Smart fill critical anomaly : only 2 decimal places!

  • 1
  • Problem
  • Updated 4 years ago
Unless I've misunderstood something, there seems to be a major and surprising anomaly in Smart Fill, which turns out to be absolutely critical to my client's corporate model.

One can adjust Properties' decimal places okay. HOWEVER one can't reflect this in the Smart Fill criteria! SF can only accommodate a maximum of 2 decimal places! This has completely blown an important model out of the water, because at 2 decimal places, SF is insensitive to critical threshold values.

To illustrate by example: scientifically derived Properties are input as percentages in the present model, to four decimal places (i.e. 98.9999%). This means one needs to select SF criteria to 6 decimals (i.e. 0.989999) - but you can't.

Just in case it's not obvious [politely meant, of course]: the decimal place definition for any SF instance needs to be independently adjustable (i.e. not slaved to a decimal place setting anywhere else), because one might require SF thresholds that are more or less sensitive than the associated Properties.

This is clearly an anomaly, doubtless unintended. Could you please fix it urgently. Even a nice safe  beta patch will do me for now!

Thanks as always,

David
London
Photo of David Abrahams

David Abrahams

  • 45 Posts
  • 4 Reply Likes
  • Frustrated

Posted 4 years ago

  • 1
Photo of Nick Duffill

Nick Duffill, Champion

  • 508 Posts
  • 169 Reply Likes
As a temporary workaround, could you use a scaled integer value (e.g. 98993 instead of 0.98993)? It is not ideal but it might be easier to tolerate than incorrect SmartFill.

A second level of workaround would be to get someone to write a macro that automatically calculates unseen integer values that are used to control SmartFill.
Photo of David Abrahams

David Abrahams

  • 45 Posts
  • 4 Reply Likes
Thanks, Nick. Noted and really appreciate your fast, helpful thoughts.

In the present case, the client is considering migration from an Excel-based model, with which they have become very familiar. So scaled integer values may confuse or disappoint, and will just highlight a shortfall in the [great] MM product - which I am encouraging them to adopt!

I frankly don't have experience of macros in MM, so can't judge whether that workaround would be simple, seamless and flexible enough in operation to replace what's missing. Anywhere I could go to learn more about where a macro such as the one you envisage would reside and how it would be invoked in a case such as this?

What I'm really hoping for is a little helpful update before too long ...

Meanwhile - many thanks!

David
London