- Ed Dozier
Do Long Lenses Not Like Filters?
I was trying to solve a mystery about why my trusty Sigma 150-600mm seemed to lose its ability to resolve fine details. I recently got a D500 and was using the Sigma while testing the camera electronic front-curtain shutter feature (to totally eliminate vibrations). I noticed the resolution measurements (using the MTF Mapper program) were much lower than expected.
Had I somehow bumped the lens and knocked the optics out of alignment? Was the camera focus system not functioning properly? Were the phase-detect and contrast-detect focus systems both out of whack? Was my vibration reduction accidentally left on while using a tripod? Am I getting sloppy and don’t even know it?
Other lenses weren’t showing any resolving problems on this camera, but that only helped to show what the problem was not. Out of desperation, I removed the 95mm Hasselblad UV filter from the Sigma. Like magic, the resolution numbers were back to where I expected them to be (more than 20% higher). Are you kidding me? My premium Hasselblad filter is no good?
I don’t have any other lenses that use 95mm filters, and I don’t have any step-up rings that big, either. I finally figured out how I could use my 24-70mm lens with its lens hood to hold the filter via friction-fit inside the hood. I took shots of my resolution target with this filter in place (at 70mm), and then removed it to repeat the shots with no filter. The MTF Mapper program showed zero resolution differences either with or without the Hasselblad filter! How is this possible? How can the Hasselblad show perfection on this lens and wreck the resolution on the Sigma?
I don’t have any other long lenses to experiment with. I don’t think most users of really big glass use filters at all, except for the kind that fit in the drop-in filter holders near the rear of the lens. Maybe there’s a good reason they don’t use front-mounted filters, aside from the big cost and added weight.
Let’s take a look at some resolution test measurements. The following plots compare shots taken with and without the Hasselblad 95mm filter.
Sigma 150-600mm at 600mm with 95mm Hasselblad UV filter. Bad resolution!
Sigma 150-600mm at 600mm without UV filter. Much better resolution.
Note in the above plots that the peak resolution with the filter in place has an MTF50 of about 28 lp/mm. The plot of the photo taken moments later under the same conditions, but without a filter, has an MTF50 peak resolution of about 36 lp/mm. That’s a difference of about 22%. I repeated this same test a dozen times, with the same results. This would lead most people to conclude that the filter is absolutely terrible. But I'm stubborn. I investigated further.
Next, I’ll show you some resolution test results using my Nikkor 24-70mm f/2.8 VR lens.
24-70 at 70mm f/2.8 with Hassleblad 95mm UV filter
24-70 at 70mm f/2.8 without filter
24-70 at 70mm f/2.8 without any filter
24-70 at 70mm f/2.8 with Hassleblad 95mm UV filter
Maybe you can tell a difference between the measurement results with or without the filter, but I sure can’t. Based upon the tests with the 24-70 lens, I would have to conclude that there’s nothing at all wrong with the Hasselblad filter.
I was taught that even a medium-quality filter would have a negligible effect on lens resolution, although it might increase light reflections or decrease light transmission. This is a whole different ball game. I now have a strong suspicion that even top-quality front-mounted filters on really long lenses have a big negative impact on resolution.
Take my experience with this filter/lens combination as a heads-up. I mainly use UV filters as cheap insurance against accidents and because they’re easier to clean than the front lens element. From now on, though, I’m just going to rely on the lens hood to protect this Sigma lens. If I ever get a short focal length lens with a 95mm filter thread, I wouldn’t hesitate to use the Hasselblad filter on it.
For long lenses, though, beware of filters over the front element. At least do some careful testing before you decide to permanently park a filter in front of your big lens. I can’t help but wonder if there are a lot of people out there that are disappointed in their long lens and don’t know that the blame lies in their filter.
