Camera Labs

Opinions on UV filters diverge a lot between photographers.
Some use them as protection for the front element, saying they won't affect IQ.
Some don't use them, saying IQ is lowered due to the "cheap" glass used on the filters.
Some use 10 dollars filters. Others use 60 dollars filters.
But how does an 10 dollars UV compares to an 60 dollars UV? Will they degrade IQ? I decided to test them today.

Test Equipment:
- Canon 40D
- 100mm f/2.8 Macro
- Hoya UV Standard filter (around 10 dollars. Got for free on the store I buy my lenses)
- Hoya PRO1 UV Filter multi coated (around 60 dollars in Brazil)
- Hoya Circular PL box as subject

How the test was performed:
- I decided to get my sharper lens to do this, but use it at its softest: f/2.8 @ MFD.

- I left the lens AF to the subject, then switched it to MF. That would avoid any focus change between the shots.

- Camera in Manual Mode, 1/20s, f/2.8, timer 2s. Same settings for every shot.

- No PP done to the pictures besides cropping to 100% and converting to JPEG with the same settings trough DPP.

Pictures:





Top is with no filter.
Center is with the cheap filter.
Bottom is with the Pro1 filter.

Here are the links to the 100% crops. I recommend you to open in different tabs so you can compare between:

No filter:
http://farm4.static.flickr.com/3330/348 ... 6ca4_o.jpg

Cheap:
http://farm4.static.flickr.com/3375/348 ... de20_o.jpg

Pro1:
http://farm4.static.flickr.com/3305/348 ... dcbc_o.jpg


Conclusion:


I can see pretty much difference in sharpness when using no filter. For my eyes, the image is MUCH more sharp without a filter.
I can't see much difference between an 10 and an 60 dollars filter tough.

If you think this test isn't worthy and have an suggestion of how I should perform it better, I'm open to suggestions.

_________________
Canon 40D and lots of lenses.
my Flickr

I think the pro1 is a bit sharper. It's also much better with CA in my opinion.

Wow. While I'd expect a filter to possibly have a negative impact, I wouldn't have thought it would be so big.

I suspect the cheap filter shot has a slight amount of vertical motion blur, but otherwise it is practically identical to expensive filter shot.

I would be interested to see this test repeated at optimal aperture (f/5.6 to 8?). That might negate possible focus errors.

Actually, I might even do this myself later...

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

I should make some tests with mine 40D and 17-40 L with mine B&W UV Haze... I am surprised how much different are shots with and without filter...

_________________
Canon 40D, 17-40/4 L, B+W UV normal; canon-50/1.4

Hey, Popo! Well... it can't be motion blur since both the camera and the subject were sitting on the table. Only if it was due to the shutter actuation.
If you see, the left side of the pics are softer than the right. That's due to misalignment of the subject, so DoF related.

I performed the tests @ f/5.6 as you said. The results were much, much better. The thing is that I wanted it to be closer @ f/2.8, since is the aperture I use the most in that lens.

Here you go:





Top: No filter
Middle: Cheap
Bottom: Pro1

Links to the 100% Crops:

No filter:
http://farm4.static.flickr.com/3408/348 ... 67cb_o.jpg

Cheap:
http://farm4.static.flickr.com/3405/348 ... 2298_o.jpg

Pro1:
http://farm4.static.flickr.com/3616/348 ... 64ae_o.jpg

_________________
Canon 40D and lots of lenses.
my Flickr

Whatever the cause, there is a slight amount of extra vertical blur in the original cheap filter result, which I think is most likely due to some sort of vibration. But overall it isn't significant, just seriously pixel peeping.

The new results are much closer, but here I think both filtered results still look fractionally different than without. Can't quite put my finger on it though.

I have to wonder if the presence of the filter might affect the AF accuracy at all.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Yep. I thought about it too. I used the lens in MF so the focus was in the exact same position for the shots WITH and WITHOUT filters.. so, maybe adding an element the focus could change?

