Yes, I know you said < 1MHz, but the 1 6502 instruction = 1 68000 instruction is also bogus. The translation will require something on an instruction by instruction basis) ... > Personally, I'm of the opinion that it should be more than possible >for a 7.14 MHz 68K to emulate a >1 MHz 6502 in software, generally. After >all, the underlying processor architectures and philosophies are similar >(any data movement sets flags, rich addressing schemes, etc.). ^^^^^ > Yeah, but watch out for the "carry = not borrow" flag. You'll have to watch out for what the last instruction was before you decide to "bcc" or "bcs". -- ------------------------------------------------------------------------------- Phil Staub tektronix!tekigm!phils (206) 253-5634 Tektronix, Inc., ISI Engineering P.O.Box 3500, M/S C1-904, Vancouver, Washington 98668