What inspired this wish list request?
I submitted my experience with Pi tags here: Pi Import Options - the Max count option - JMP User Community and a Tech Support case TS - 00192427. Tech support said they are looking into options and asked me to post the request here.
The effect of limiting each Pi tag output based on the total # of results allowed divided by the number of tags requested severely limits the output of frequently collected Pi tags and makes it appear that there is a problem with the tag when the data abruptly stops while all other tags (with less frequent collection) continue. (Example graph - green stopped outputting before red, even though I thought they were identical. Both stopped way before purple.) It also limits how many tags you can pull at a time - even if most are very infrequent (once per day or once per eight hours) while others are every 30 seconds. This error message limiting me to 37500 tags is for the "system limit" of 150,000 results divided by pulling 4 tags at the same time. This applies to JMP 18, which is the only version where this feature for multiple tag import exists (I am using 18.1.1).
What is the improvement you would like to see?
In order of difficulty:
1) Add documentation to the JMP help menu / webpage about how Max Count feature works for Importing Pi tags and what the limitations are
2) Add a Warning or Error to Pi tag outputs when pulling multiple tags if some tags reach their limit before the time range while other continue through the full range.
3) Change the method of pulling tags so that the full value of # of tags (max count - 150,000 "system limit") can be used - either by
a) going left to right across the tags
b) by pulling each individually and combining (recommended)
Why is this idea important?
I cannot recommend this great JMP feature (pulling multiple Pi tags at once starting in JMP 18) to coworkers if we can't rely on the outputs. I can put together complex scripts to loop through, but each area in my plant is very different and I can't cover them all with unique code. The interface for manual Pi tag pulls is great for the more casual user, but it has to be robust or it creates more frustration and busy-work to understand the odd outputs than it creates opportunities to learn from the data.