More truth about UV/Protection filters. A comparison challenge. Step up

[LightSpeed]

After the heated debate, I decided to test the scenario and challenge some of the big timers, in a friendly manner, to pick the correct image. It was late when I got home so I didn't have much time to devote to this. I will probably end up doing it again. This was sort of on short notice so bear with me.

Both images were edited identically. One of the images was captured with no UV filter. The other image was captured using a Marumi DHG ( digital high grade ) UV filter. Both shots were hand held and taken at slightly different angles. I apologize for not having time to put a tripod on this. I did notice some things before the edit, but I'll get into that later. I noticed no difference in color from one to the other. I did use a flash on this. Now here is your chance to debunk UV filters and their usefulness. Make your pick, and then give an explanation as to why you picked number one or number 2. Let try to keep it clean and gentlemanly. Thank you for your time and participation, beforehand.

Image 1

Image 2

 

[Ron Evers]

Sorry, but being hand held/different angle/distance negates the objectivity of the experiment.

You must eliminate as many variables as possible for this to be a reasonable test.

 

[Light Artisan]

I agree.

Also, edited? Should be SOC in my opinion...

 

[mrpink]

It was late when I got home so I didn't have much time to devote to this. I will probably end up doing it again. This was sort of on short notice so bear with me.

Why don't you get back to us when YOU care.










p!nK

 

[Josh66]

For the re-do, I would maybe take a picture of something where there is actually a chance of UV filter induced defects showing up. Maybe the lights on your Christmas tree (no offense if you don't celebrate Christmas). Now, I'm not saying they should fill the frame, but at least visible somewhere in it. That's the first thing that comes to my mind anyway...

Something with a point light source in it...

(I would do it, but I don't have a UV filter...lol)

I would be interested in seeing that though.

 

[Ron Evers]

Why don't you get back to us when YOU care.

p!nK

I think that was an unfair statement Pink.

The OP cares to make a point but was premature in thinking the test out. I for one would like to see the result of a controlled test.

 

[Bitter Jeweler]

Really?

 

[tirediron]

Some people UV, some people don't UV - can't we just let this die?????

 

[Bitter Jeweler]

Why? Cuz somebody has to win the battle!

:roll:

 

[mrpink]

Why don't you get back to us when YOU care.

p!nK

I think that was an unfair statement Pink.

The OP cares to make a point but was premature in thinking the test out. I for one would like to see the result of a controlled test.

Cares enough to try and make a point, but not enough to perform and submit a legitimate test.... he must work in government.

I think the statement was spot on.

Also, I think this dead horse has been beaten enough.








p!nK

 

[minpingurl1]

Im going with image 1 has the filter. I see some blurring on some of the petals.

 

[Josh66]

I see some blurring on some of the petals.

I see that too (left side), but due to the uncontrolled nature of the test shot, it's kinda hard to say what caused that.

 

[phiya]

Ford or Chevy?
Mac or PC?
Cats or Dogs?

Need I say more?

 

[Josh66]

Ford or Chevy?
Mac or PC?
Cats or Dogs?

Need I say more?

Not quite the same... If we were talking Nikon or Canon - yes, I would agree.

I think there is a chance that under the right circumstances, a UV filter WILL cause undesirable things to happen. You can't say the same for using Nikon as opposed to Canon, for example.

 

[pgriz]

Let’s review what we’re talking/debating about.
1) UV filters protect the lens against dust, scratches, and other possible damage.
2) UV filters improve the picture by filtering out UV light.
3) UV filters diminish the quality of the picture. Possible effects:
a) less sharpness (due to quality of glass, mounting, etc.)
b) more glare (due to poor anti-reflection coating, etc)
c) Focus shift due to another layer of glass being in front of the lens.

Each assertion above can be tested. However, any test will have to, as Ron has pointed out, minimize changing of any variables other than those being tested. So standard test conditions require a tripod (eliminate possible unsharpness issues due to camera shake), mirror lockup, remote trigger, manual focus, consistent light source, detailed and non-moving subject, same shutter and f/stop, etc.

Each test should have at least two shots, preferably without any processing, but if processing is required, then at least the exact same processing steps, which can be reproduced by others.

Furthermore, since the debate also touched upon the “quality” of the UV filters, we should be able to identify the filter brands on the low-end of the quality scale, and those at the top/high end of the quality scale. Since even with good stuff, individual variations exist, we also need to test at least two examples of the same filter to be sure that individual defects are not contaminating the overall results.

Doing controlled tests is not straightforward. On the other hand, the potential benefits of knowledge of a well-designed and well-executed controlled test, are very high. Some of the key elements to have a controlled test involve:
1) Publication of the exact test setup, so that others can carry out the test and verify the results.
2) The testing protocol is published.
3) The testing sensors/tools are clearly identified.
4) The testing criteria are clearly identified.
Because this level of rigor is difficult for most people to achieve, tests that are done informally do not really prove much, as too many things not controlled by the test, can be contributing to the end results.