VERICUT Users' Forum

RobD

Member

From: Connecticut

Registered: 2005-02-07

Posts: 20

Logical Product AND

Hello,

My Fanuc 15MA with Macro B capability seems to have a difference of opinion with Vericut V5.4.5 and V6.0.1, and I am wondering if anyone else has seen this.

BACKGROUND:
==========
AND is defined by Fanuc as "Logical product (at every bit of 32 bits)".

On the Fanuc, for example, [3AND3] = 3.  I am not electrical, but I am told that this is the result of some BCD math wizardry.

PROBLEM:
=======
In Vericut, the answer is not the same.  In Vericut, [3AND3] = 1.  Oops.

SUMMARY:
=======
This makes a big difference to me because I am looking at the Vericut probing module, and I am using standard subroutines provided by the probe manufacturer.  Their standard routines use AND.  The simulation was not returning the correct values in certain key registers.  I discovered that I must make edits to their subroutines on my PC to make them work for my programs in Vericut.  I do not really feel great about doing this.  I would be willing to bet that OR and XOR misbehave as well.

CONCLUSION:
=========
So I am wondering if anyone has encountered this and what you did about this.

Thank You

jsmillett

Member

From: Chatsworth, CA

Registered: 2005-02-25

Posts: 42

Re: Logical Product AND

In the the Control File menu: Configuration -> Word Format

change:

Name       Type     Sub_Type    ...
AND         Logical  AND
OR           Logical  OR

to:

NAME       Type      Sub_Type    ...
AND         Logical   Bitwise_AND
OR           Logical   Bitwise_OR
XOR         Logical   Bitwise_XOR

Just watch out for operator precedence issues.

---

Jerry Millett

RobD

Member

From: Connecticut

Registered: 2005-02-07

Posts: 20

Re: Logical Product AND

Thank You.  That worked in V6.  I had to edit my .ctl file, because none of the V6 library fanuc files had the "Logical_Bitwise" definitions within.

I did not totally understand your caveat "Just watch out for operator precedence issues."  I decided to do more testing.  Oh, shucks!  I ended up with conflicting results.  I created a simple test script and ran it three ways:

1. Fanuc control
2. Vericut with legacy AND
3. Vericut with Bitwise AND

All three ways give me different results.  Here is my test script:

#551=[3AND3EQ1]
#552=[[3AND3]EQ1]
#553=[3AND8NE8]
#554=[[3AND8]NE8]

I hoped that Fanuc and Bitwise AND would agree.  Wrong!  Here are the results:

Register   Fanuc   Bitwise AND   Legacy AND
===========================================
#551          0         0           0
#552          0         0           1
#553          1         0           0
#554          1         1           1
===========================================

I can't see a way out of this one.  I'll have to edit the vendor's sub-routines and submit a trouble ticket.

Thanks Again.