Byron Q. Desnoyers Winmill escribió:
IIRC, the 6502 could execute many instructions in one or two clock
cycles, while the 80x86 required several clock cycles. But my
memory is vague.
Something simmilar. In the Apple II (and all other 68xx and 65xx-based
computers) the processor clock matched exactly the bus clock. That means
that a bus cycle (a memory read/write or an I/O access) lasts *exactly*
one processor clock cycle. Instead, in computers based in 8080, Z-80,
8086 and related processors, the processor clock ussually was three or
four times faster (Z80 computers used to run at 3 or 4 Mhz in the early
80s), but bus cycles takes four processor clock cycles each. So a
simmilar machine instruction could take four or five cycles on the 6502,
but fifteen or twenty on a Z80.
This means that an Apple II with an 1 Mhz 6502 is roughly as fast as a
CP/M machine based on a 4 Mhz Z80, or just a little slower than the
original, 4.77 Mhz IBM PC (mainly because the 8088's capacity to operate
on 16 bit intergers in less processor cycles). That a 2.8 Mhz IIgs is
actually faster than a 10 Mhz 80286 PC. Or that an 8 Mhz accelerated
IIgs can compete without problem with a 33 Mhz '386 PC (if you don't
mind not having hardware memory protection, that is).
If you sum to this that almost all early PC and CP/M software did all of
their I/O via the DOS and the BIOS, and that the PC BIOS was less than
optimal, you will be able to understand how amazed was I at the slowness
of an original PC (after issuing a DIR command, you could almost see how
lines were written in the screen) after years of working with my Apple
//c (and some faster PC clones).
Apple did really poor marketing of processor speeds, and still does
that, because the processors used by their products had allways been
able to do more per clock cycle than Intel equivalents (6502 vs. 8088,
65826 vs. 80286, 68030 vs. 80386, PowerPC 601 vs. Pentium, PowerPC G5
vs. Pentium 4, etc.). Other companies (i.e., AMD) have had the same
problem, but have solved it more creatively. For example, my Sempron
2400+ actually runs at 1.67 Ghz... and performs better than a 2.4 Ghz
Pentium 4 (the "2400+" name comes from that). Many PC users near me say
things in the line of "Macs are really slow, there is no model faster
than 2.0 Ghz, but there are 4.0 Ghz P4 computers". They don't understand
that a 2.0 Ghz G5 is actually faster than a 4.0 Ghz P4, and that is
Apple's blame - and nobody else's. Maybe that is one of the reasons in
the switch to Intel processors anounced for the next year (Pentium 4 is
a killer in raw clock speed), but I think this is another matter for
another thread :o) .
Greetings,
Antonio Rodríguez (Grijan)
<ftp://grijan.cjb.net:21000/>
--
Apple2list is sponsored by <http://lowendmac.com/> and...
/ Buy books, CDs, videos, and more from Amazon.com \
/ <http://www.amazon.com/exec/obidos/redirect-home/lowendmac> \
Support Low End Mac <http://lowendmac.com/lists/support.html>
Apple2list info: <http://lowendmac.com/lists/apple2.html>
--> AOL users, remove "mailto:";
Send list messages to: <mailto:apple2list@mail.maclaunch.com>
To unsubscribe, email: <mailto:[EMAIL PROTECTED]>
For digest mode, email: <mailto:[EMAIL PROTECTED]>
Subscription questions: <mailto:[EMAIL PROTECTED]>
Archive: <http://www.mail-archive.com/apple2list%40mail.maclaunch.com/>
iPod Accessories for Less
at 1-800-iPOD.COM
Fast Delivery, Low Price, Good Deal
www.1800ipod.com
