THE WELCH COMPANY
440 Davis Court #1602
San Francisco, CA 94111-2496
415 781 5700


S U M M A R Y


DIARY: September 4, 1997 01:22 AM Thursday; Rod Welch

Upgraded BIOS CPU #8 does not fix problem; IBM advises video problem.

1...Summary/Objective


..............
Click here to comment!

CONTACTS 
0201 - IBM OS/2                           407 443 2000
020101 - Ms. Susan Proiepti-Conti; Technical Specialist =512 823 1245
020102 - Service & Support  =800 237 5511 1 or 800 799 7765 2

SUBJECTS
SDS Reports, Subjects
251 Pointer Update routine
Summary pointers, 910304
OS2, Copy, DOS command
10236,psc  Copy Command in 25.bat for
Video, 930817
SDS Data Base Update Failure under Os2,
Video MGA IMP +/P REV 201, SN AA62555C

1110 -    ..
1111 - Summary/Objective
1112 -
111201 - Followed up work at ref SDS 18 line 78, ref SDS 16 line 77.
111202 -
111203 - I tried the SDS data base update routine after installing the BIOS
111204 - upgrade, per ref SDS 17 line 131.  Had two successful runs and then
111205 - the thing crashed on the third try.
111206 -
111207 - Got call from Susan saying problme is the video driver.
111208 -
111209 - We tested this theory by running 25.bat under VGA mode, and it was
111210 - successful. Susan will send a letter explaining IBM's findings and
111211 - recommendations.
111212 -
111213 -
111214 -  ..
1113 -
1114 -
1115 - 1323 received call from Susan
1116 -
111601 - She indicates their review shows the problem may be related to the
111602 - video driver.  They want me to try the VGA driver to see if the SDS
111603 - report will run successfully.
111604 -
111605 - We changed the video driver to VGA, and then tried the SDS data base
111606 - update opertion from within SDS.
111607 -
111608 - This resulted in the program running 6 or 7 times without a problem.
111609 -
111610 - I later ran it a few more times and so this seems to establish that
111611 - OS2 works better in VGA, and can do as well in VGA as W95 does in 1280
111612 - x 1024.
111613 -
111614 - I asked Susan for help changing back to 1280 x 1024.
111615 -
111616 -  ..
111617 - Susan suggested that before we change back, we check with Matrox to
111618 - see if they have a newer driver, that will work better with os2.
111619 -
111620 -     I explained having obtained updated drivers from Matrox under
111621 -     similar cirsumstances of a request by IBM on 941219, ref SDS 1
111622 -     C19H
111623 -
111624 -     This was done on 941220. ref SDS 2 1V5M
111625 -
111626 -  ..
111627 - I advised of having installed an updated BIOS chip for the Matrox
111628 - controller on 970427, ref SDS 5 line 50, and on 970418 I downloaded
111629 - their most recent drivers, ref SDS 4 line 61.  This was recommended by
111630 - Microsoft to solve a problem of W95 not loading consistently.  It did
111631 - not solve the problem.
111632 -
111633 -  ..
111634 - Susan checked while we talked and said the Matrox Web site shows there
111635 - lastest video drivers for the Matrox Impression Plus, is dated in
111636 - April 1997, which seems to be when we updated, ref SDS 4 line 59. One
111637 - possibility is that we installed updated W95 drivers and not Os2
111638 - updated video drivers.  That is what the record seems to show, ref SDS
111639 - 4 line 63.
111640 -
111641 -  ..
111642 - Susan said I should contact Matrox to ask them to modify their video
111643 - drivers to solve this problem.
111644 -
111645 - I asked her to contact Matrox, since IBM is better positioned to
111646 - explain the problem.
111647 -
111648 - Susan said IBM does not have a working relationship with Matrox.
111649 -
111650 -  ..
111651 - I asked her to send me the technical explanation of the problem, so
111652 - Matrox will informed about what corrections to make.
111653 -
111654 - Susan said she will get this information from the person who analysed
111655 - the memory dump file.
111656 -
111657 - I asked Susan to send me a letter.  It can be via email.  She should
111658 - explain IBM's findings about the cause of the problem.  She should
111659 - advise of alternate video controllers which IBM supports, that will
111660 - avoid this problem.
111661 -
111662 -     [See follow up letter from Susan, ref SDS 19 line 92.]
111663 -
111664 -
111665 -
111666 -
111667 -
111668 -
111669 -
111670 -
111671 -
111672 -
111673 -
111674 -
1117 -
Distribution. . . . See "CONTACTS"