Hi,
I set up an "active" (first) metric g and a "frozen" (second) metric a on the tangent space of a single manifold. Then formed the raised covariant derivative AD[c][v[b]] of a vector, which is the lowered compatible covariant derivative of the frozen metric a, then raised by the active metric g. Then applied SeparateMetric[g] to this, as well as SeparateMetric[a], and got two different results that don't agree with each other. A notebook giving a side-by-side comparison of the results is attached.
I'm an xAct newby, but I thought that the results should be equal in this case (they may initially look different, but would be found to be equal when manipulated in a side-by-side comparison). Am I misinterpreting this?
If they should be equal, perhaps just like ContractMetric only working for the active metric, should SeparateMetric have the same limitation to avoid this problem?
Regards,
Bill