Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
05-12-2015, 07:31 AM   #1
Loyal Site Supporter
Loyal Site Supporter
BigMackCam's Avatar

Join Date: Mar 2010
Location: North-East of England
Posts: 15,215
Always test your filters!

I thought this might prove a useful and cautionary example of how filters can potentially degrade images, and why it's essential that we check anything we put on our lenses before starting a shoot...

I recently treated myself to a new HD DA 70mm Limited. When it arrived a few days ago, I mounted it on my K3 and took a few test shots in the garden and indoors to make sure it was working as expected. Happy that all was well, I put it back in the cute little leatherette pouch until I had some time to try it out properly.

Now, I'm in the habit of fitting multi-coated UV filters to my lenses - both to protect them, and - primarily - to make it easy to clean the front element while I'm out and about. It can get quite windy in my part of the country, and we get our fair share of rain too. Plus, I'm lazy and don't always have an uncontaminted lens cloth to hand. With a UV filter fitted, at a pinch I can breathe on the filter and use my cotton shirt tail or a tissue to clean it, safe in the knowledge that the lens glass and coatings won't get scratched.

Yesterday, I had some errands to run and the weather wasn't too bad, so I put a new UV filter on the DA70, fitted it to the K3 and headed out.

On my way back to the house I stopped at a field where there are some beautiful Appaloosa horses, and took a few shots.

Last night, I imported the photos into LR6 and was troubled by what I saw... On a number of pictures, I could see diagonal lines or bands in some of the out-of-focus areas. The bokeh on the DA70 can be a little busy, but this was something else - obvious striations running from top left to bottom right.

The first attached photo is a resized crop of the picture (beautiful horse, isn't it?), taken wide open at F2.4. The second attachment is a crop of the area just below the horse - can you see the lines I'm talking about?

So, today I took some test shots of my lawn out back, both with and without the UV filter fitted.

Attachment number three is a full size view of an area of the test shot, taken at F2.4 with the UV filter fitted. The diagonal lines in the bokeh are most clearly visible at the centre-top of the image, but can be seen throughout as thin, whispy lines.

The final attachment is the same test shot taken without the filter. See? The lines have gone! In fact, the bokeh looks better in general to my eyes.

I can repeat this again and again, so I'm 100% certain the filter is causing it. I had a cheap non-MC UV filter lying around from a flea-bay lens purchase and tried that on the lens too... Worked perfectly, no lines.

The filter with the problem is a supposedly-genuine Kenko model, labelled "Kenko *Digital Filter MC UV". I accept that it's not a premium quality filter, but not a dirt-cheap unbranded model either.

I'm now going to do A/B checks with all of my filters to see if any others cause obvious IQ degradation...

On a related note, I'll admit to being just a little underwhelmed initially by the HD DA70's bokeh. It's not objectionable - in fact, by F4.0 it's quite nice - but wide open it's not what I'd describe as smooth and creamy - it has something of a "brush-stroke" effect. That said, the DA70's a lovely piece of kit - sharp, and with truly beautiful colour rendition. A keeper, for sure

Attached Images
       

Last edited by BigMackCam; 05-12-2015 at 08:04 AM.
05-12-2015, 08:48 AM   #2
Loyal Site Supporter
Loyal Site Supporter
Madaboutpix's Avatar

Join Date: Dec 2011
Location: North Rhine-Westphalia
Posts: 921
QuoteOriginally posted by BigMackCam Quote
With a UV filter fitted, at a pinch I can breathe on the filter and use my cotton shirt tail or a tissue to clean it, safe in the knowledge that the lens glass and coatings won't get scratched.

Your front element, of the lens that is, may be safe this way. However, high-quality filters usually come multi- or even nano-coated as well, and if you use your shirt tail or tissues on them, you are likely to leave smears and lint on them, or can even scratch them. (Since the advent of microfiber optics cloths, our lives have got a lot easier, and I always make sure that I carry a clean one in my bag.)