Any idea for me to test it? Because I think if I let the camera AF for each shot, even the smallest difference in the AF calculations could lead to an different result. Maybe MF with Live View? What do you think?

_________________
Canon 40D and lots of lenses.
my Flickr

Thinking more, while the glass in theory will affect the light path, if it will be significant or not I'm not sure. So I think you're right, it might affect the focus position if you don't compensate for it. In theory AF would be able to take that into account. In part that's why I was interested in smaller aperture tests too.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

Doing same sorta thing here. Filters are:
Hoya HD UV (58mm) - pretty expensive, supposedly using high strength glass so able to be made thinner than normal.
Hama UV 390 (62mm) - low end brand. Claims to be coated.
Jessops UV (72mm) - I didn't get this from new (came with a used lens) so I don't know any claims on it. It doesn't look like it has any coating.

As the filters are different sizes, I used adapter rings to fit them.


None


Hoya HD UV


Hama UV 390 coated


Jessops UV

Above are 100% crops. Sony A350 with Tamron 90mm macro at f/5.6. Taken using remote release on tripod. Camera AF from infinity each time. Target is a Tamron lens box. Looking at the above I can't be 100% sure the AF was picking the same distance each time.

How do the filters look transmitting light?



The Hama filter has an obvious but slight yellow cast. The hoya might have a slight cast but hard to tell. The Jessops is comparatively clear.

What about reflections?


Windows at back of shot.


Windows to left side of shot.


Windows behind shot.

At shallow angles, all reflect light, with the Hoya being the only one to introduce a purple cast. As the angle is increased the Hoya shifts to a green cast but the intensity reduces significantly. The Hama might have a small advantage over the Jessops.

_________________
Canon DSLRs: 7D, 5D2, 1D, 600D, 450D full spectrum, 300D IR mod
Lenses: EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 28-80 V, 70-300L, 100-400L, TS-E 24/3.5L, MP-E 65, EF-S 15-85 IS
3rd party: Zeiss 2/50 makro, Samyang 8mm fisheye, Sigma 150 macro, 120-300 f/2.8 OS, Celestron 1325/13
Tinies: Sony HX9V.

personally I'd like to see that shutter speed at 1/125 at a minimum even on a tripod...........take out as many of the variables as possible.

Why? If the camera is over the table with no hand contact, the shutter speed shouldn't be an issue..

_________________
Canon 40D and lots of lenses.
my Flickr

Table could move........vibrations form passing vehicles, trains whatever. Higher shutter speed removes those possible variables. Just as you removed focusing variables by conducting another test at a smaller aperture.

Just my opinion.

Okie Dokie... so tomorrow when the sun's back I'll take a few more shots

_________________
Canon 40D and lots of lenses.
my Flickr

I found this post interesting and decided to do it my self. Now it is not ideal conditions but here is what I did. Put a light close to a box of raisins and used a tripod with shutter release. Each time I took the picture I focused the camera to infinity and let the camera focus each time and did five of each and looked for the best one.

I used the HOYA HMC 77mm. The lens was a Canon 70-200mm f/2.8 L (non-IS)

Frist is WITH FILTER 1/640 f/4 1600 ISO 200mm




Second is WITHOUT FILTER interestingly 1/800 f/4 1600 ISO 200mm




Now it is hard to see the difference but I could tell a very slight difference between the two. Also I know ISO 1600 is not good to do this test since these are 100% crops so I might have to try in the sun light with ISO of 100 to see if there is any difference but I will have to wait for the sun to show its self.

_________________
Canon 350 : 70-200mm f/2.8 L : 400mm f/5.6 L : 18-55mm f/3.5-5.6 : 430EX II : Canon 1.4x II :

Even if there is a slight improvement in the image quality between the two filters (i honestly cant see the difference in the comparisons seen here, except for the initial post), that doesn't give me a good enough reason to fork out 60 bucks (about AU$100 atm) for the "high quality" product.