VEE 2.3 Introduction Notes (jun 94)

last update 28 aug 98 / greg goebel / public domain / vwv_2_3

* This document provides introduction notes for the VEE 2.3 release in the spring of 1994.

[1] OVERVIEW
[2] BACKPLANE SUPPORT
[3] BUG FIXES

[1] OVERVIEW

* A new release of VEE for Windows is now available that provides support for direct VXI backplane on the RadiSys EPC-7 embedded VXI controller with VEE for Windows. HP VEE for Windows now allows full access to VXI instrumentation, both register and message based, both with and without ID driver panels, without the need for a hardware commander module.

While EPC-7 support was the only planned reason for this revision, it does bring other things to the party:

Support of the new 82340A and 82341B HPIB cards.
A new feature that allows you to set the DDE timeout period in milliseconds using the "DDETimeout" entry in the V.INI file: DDETimeout=5000
A new Strip Chart menu entry, "Config Buffer", that allows you to set the number of points saved by the Stripchart object.
Many important bug fixes.

Note that this new revision has the rev number 2.3, which is a different format from our previous releases of VEE. Originally this revision was numbered B.03.00, but we are now using a different numbering scheme on VEE releases.

[2] BACKPLANE SUPPORT

* Using VEEWIN with an EPC-7 also requires use of version 4.0 or later of Radisys's "EPCONNEC/VXI For DOS & Windows"; this is pre-installed with VEE on VEEWIN/EPC-7 bundles and is otherwise provided with current EPC-7 product.

Backplane support also requires a new software module for VEEWIN, known as the "soft commander" or "portable SCPI" (PSCPI). This module allows software emulation of a VXI mainframe so that VEEWIN can directly access VXI register-based cards; PSCPI contains VXI card drivers that convert message-level SCPI commands sent from VEE to register-level accesses required by the cards. (PSCPI is now included as part of the product.)

Note that to operate on the EPC-7, you will need to obtain the EPCONNEC 4.0 product (not to be confused with SURM 4.0, which is available on older EPC-7s). This product is available as HP part number RADI-EPCB (yes, that's correct).

Backplane access is through select code 16; VXI cards are addressed through the backplane by their logical address. For example, a card at logical address 24 is accessed as 16024 through PSCPI. Note that if VEE wanted to access this same card through an actual HPIB commander module, the card would be accessed as an HPIB secondary address that is the logical address divided by 8; the same card would be accessed in that case as 70908.

The following register-based cards are currently supported by PSCPI:


   DMMs:

      E1326B / E1411B

   Multiplexers & switches, capable of being controlled via a DMM:

      E1343A / E1344A / E1345A / E1346A / E1347A / E1351A / E1352A 
      E1353A / E1355A / E1356A / E1357A / E1358A / E1460A / E1476A

   Stand-alone multiplexers & switches:

      E1361A / E1364A / E1366A / E1367A / E1368A / E1369A / E1370A
      E1442A / E1463A / E1465A / E1466A / E1467A / E1468A / E1469A
      E1472A / E1473A / E1474A / E1475A

   Counters:

      E1332A / E1333A

   Digital I/O:

      E1330B / E1458A

   DACs:

      E1328A / E1446A

   D20 modules:

      E1450  / E1451A / E1452A

   Arbs:

      E1340A

   ADCs:

      E1413A/B

   Memories:

      E1488A

   Other HP-distributed cards:

      DDCC-1553 MilStd-1553 card
      BANC 350 IRIGB Clock/Timer

[3] BUG FIXES

* The following bugs have been fixed in VEE 2.3:

There was a problem where programs containing spectrum analyzer IDs could be stored but not reloaded.
If you configured a VXI driver to a secondary address greater than 15 (such as 70916, 70917, and so on), then the ID would send a reset each time the VEE model pinged it.
There were various problems that caused the serial port to be locked out.
Allocating records to a dataset resulted in a big-time memory leak.
A Strip Graph fed by a Sliding Collector resulted in bogus (flat) data after about 8,000 points.
There was a major bug in performing PRINTs or PLOTs from graphics objects
have too many points and not all the points will be printed or plotted.
DDE operations sometimes caused the final character of a string read from a server app to be lost.

All these defects have been repaired.