15 Mar
2005
15 Mar
'05
4:30 p.m.
I wrote:
These are easily programmed using zero-cost
development software you
can download from the Xilinx web site, and a simple parallel port
cable you can build yourself easily.
Tony wrote:
The software may be 'free',
I specifically didn't use that word because it definitely is not
"Free Software".
but the machine to run it on, and the OS to
run it under, most certainly aren't.
I didn't pay any money for the Linux OS I run it on. And it will
run (perhaps slowly) on cast-off Pentium machines that people simply
throw away.
When DECUS gives you a "free" C compiler for the PDP-11, do you
complain about having to pay for the PDP-11 computer and the
RT11 operating system?
And I do have an objection to
trusting my design to a piece of binary-only software that I have no easy
way of checking if it's doing the right thing
The Xilinx FPGA Editor lets you see *exactly* how the resulting logic
is wired in the device. However, unfortunately the FPGA Editor is not
included in the no-charge dowloadable software; it's only in the paid
version, which costs US$500 and up.
(it's very difficult to be
sure a CPLD or FPGA is doing the right thing under _all_ conditions).
It's difficult to tell whether any collection of thousands, tens of
thousands, or hundreds of thousands of gates are doing the right thing
under all conditions, regardless of whether they're in an FPGA, or a
bunch of 7400 NAND gates you've wire-wrapped. In fact, it's
essentially impossible.
I would mind if they'd properly document the chips
so I could write my
one CAD software if I wanted to. But they don't. AFAIK there is no 100%
documented CPLD or FPGA available (100% documented meaning you can go
from design to chip without proprietary software or a proprietary
programmer).
There are plenty of completely documented CPLDs, and no completely
documented FPGAs.
Xilinx did completely document one FPGA family (XC5200? XC6200? I've
forgotten). It didn't sell well, so they discontinued it.
However, I don't *really* think you want to spend the rest of your
life writing your own FPGA development software. It's a hard problem
and there are hundreds of thousands of man-years of development effort
in the Xilinx software. By the time you got your own software working
for one family, that family would have long since been discontinued.
Eric