I've recently begun using a Sekonic 558, mainly for spot-metering landscapes when shooting slides and B&W print film (I use Canon FD SLRs). As an interesting test, I thought I'd try metering through my filters to see if the readings agreed with the published filter factors from the manufacturers.

My filters are not particularly high quality - all being Cokin P or HiTech (non-pro) 85mm gels.

I found quite a few discrepancies which I'm not sure if I should attribute to manufacturing/publishing variations or perhaps varying sensitivities to certain wavelengths in the Sekonic itself.

The results (EV published | EV measured):

* Cokin Red - 3 1/3 | 2 1/2
* Cokin Orange - 1 2/3 | 1 2/3
* Cokin Yellow - 2/3 | 2/3
* Cokin Green - 2 2/3 | 2 1/2
* Cokin 120 G1 Grad - 1 2/3 | 1 2/3
* Cokin 121L G2 Grad - 1 | 1
* Cokin 121M G2 Grad - 2 | 1 2/3
* Cokin 121S G2 Grad - 3 | 2 2/3
* Cokin 81B Warming - 1/3 | 1/3
* Cokin 82B Cooling - 2/3 | 2/3
* Cokin Linear Polarizer - 1 2/3 | 1 2/3
* HiTech 47B Blue - 2 2/3 | 4
* HiTech ND 0.6 - 2 | 2
* HiTech ND 0.9 - 3 | 2 1/2
* HiTech ND 3.0 - 10 | 5 1/2

In other words, who do I believe? I don't have any film back that was metered with the Sekonic but I did a couple of digital tests on the HiTech NDs and it seems the in-camera meter on my NEX-5 measures something close to the published spec.