jray
Offline
• • •
Upload & Sell: Off
|
Hmm... it's this thread again 
I have heard all the "Filters degrade your images", and "Don't worry about the scratch on your front element, it won't affect the image." statements time and time again. Well if a scratch on the front element can't be detected in the final image, then I would imagine a piece of ruler flat optical glass shouldn't cause too much of a problem. OK, perhaps a bit of additional flare from time to time, but how much? I've done a few tests to see just how much UV filters really affect image quality, but I still have additional tests to perform before I come to any firm conclusions. In particular I have not been able to test for flare problems since it's been overcast and raining in my area for the past week  .
Below are a few of the test sets I've done so far. They are performed using no filter, and then with a stack of 4 filters to compound any image degradation so it can be more easily detected. One of the filters is a Kenko MC UV filter which is a Hoya HMC filter (not super HMC), and the other three are S+W no-name UV filters ($8.50 each for 77mm). These are from a few projects I am currently working on. I would also like to compare 5 stacked UV filters of the types; Hoya Super HMC, B+W MRC, and Heliopan SH-PMC, but unless I can convince a filter reseller to loan me some, I don't foresee these tests happening very soon. I have Hoya Super HMC filters on my lenses, 3 of which are 77mm, but that's it. I want to test no less than 5 stacked when testing high quality filters.
Anyway, here are images from test set 1 and 3. I'll place more on a web page as soon as the weather and my schedule agree. Note: Images were critically focused using an Angle Finder 'C' at the 2.5x position.
http://www.d30.info/tests/uv_filters/filters_s1.jpg
http://www.d30.info/tests/uv_filters/filters_s3.jpg
The way it looks to me is the primary image degradation is due to a slight contrast reduction, and a bit of color shift, which may or may not be evident when testing with high quality filters. For those whose seem to see wider edges in the tests with filters, make sure you count the number of pixels across a edge band. You'll see that the number is the same for both in almost all cases. It's the contrast difference that seems to create this effect, and remember, three of these filters are the cheapest non-coated filters I could find. I don't imagine that even these would cause much problem when used one at a time. I won't really know until I can conduct more tests.
As a last point... I get all kinds of crap on the front of my lenses (actually filters) when crawling around in wet grass, splashing through marshes and bogs, getting hit by flying mud and small rocks at Motocross events, etc. The last thing I want to do is open up my bag, or sit up to get in my vest to get lens cleaner, or stop shooting long enough to do so at events. I just use the tip of my tee-shirt or other soft rag and get on with things. Life's too short, and I can always buy a new filter if I scratch it. If not, I can move most of my filters to future lenses I purchase. In addition, I don't have a problem with removing an extremely grimy filter and dunking it into a small pan of water to soften any crap that gets into the crevices. I wouldn't want to do this this with any of my lenses.
Anyway, just my 2000 cents...
Edited by jray on Mar 01, 2004 at 08:14 AM GMT
|
