While I don't know what the "real" numbers are, I'm inclined to believe this
(4000) number, at least relative to the others just based on the internal
resources.
What I find odd, by the way, is that this ratio, seemingly quite reasonable,
doesn't show up in the IP-cores available today. While it's quite reasonable to
believe the 650x core with only four internal software-accessible registers
(A,X,Y, SP) would have a substantially lower gate count than a Z80, which has
lots of register resources, (A,B,C,D,E,H,L,IX,IY,SP, plus a second set of the
same) and it does, but nowhere near the ratio that these registers suggest. Of
course there are several ways of looking at the definition of "gate" but it's
odd that the ratio of gates consumed by each of these cores doesn't approach the
>2:1 that this estimate reflects. The production level pricing, basically a
cost based on silicon by the pound, seems to reflect this same ratio as do the
comparisons of the era when the 6502 was current.
What would be interesting, since it's conspicuously absent from the list quoted
below, is the transistor count in the 6802 (a 6800 with internal clock
generator) and 6809.
One thing that I found fascinating back in '81-'82 when MOT started a big push
for their two relatively new brain-children, the 68K and the 6805, was that the
6805 was, in fact, a bigger chip. They glued a bare chip to each data sheet
they passed out at their sales seminars back then and the 6805, with its
peripherals, timers, ram, ROM, etc, was markedly bigger. It made me wonder how
they fit the chip to the lead frame I was accustomed to associating with the 28
and 40-pin packages used for them.
Dick
----- Original Message -----
From: "Jim Battle" <frustum_at_pacbell.net>
To: <classiccmp_at_classiccmp.org>
Sent: Saturday, May 05, 2001 12:57 AM
Subject: Re: How many transistors in the 6502 processor?
> At 02:18 AM 5/4/01 -0500, you wrote:
> >Does anyone know how many transistors made up the 6502? These days with
> >Intel's boasting of the number of transistors their latest processors use,
> >it'd be interesting to know what we used to get by using. What, it can't
> >have been more than a few thousand, right?
>
> Wow -- such a long thread, and no answer yet that I've seen. The following
> came from a few web sites, but mostly from these two:
> http://www.intel.com/intel/museum/25anniv/hof/tspecs.htm
> http://www.mznet.ne.jp/svd/multimedia.htm
>
> 4004 -- 2,300 (10 um)
> 8008 -- 3,500 (10 um)
> 8080 -- 6,000 (6 um)
> z80 -- 8,500
> 6502 -- 4,000 one source, 9000 another source (I tend to believe the 4K
> number more)
> 8086 -- 29,000 (3 um)
> 68000 -- 68,000 (lots of that in microcode)
> 80286 -- 134,000 (1.5 um)
> 80386 -- 275,000 (1 um)
>
>
> -----
> Jim Battle == frustum_at_pacbell.net
>
>
Received on Sat May 05 2001 - 09:20:02 BST