5 Sep
2007
5 Sep
'07
8:32 p.m.
On Wed, 5 Sep 2007 10:51:05 -0700 (PDT)
Chris M <chrism3667 at yahoo.com> wrote:
So the benefits of pals/gals are not so much
integration or a high degree of, but rather
modification?
There are many different variants of PALs. Clocked, registered,
tristate, ... I can't remember waht was available. GALs have
configurable outputs. With a PAL you have to choose a special chip to
get tristate outputs. You need an other chip for open collector outputs.
(Or somthing like that.) A GAL has all this build in. You can switch the
type by programming it. So a single type, a GAL 16V8, can replace many
different PAL variants.
You wouldn't want to, for instance,
attempt a *reproduction* (of say a unobtainium piece
of silicon) by using a bunch of galpals instead of an
FPGA?
GALs are much, _much_ lower integrated as FPGAs. FPGAs are more flexible
then GALs also.
How much does the typical programmer cost? Are they
*easy enough* to build?
A long time ago I was building a GAL programmer baised on a
BASIC stamp.
Given the limited capabilities of the BASIC stamp you can imagin that it
isn't that hard to programm a GAL.
--
tsch??,
Jochen
Homepage: http://www.unixag-kl.fh-kl.de/~jkunz/