That's why I rarely use filters, unless I want to create a special effect. For instance, an ND filter to force a longer shutter speed for motion blur. Otherwise, I see little merit in defeating all the work lens designers and manufacturers put into giving us the best possible IQ at a given price. As your sample images would seem to indicate, that may even apply to bokeh. Thanks for posting those.
05-12-2015, 09:00 AM   #3
Site Supporter
Site Supporter
Stavri's Avatar

Join Date: Aug 2014
Location: at a Bean & Leaf
Photos: Albums
Posts: 1,757
I have a couple of high quality Hoya Pro1 and B+W filters which actively swap between my lenses during use, the rest of my filters are used for protection only. It's always useful to get a first hand account of what brand of filters work best.
05-12-2015, 09:11 AM   #4
Pentaxian
lightbox's Avatar

Join Date: Mar 2015
Posts: 599
It's interesting to see how filters affect the image. I expected a softer image with the filter attached, but it looks "sharper," for lack of a better word.

I'm having a little trouble discerning if the lines you mention aren't in fact some kind of illusion or pattern produced by the blades of grass. Are you able to reproduce the effect on other subjects (ones without a pattern)? Also, any comparisons at other apertures would be interesting.

05-12-2015, 09:20 AM   #5
Loyal Site Supporter
Loyal Site Supporter
BigMackCam's Avatar

Join Date: Mar 2010
Location: North-East of England
Posts: 15,215
Original Poster
QuoteOriginally posted by Madaboutpix Quote
Your front element, of the lens that is, may be safe this way. However, high-quality filters usually come multi- or even nano-coated as well, and if you use your shirt tail or tissues on them, you are likely to leave smears and lint on them, or can even scratch them. (Since the advent of microfiber optics cloths, our lives have got a lot easier, and I always make sure that I carry a clean one in my bag.)

That's why I rarely use filters, unless I want to create a special effect. For instance, an ND filter to force a longer shutter speed for motion blur. Otherwise, I see little merit in defeating all the work lens designers and manufacturers put into giving us the best possible IQ at a given price. As your sample images would seem to indicate, that may even apply to bokeh. Thanks for posting those.
Hi. Thanks for the response - of course, you're correct. But, I've said it before, and I'll say it again - we all have different priorities. I don't make a habit of using my shirt tail to clean the face of my UV filters (like you, I prefer to use a microfibre cloth) but I will do and have done so on occasion, and I've never been unable to remove smears. I realise I could scratch the multi-coating on a filter, but it's just a filter - the lens will be fine. Interestingly, I see no degradation in IQ from even my most heavily abused UV filters. But, that's just me!
05-12-2015, 09:41 AM   #6
Forum Member
Multi-coated's Avatar

Join Date: Jan 2013
Photos: Albums
Posts: 95
I had a somewhat similar experience with a brand new mid-range filter a few years ago. I think it was a Kenko. Definite IQ degradation with filter. Out of curiosity I passed the filter off to a neighbor who is a pro photog. He ran test shots and thought the filter substandard. Latter I pass the filter off to another neighbor who is an engineer. He brought to the filter to work where he and co-worker ran some sort of tests for kicks, don't ask me what, but they too thought the filter crap. They all thought the filter probably not a counterfeit since the coating quite good. The problem was the glass itself. Badly flawed according to the engineers.

So it seems it was just my bad luck that I bought a flawed filter that somehow made it past QC. Perhaps the same for you?

