We are trying to migrate PI ACE calculations to PI AF. One issue we've had is replicating PI ACE's latency functionality within PI AF. We have many PI ACE calculations that are very simple and do little more than output the value of one PI tag to another tag when a certain trigger occurs. However these calculations also have a variety of latency periods.
The best approach I've had so far is to use the "True For [x] seconds" feature of event frame generation analyses, so that the analysis doesn't trigger and write outputs until [x] seconds after the trigger tag receives an event. However it feels awkward to generate an event frame for every such analysis, and we have to make the [x] parameter static within the analysis template, instead of having the parameter derive from an attribute template. This does not allow us to account for the different latency periods across each context.
Thank you for any advice you can provide.