Screamin Scott wrote:
How do they work with lenses that have plastic filter mounts ? (many of the kit lenses do)
Actually, virtually all Canon lenses from kit through L-series now use plastic filter threads, so binding is rarely an issue.
However, it can't hurt to have brass mount filters, in case you ever need to use them on an aluminum barrel (that's where thread galling is the most likely, when it's aluminum to aluminum).
I would NEVER use anything magnetic around a digital camera.... I suspect it would be very easy to corrupt any memory card that got near the magnet. Heck, I've seen credit cards and similar, with their magnetic stripes, corrupted by being carried in the same pocket as a cell phone!
"Universal" rectangular filter systems such as Lee are great for certain things... but are uncoated or single coated optical plastic that's easily scratched, plus can be difficult to shade very well. I prefer to use round, screw-in filters as much as possible, that can be fitted under the lens' matching hood.
OP, I mostly use the B+W MRC or Kaësemann C-Pols. They are excellent quality Schott glass, mounted in brass and multi-coated to minimize flare issues. The Kaësemann is further sealed against moisture.
For use on a 24mm lens, with B+W you most likely would
not need a more expensive "slim"version. B+W's standard mounting rings are pretty narrow to begin with. I've used them on 16mm, 17mm and 20mm wide lenses on full frame (10mm wide on crop cameras) without any vignetting issues at all. Some "slim" filters also don't have front threads, so cannot be capped with a typical lens cap. A special cap is sometimes available for use with them.
I've also used Hoya HMC, SHMC and HD/HD2... and found them to be excellent quality, too.
And I've heard a lot of good reports about Marumi, though I've never yet used them.
Heliopan, Nikon, Zeiss, Singh-Ray are more top quality filter manufacturers. Vivitar and Tiffen have offered some good ones, too.
However all manufacturers offer a variety of grades. I know B+W offers 4 or 5 grades of C-Pol and Hoya offers f or 6... at different price ranges.
Many "name brands" don't actually make their own filters at all... they just buy them from someone else and have their name printed on them. Vivitar has always been one of the largest "rebranders" in the photo industry. Kodak and Polaroid do a lot of this, too. But Vivitar practically invented the practice beginning in the 1960s and 1970s. For the past 30 years or more, nearly everything sold under the Vivitar name has been outsourced. Not that this is necessarily a bad thing. Outsourced items can be quite good, too.
Sigma appears to relabel filters, too. I suspect Canon does as well (and all Canon filters appear to be single coated, which would make them very overpriced). Don't know, but wouldn't be surprised if Nikon does this too. Usually camera and lens manufacturers don't make their own filters.
To get a good C-Pol, look for high quality glass and a neutral filter that won't add a tint to your images. I'd also recommend multi-coatings, mainly to minimize flare issues (some of the more recent one such as Hoya's HD2 and B+W's "Nano" are more scratch resistant as well as easier to clean). And don't be too cheap. I don't know, but would strongly suspect a $20 C-Pol would best be used for skeet shooting or as an air hockey puck or a coaster to keep your ice tea from sweating on your nice wood coffee table. You generally "get what you pay for".
EDIT:
Quote:
http://www.lenstip.com/115.4-article-Polarizing_filters_test_Results_and_summary.html
I take the above comparison with a grain of salt. Notice how the B+W "standard" Kaësemann fails miserably, while the B+W "slim" KSM is rated the best? Reading that closely, the author/tester mentions that the "standard" KSM they tested turned out to be a linear polarizer that somehow was mislabeled, a very obvious error. No wonder it failed!
The B+W standard and slim filters are the same, other than their mounting frames, so they should have give identical test results (except the "slim" would lose some points in this particular comparison... for being pricier). A reputable tester would have discussed with the manufacturer and gotten a replacement for the obviously mislabeled filter... or at least would have redone the tests and updated the info sometime in the years since that test was initially done. It's almost as if they want to keep the bad data from the mislabeled filter in the mix, in order to skew the results.