RE: MOS chip revisions

I asked Dave Esposito who was a long time layout person at MOS/CBM:
Dave's comments are below except where I chimed in.

The simple R1, R2, R3  means full layer revisions.
The other numbers are layer revisions.

Example 1:   6567R56A
6567                R5                6A
Base part number    Mask revision     layer revision of the highest Metal
layer

    This is the 5th full mask revision  plus one more layer Metal 6
revision
    But that # is for identification only.  In reality this part probably
had a M6 VIA5 and M5 changed.  They only list the highest layer rev.


EXAMPLE 2:   8563R7A
8563                R7              A
Base part number    Mask revision   I'm not sure what the A/B/C means.
The 'A' could be packaging related. Ceramic vs plastic.  But I'm really
not sure.

BilHerd->I  think I saw this mean a backbias generator added once maybe a
change of implant energy once.  It could mean whatever the design team
meant as long as they didn't think the masks could get mixed up in the
process.  Dave DiOrio got burnt on the VIC on the C128 when they inserted
the desired mask to fix something and then inserted a mask from an older
rev and brought back a problem.


Also, after a while (when the chip was considered stable/bugfree?) MOS
dropped the 'Rx' after the part number and only printed the part number on
the IC. An indication that this chip was considered 'done'? Examples were
the 6526, the 8520 and the 8500 which had R-Numbers at the beginning but
then dropped them.

Correct.  Once the Part went into full production it was common practice
to drop the Revision numbers.  Sometimes it was even released as a new
part number.   Internally we knew it took (n) revisions to get it right.
But the part number in production does not need to reflect that level of
detail.


-----Original Message-----
From: owner-cbm-hackers@musoftware.de
[mailto:owner-cbm-hackers@musoftware.de] On Behalf Of Gerrit Heitsch
Sent: Monday, June 25, 2012 1:29 PM
To: CBM Hackers List
Subject: MOS chip revisions

Hello,

here's a question for Bil if he still remembers details...

How did MOS pick the revision numbers printed on the ICs?

The simple ones 'R1', 'R5' and so are not what I'm wondering about, but
there were also rather strange numbers that fell out of the standard
sequence like the 6567R56A or the 8563R7A and R9B or the 8566R3X.

Also, after a while (when the chip was considered stable/bugfree?) MOS
dropped the 'Rx' after the part number and only printed the part number on
the IC. An indication that this chip was considered 'done'? Examples were
the 6526, the 8520 and the 8500 which had R-Numbers at the beginning but
then dropped them.

  Gerrit


       Message was sent through the cbm-hackers mailing list

       Message was sent through the cbm-hackers mailing list