Looking back I think the four of us experienced the perfect storm of too much time on our hands. :-)
05-12-2015, 09:56 AM   #7
Veteran Member
LFLee's Avatar

Join Date: Aug 2009
Location: Southern California
Photos: Gallery | Albums
Posts: 2,292
this might not be filter problem....

my filter is pretty dirty but don't see them show up in picture, especially you are using pretty 'wide' aperture.
I only clean my filter it they are really dirty and when spots show up when I shot directly into sun...
05-12-2015, 09:59 AM   #8
Loyal Site Supporter
Loyal Site Supporter
BigMackCam's Avatar

Join Date: Mar 2010
Location: North-East of England
Posts: 15,215
Original Poster
QuoteOriginally posted by lightbox Quote
It's interesting to see how filters affect the image. I expected a softer image with the filter attached, but it looks "sharper," for lack of a better word.

I'm having a little trouble discerning if the lines you mention aren't in fact some kind of illusion or pattern produced by the blades of grass. Are you able to reproduce the effect on other subjects (ones without a pattern)? Also, any comparisons at other apertures would be interesting.
Yes, I noticed that the image looks sharper with the filter. At first I put it down to a slight difference in focus between the test shots, but in fact it's consistent for all the "with filter" shots versus those without. Perhaps the fine lines in the bokeh are giving an impression of sharpness that isn't there? Or perhaps some of the light is somehow being focused as if through a smaller aperture by the defect in the filter?

The lines aren't a pattern produced by the grass, in my opinion. Firstly, they only run from top left to bottom right, and the angle of them appears identical between the horse shot and that of my lawn. These were taken at different angles to the prevailing north-westerly wind we get here, so the grass wasn't leaning in the same direction on both examples (if that makes sense).

I see the same effect on shots taken at F4, and to a lesser extent as far as F6.3 - after that, it becomes more difficult to see (at least, based on the photos of my lawn).

See attached two more full size crops of the same test photos, at F4 - the first one taken with the offending filter, second one without. The slight difference in colour rendition may be due to the sun peeking through the cloud at the time I took the shot with filter; this auto-exposed at 1/1000, whereas the one without was at 1/500. Either way, you can clearly see the diagonal lines in the version taken with the filter...

---------- Post added 05-12-2015 at 06:02 PM ----------

QuoteOriginally posted by LFLee Quote
this might not be filter problem....

my filter is pretty dirty but don't see them show up in picture, especially you are using pretty 'wide' aperture.
I only clean my filter it they are really dirty and when spots show up when I shot directly into sun...
Understood... but the problem doesn't occur without the filter, only with... quid pro quo, the filter is causing the problem (unless I'm missing something, which is entirely possible!!)

Attached Images
   
05-12-2015, 10:15 AM   #9
Pentaxian
Ontarian50's Avatar

Join Date: Mar 2015
Posts: 514
I'd like to see the science

Obviously we all want our protective filters to be as invisible as possible. Clean glass, good coatings, etc. etc.
But I recall a test published in Shutterbug many moons ago where a fellow put his filters to a scientific test. He photographed a test target from a tripod, with and without various filters. Even the acrylic (drop in) filters had no effect on his resolution, but when he put a seriously dirty filter on his lens, he saw a loss.
I have come across budget polarizers that were made from unflat glass - presumably cut from sheets, rather than cut and ground from a block of optical glass. The slight optical warp made a mess of the autofocusing.
Today, I think the biggest issue with filters is ghosting. The better "digital" filters do a better job of suppressing reflections off the sensor (which then travel back through the lens, reflect off the flat rear filter surface, and then back toward the sensor). It's mostly only visible with really bright highlights - but if bright lights are obvious, then presumably less bright things are making less obvious ghosts.
I photographed my son's birthday party the other day, and I had a vintage Pentax filter on my 50mm 1.4-A. Looking at the images later, I saw green spots near his face - the birthday cake candles had reflected back off the filter. While the old Pentax filter was multicoated, it apparently isn't up to current standards.
I have tested various grades of UV/lens protection filters, and sure enough, the more costly ones show less ghosting. To get rid of the ghosting entirely, take the filter off.
That said, I've seen many, many lenses saved by the use of a protection filter. It sucks to scratch, crack or ding a good quality filter, but if it saves costly repairs to the lens behind it, it was worth it.
05-12-2015, 10:21 AM   #10
Veteran Member
Sagitta's Avatar

Join Date: Feb 2011
Location: Maine
Photos: Gallery
Posts: 2,949
Your filter is most likely causing a loss of light, forcing the camera to compensate. I'm willing to bet your EXIF is going to show a change of some sort whether in aperture, ISO or shutter speed. A smaller aperture will introduce sharper bokeh, a slower shutter speed more blur, etc etc.

