Reliability
Where calculation reliability stands
VoLCA is compared with results published by database producers, using the same source data and the same method, in the implementation each comparison names.
Every release is measured four ways, and every measurement is published, the ones that disagree included: what does not agree is shown for the same reason as what does, so that anyone can see where the engine stands.
Does VoLCA find the results database producers publish?
Every impact category of a publisher’s table, scored by VoLCA from the same source data and the same method, with the verdict a release is held to.
Published results →Did a new engine version change a result?
The same published results, compared against every engine version in turn.
Engine history →Do two implementations of one method give the same score?
One database scored under the implementations several publishers make of one method, two at a time.
Implementations of one method →What does it cost to read a database and to score it?
Loading time, scoring time and peak memory for each engine release, measured from a deleted cache.
Time and memory →