The problem lies in the fact that the
Trend Server quantises the data for millisecond trending on second
boundaries. So for a 100ms trend when requesting 485 samples (VGA
Trend template) it bumps it up to 490. The actual problem is that
when the client gets the 490 samples it says it only wanted 485 so
only uses the first 485 and ignores the last 5 - and hence the gap
- it should be using the last 485 and ignoring the first 5.
CIT has confirmed this to be a problem in Citect for Windows
versions 5.00. This problem has been fixed in version 5.00 Service
Pack D.
|