If all things are kept the same (shutter speed, ISO, aperture) then the shot with the filter will almost be guaranteed to underexpose just a bit more than the other.

As an aside, unless you're enlarging these photos to a very large size and printing them out, this is (mostly) a non-issue anyway because whatever minor flaws may be getting introduced will be almost unnoticeable in the final print. We're in pure pixel peeping territory here.
05-12-2015, 10:58 AM   #11
Loyal Site Supporter
Loyal Site Supporter
BigMackCam's Avatar

Join Date: Mar 2010
Location: North-East of England
Posts: 15,215
Original Poster
@Ontarian50 - that's very interesting. I'd really like to find a brand and level of UV filter that provides a balance between price (as I'm hard on my filters) and performance. As I mentioned, I had a cheap uncoated filter lying around that worked just fine in this test, but I'm sure that same filter would be at risk of ghosting and won't be as efficient at passing light through to the lens. I also agree that to get the best from the lens, take the filter off. For me, I'm more comfortable working with a UV filter in place (as long as it doesn't materially degrade the IQ, of course!!)

QuoteOriginally posted by Sagitta Quote
Your filter is most likely causing a loss of light, forcing the camera to compensate. I'm willing to bet your EXIF is going to show a change of some sort whether in aperture, ISO or shutter speed. A smaller aperture will introduce sharper bokeh, a slower shutter speed more blur, etc etc.

If all things are kept the same (shutter speed, ISO, aperture) then the shot with the filter will almost be guaranteed to underexpose just a bit more than the other.

As an aside, unless you're enlarging these photos to a very large size and printing them out, this is (mostly) a non-issue anyway because whatever minor flaws may be getting introduced will be almost unnoticeable in the final print. We're in pure pixel peeping territory here.
Well, I'm shooting in aperture priority mode, so the aperture remains fixed at where I set it, as does the ISO. Exposure does change, yes - however in those last two shots I posted, the shot with the filter exposed at 1/1000, while the one without exposed at 1/500. That's due to lighting at the precise moment I took the shots. Even so, a static shot hand-held at 1/500 with a 70mm lens, wouldn't show any appreciable blur. I'm willing to counter-bet that if I set the camera to full manual, same ISO / aperture / shutter speed, I'll still get these "bokeh lines" in the image taken with the apparently-flawed filter, and no lines without the filter. As for the last point re pixel peeping... In LR6, I saw an unnatural look to the bokeh with the horse picture at less than 1:1 zoom - which is why I started to investigate further. I only clearly saw the nature of the problem when zoomed in at 1:1, but I could tell something wasn't right at less magnification. However, you're right - at 1:2 or less, this isn't really noticeable...
05-12-2015, 11:07 AM   #12
osv
Veteran Member




Join Date: Mar 2007
Location: So Cal
Photos: Gallery
Posts: 2,080
QuoteOriginally posted by BigMackCam Quote
Well, I'm shooting in aperture priority mode, so the aperture remains fixed at where I set it, as does the ISO. Exposure does change, yes - however in those last two shots I posted, the shot with the filter exposed at 1/1000, while the one without exposed at 1/500.
that shutter speed should be high enough to prevent it from being a factor, but there appear to be some pretty big differences in the way that the lens is focused, in those last two pics.

i have some cheap uv filters that are made out of plastic, not glass... take 'em out of the threaded housing, and they can be bent way over.
05-12-2015, 12:13 PM   #13
Loyal Site Supporter
Loyal Site Supporter
BigMackCam's Avatar

