well, I must confess I also like and use global variables in programs!
there, it's out! I'm a globaller.

easier is not necessarily worse just as harder is not necessarily
better, just look at cost as a measure for goods or services, cost does
not correlate very well to quality or value received despite the old
adage 'you get what you pay for'.

I understand the attendant pitfalls of the usage of global variables in
programs, it all depends on the total complexity involved.
The main point, at least here, is to get working boards out the door in
manner which is as quick as possible while consistent with creating a
product of adequate (but not 'excessive') quality in the words of a
former business guru of mine.  If you wish I will attempt to explain his
concept of excessive quality at a later time.

I certainly understand that there are appropriate uses for the
additional overhead of hierarchical structures and their trappings, I
just appreciate that we are not forced into using them.  With designs of
20 or maybe 40 chips and several hundred components total (which count
is typically bloated with bypass caps and pullups) I find the global
approach both easier and better in the sense of being more fool proof.  
I'm sure there are many designs for which the approach I advocate is not
ideal.

Regarding a typo causing a disconnection you are certainly correct, for
this reason I try to always cut and paste net labels.  
Of course a typo would have the same effect in a hierarchical design.  
I got my tail caught in the door once (long ago!) when two busses were
not connected because one was ADx and the other was ADRx cut from
different schematics, the intended connections were obvious to me, but
not to the machine (lest you say 'aha!' they were on the same sheet).

Anyway, with the half-life of many products approaching seemingly days
and considering that so many projects out there never see the light of a
profitable day I am simply advocating the most expedient approach as I
see it for a certain scope of design.

As long as I'm ranting, I have found that in the (somewhat unlikely ?)
event that a given schematic actually leads to a successful and long
-lived board, the usefulness of the schematic itself tends to decrease
over time (flames are coming).  The reason I say this is that these days
things have to work right out of the chute or they are replaced or you
go out of business.  All most end customers/users want is a working box,
not a schematic.  Add to this the exponentiating rate of obsolescence.
In the early phases of course the schematic is vital for
troubleshooting, revisions and repair and such, but later if it is still
alive the schematic really ultimately assumes the form of a Bill of
Materials and Assembly Instructions and only us geeks can even
peripherally understand the original schematic anyway.  

Of course forge ahead with whatever works for you this is just my 3
cents ...

Dennis Saputelli
 

"van de Werken, Matthew (DEM, PH)" wrote:
> 
> I don't see how this is so. If for example there is a typo on the port on a
> single sheet, that connectivity still doesn't flow through whether the ports
> are global or not. A good reason to *not* use global ports is it allows you
> to use the same port name in more than one place, and also allows duplicates
> of the same sheet more easily. I also like using a hierarchical design with
> increasing amounts of detail as you delve into the hierarchy. IMO this makes
> a complex design much more readable.
> 
> Global ports/net names is analogous to global variables in programming.
> 
> Cheers,
> MvdW
> 
> >
> > to avoid this we just use nets and ports global and forget about all
> > that hierarchical stuff, sheet entries and module ports
> > we generally just have sheet symbols on a 'top' sheet which
> > is basically
> > unused except to gather sheets into a 'project'
> >
> > it's also a good idea to have all ic's NOT have hidden pins, then you
> > can control the power scope easily
> >
> > If you must use the sheet entries, etc. I think there is a way to list
> > unconnected pins, that should be easier than combing thru the netlist,
> > although I guess that would not cover groups of pins that are
> > connected
> > together on their own sheets but not across sheets
> >
> > that's another argument for 'nets and ports global'
> >
> > Dennis Saputelli

-- 
___________________________________________________________________________
www.integratedcontrolsinc.com            Integrated Controls, Inc.    
   tel: 415-647-0480                        2851 21st Street          
      fax: 415-647-3003                        San Francisco, CA 94110



* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
*  This message sent by: PROTEL EDA USERS MAILING LIST
*
*  Use the "reply" command in your email program to
*  respond to this message.
*
*  To unsubscribe from this mailing list use the form at
*  the Association web site. You will need to give the same
*  email address you originally used to subscribe (do not
*  give an alias unless it was used to subscribe).
*
*  Visit http://www.techservinc.com/protelusers/subscrib.html
*  to unsubscribe or to subscribe a new email address.
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

________________________________________________________

To leave the EDAFORUM discussion list, send a email with
'leave edaforum' in the body to '[EMAIL PROTECTED]'

More Information : http://www.dolist.net

Reply via email to