OK, first in with Metric vs Imperial for conservatism ...
Trying not to give any spoilers, but 9-10 minutes (13%ish ) of calculated difference in bottom time on the second dive to get the same PG outcome between metric and imperial is ... remarkable.
First, my 13% difference above was calculated based on the effective TBT (Total Bottom Time) including the RNT (Residual Nitrogen Time).
I'm only familiar with the imperial RDP. I assume there is a (slight?) difference in how things work with metric (due to rounding, lack of decimal points perhaps?)? If so, could you be slightly under the NDL with one and slightly over with the other?
Doing the calculations on the imperial table, RNT is 31 and you end up with a TBT of 76 in the 50' column, r-up to 80 -> group X. 80 is NDL for 50'
With the metric, RNT is 35 and you end up with a TBT of 80 in the 14m column, r-up to 82 - group V.
Group W in the metric 14m column is 87 and group X is 92. 98 is Group Y and NDL.
So if you are running the metric tables, to get to group X you would extend your 45 minute dive by between 8 and 12 minutes.
Bumping up against NDL would happen at +4 minutes if you were working in imperial and +18 minutes in metric.
I'm beginning to think I'm crazy.
The whole "Measure with a micrometer, mark with chalk, and cut with an axe" trope about deco theory really seems
grossly understated when we are talking about a
14 minute difference in calculated NDL between metric vs imperial version of tables from the same agency, based on the same model.