At 08:39 AM 4/5/01 -0700, Brad Velander wrote:
>Reading this thread, I can't help but ask if something I used to do back
>many years ago in Orcad would work with Protel for buses. I also used to
>deal with bussing large numbers of control signals in Orcad. It was not that
>hard to do with Orcad because it would eloquently handle multiple netnames
>on nets.

OrCAD net renaming was far from elegant. One of the worst schematic 
disasters I had to work with was an OrCAD schematic that had been put 
together from various pages from other projects. In order to get these 
pages to work together, nets had been renamed. Only problem is that it is 
less than obvious how to control net renaming: when you short two nets 
together, the combination might take one name or the other. I'm sure that 
there is some underlying rule but it is, as I said, less than obvious. 
Then, on this project, where the nets were shorted together on more than 
one page, it happened that the renaming went one way on one page and the 
other way on the other page.

It was *quite* difficult to disentangle all this. In my opinion, net 
renaming is usually more trouble than it is worth. Under some controlled 
circumstances, it could be useful. In many other situations, it makes 
spaghetti.

Protel presently supports net renaming only under one condition: nets take 
on the names of ports which connect to them, at the higher level. This 
allows nets to have one name on one sheet and another name on another sheet.

I would support another method of controlled net renaming, where hidden 
nets could be renamed, sheet-by-sheet, *not* globally*, by a visible 
device, as was done in Tango (and OrCAD supports the same thing, as I 
recall, but I am not sure it was well-controlled. Tango was.)

>         Used to work just fine in Orcad, does it work for Protel? It could
>cause some minor confusion to the un-initiated but after a minute or two and
>a brief explanation, everyone worked with it just fine.

Minor confusion? How about a week of work *after* we figured out what was 
going on, and still some stuff got by us. No, Protel does not support net 
renaming, or, at least, an error is generated. Like any error, it can be 
suppressed. Not a good idea.

Consider what has been suggested, as I understand it. One could take a set 
of control signals, say WR*, RD*, etc., and rename them CTRL1, CTRL2, etc.

One could then take these signals off the sheet using a bus port 
CTRL[1...]. Now, my question is, "Why is this superior to simply taking WR* 
and RD* off the sheet?" Sure, the sheet entry will take less space on the 
sheet symbol, but you have to add the renaming net labels, plus ports, and 
then, in order that the schematic make sense, one must rename the CTRL 
signals back to their original names on the other sheets. Or the reader 
will be face with wondering, "Now, was WR* CTRL1 or CTRL2?"

Why not just place a port on the wire in the first place, called WR*? And 
another, RD*, etc.

Numerical buses are useful and not confusing because we know what nets are 
associated with the bus, at a glance. There is no mystery. Allowing random 
names to be associated with a bus would either introduce confusion aplenty, 
or it would be a bicycle for a fish, that is, if sufficient signs are given 
to make the bus names clearly known, you could accomplish the same 
information with the same number of object or fewer.

If you must have a bus line to show that a set of control signals are 
associated together in some way, then, by all means, draw the bus line; but 
please don't confuse the situation by assigning multiple names to the same 
net *on the same sheet*. It's bad enough when this is a power net, but, 
unlike some of us who simply think that hidden power pins should never, 
ever be used, I think net renaming would be useful in this case only. If I 
see a power object with VCC on it, and a short piece of wire with the net 
label +5V, and these are connected to nothing else, and there are devices 
on the page with no power pins showing, I know immediately what is going 
on, it is not confusing. (In fact, such a renaming device was a clear 
indication, on Tango schematics, that there were hidden pins on the page; 
there was no other reason to do this.)
[EMAIL PROTECTED]
Abdulrahman Lomax
P.O. Box 690
El Verano, CA 95433


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To join or leave this list visit:
* http://www.techservinc.com/protelusers/subscrib.html
*                      - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Reply via email to