The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: Rick Filipkiewicz
X-Mailer: Mozilla 4.75 [en] (WinNT; U)
Subject: Re: Need advice wiring up a CPLD
References: <3D9B8FDE.firstname.lastname@example.org> <3D9C92AE.email@example.com>
Organization: MIPS Technologies (UK) Ltd
X-Cache: nntpcache 2.4.0b2 (see http://www.nntpcache.org/)
Date: Fri, 04 Oct 2002 08:49:06 +0100
NNTP-Posting-Date: Fri, 04 Oct 2002 08:49:07 BST
"Christopher R. Carlen" wrote:
> Ken Smith wrote:
> > In article <3D9B8FDE.firstname.lastname@example.org>,
> > Christopher R. Carlen wrote:
> > [....]
> >>Everything is pretty well thought out so far, I think. The only problem
> >>is, the CPLD has 84 IOs available, of which I plan to use up to 40. 32
> > I suggest you spread the I/O connection among the logic blocks in logical
> > groups.
> > If you are fairly certain the other lines will not be needed, you can hook
> > pairs of them together so that you have another way to get signals between
> > logic blocks or you can wire up a socket for a 22V10 or spare I/O. This
> > will help keep your options open.
> Perhaps it will suffice to bring the unused IOs out to vias. Then I can
> tie them together later if need be.
> It may even be less trouble to build a new daughterboard for the XCR3256
> chip later on. But the XCR3128 seemed like it would give a lot of room
> for growth over the XCR3064 which I originally considered. But the main
> problem their was having to share the JTAG pins, which I wanted to avoid
> for simplicity.
I don't think there's anything to stop you having multiple alternative
connections to the same signal as long as, obviously, all but one are tristated.
So pick the best one you can think of for the 40 I/Os and then double up the
others until you run out of pins (e.g. allocate the alternates on the other side
of the chip from the default).
Another thing would be to have multiple `configurations' in the chip at the same
time (one per engineer if poss) I.e. each pin's logic function is selectable via
a header. Each magic box user would have a plug which they push onto the header
to ``personalise'' (or DIP switches or, really slick, a little keypad (*)).You
could go further so that when you reach the point where the engineers are
re-programming the device they would each have their own code `module' ...
(*) At this point you could even start adding some user id checks. If any of the
engineers offends you in any way you can cut them out and display a message
``Access denied, please contact magic box administrator'' :-).
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup