Eons ago, a new feature came to SLR cameras which caused then-common linear polarizing filters to fail. I think the problem related to the partially-silvered mirrors, which allowed light to pass through the mirror for light metering, but it was decades ago.
So I thought I’d ask: If mirrorless cameras no longer have partially silvered mirrors, can linear polarizing filters be used once more? As I recall, they were more effective and less expensive than circular polarizers.
Sorry none of the Physics experts came along to give you a useful answer!
A Google search revealed a lot of not very helpful articles on the subject. This one seems pretty good, however.
The quick answer is that, as you say, while modern mirrorless cameras, unlike recent SLR cameras, will usually work with a linear filter, these are currently difficult to come by and not nowadays significantly cheaper than circularly polarizing filters.
So the message is to stick with circularly polarizing filters — until some technological revolution reverses the accuracy of this advice!
David
I was confident that the metering would be fine, because you only need to meter with the sensor what the sensor can actually see and record. What I did wonder about, however, was if there was some kind of interaction between AA filters and polarizers that might be different with linear ones.
Somewhat related, I've always thought that perhaps the best way of getting rid of glare on smooth surfaces would be to have a motorized, remote-control polarizing filter that took the same static photo several times as the filter rotated, and then did a "minimum" Z projection - take the darkest pixel for any image position, from the numerous frames (and for a flat-color special effect, a maximum could be done). I wonder if one type of filter would be better than the other for this?