Join Date: Mar 2010
Location: North-East of England
Posts: 15,215
Original Poster
QuoteOriginally posted by osv Quote
that shutter speed should be high enough to prevent it from being a factor, but there appear to be some pretty big differences in the way that the lens is focused, in those last two pics.

i have some cheap uv filters that are made out of plastic, not glass... take 'em out of the threaded housing, and they can be bent way over.
OK, I agree re focus. So I just took another A/B test shot. Here's my process:

1. Fit filter to lens
2. Set camera to Av, aperture F4, ISO 200 (a shutter speed of 1/125 was automatically selected)
3. Use AF to focus on reference point on lawn and take the shot
4. Switch to MF and M mode
5. Carefully remove filter without touching focus ring (was VERY careful here)
6. Set shutter speed to 1/125, aperture F4, ISO 200
7. Centre on the same reference point without re-focusing and take next shot

Both attached photos show the exact same area at 1:1 zoom. Look in the top third of the cropped images... Notice how in the first shot (taken WITH filter attached), those diagonal lines are visible... but in the second shot (WITHOUT filter), no lines... I did this a second time to ensure I got consistent results. Yup, same results.

And again, despite exactly the same focus, the shot WITH the filter looks as if it is sharper than the one WITHOUT. No difference in focus. None.

As for the plastic filters - oh dear. I'll bet at least a couple of my very early UV filters are just as bad!!
Attached Images
   
05-12-2015, 01:00 PM   #14
Pentaxian
Na Horuk's Avatar

Join Date: Feb 2012
Location: Slovenia, probably
Photos: Gallery | Albums
Posts: 11,183
Almost reminds me of the crosshatch filter that I have, which affects bokeh and gives it a checkered texture.
I read a couple times online that people bought a good filter online, but ended up getting a fake, sometimes in an opened genuine case, other times in a case that is very similar, but not the same as original brand. Anyway, I hope you get things sorted out. This is why so many people are against UV filters on digital cameras, as they can only degrade the IQ. There are now brands selling "protection" filters, but I don't know how effective and protective those actually are.
05-12-2015, 01:16 PM   #15
Loyal Site Supporter
Loyal Site Supporter
BigMackCam's Avatar

Join Date: Mar 2010
Location: North-East of England
Posts: 15,215
Original Poster
QuoteOriginally posted by Na Horuk Quote
Anyway, I hope you get things sorted out. This is why so many people are against UV filters on digital cameras, as they can only degrade the IQ. There are now brands selling "protection" filters, but I don't know how effective and protective those actually are.
Thanks I'm just happy that the lens isn't to blame. I'll mark the bad filter and put it away so I can show other photographers if the subject arises. Then, over the next few days, I'll test all my other filters and ditch those that have any obvious impact on IQ. To be honest, this is only the second filter I've ever had problems with - the other looked fine, but caused AF and contrast issues. That was a long time ago in my Nikon days, and the filter went straight into the bin!!
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
angle, attachment, bokeh, camera, da70, f2.4, f4, filter, filters, grass, image, k-mount, lawn, lens, lines, model, pattern, pentax lens, photo, photos, shot, shots, slr lens, test, uv
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
B&H: How to Test Your Lens interested_observer Pentax SLR Lens Discussion 2 02-14-2015 11:36 AM
New IIs; What's your test checklist? fewayne Pentax K-5 16 05-19-2013 05:46 PM
Reminder: Always mount your lens hoods! L33tGreg Pentax SLR Lens Discussion 6 03-15-2012 04:00 PM
Streets Always wear your sunglasses KarenH Post Your Photos! 23 02-17-2011 02:14 AM
do you always straighten your photos in PP? aleonx3 Digital Processing, Software, and Printing 12 03-19-2010 04:55 PM



All times are GMT -7. The time now is 03:37 PM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top