Distance to subject for lens test

I just bought a GX9 with 14-140 and would like to test the lens sharpness (I know, the GX10 could be coming, but I expect it to take a while and not have major IQ upgrades and the GAS was too strong).

What would be a good distance from the camera to the test subject (e.g. a brick wall) to test? I looked at a few video's, but the distances used were not discussed, but seemed relatively close, like a few meters. Wouldn't that impact the corner sharpness to much because the distance from the corners (left and right of the subject) to the camera vs middle of the subject to the camera differs somewhat? (relatively speaking and especially at the shortest FL)?

Thanks for the help!

Chris

ChrisFB said:

I just bought a GX9 with 14-140 and would like to test the lens sharpness (I know, the GX10 could be coming, but I expect it to take a while and not have major IQ upgrades and the GAS was too strong).

What would be a good distance from the camera to the test subject (e.g. a brick wall) to test? I looked at a few video's, but the distances used were not discussed, but seemed relatively close, like a few meters. Wouldn't that impact the corner sharpness to much because the distance from the corners (left and right of the subject) to the camera vs middle of the subject to the camera differs somewhat? (relatively speaking and especially at the shortest FL)?

Thanks for the help!

Chris

Click to expand...

Most lenses can focus to at least 1 m, and not a few have close focus distances much closer than that.

At the stated close focus distance, a lens is typically still has decent resolution, by observation the optics are dealing with the difference in distances.

There is also a prctical issue.

With your lens at 140mm I'm sure you can find a brick wall big enough that you can back up 20m and still have the wall cover the field of view.

Most unlikely that you can find a brick wall big enough to do this at 14mm.

ChrisFB said:

I just bought a GX9 with 14-140 and would like to test the lens sharpness (I know, the GX10 could be coming, but I expect it to take a while and not have major IQ upgrades and the GAS was too strong).

What would be a good distance from the camera to the test subject (e.g. a brick wall) to test? I looked at a few video's, but the distances used were not discussed, but seemed relatively close, like a few meters. Wouldn't that impact the corner sharpness to much because the distance from the corners (left and right of the subject) to the camera vs middle of the subject to the camera differs somewhat? (relatively speaking and especially at the shortest FL)?

Thanks for the help!

Chris

Click to expand...

I would find distant panoramic cityscape and take shots with horizon running through the image corners diagonally at a few focal lengths at different apertures. For me it works better than a brick wall.

buratino said:

ChrisFB said:

I just bought a GX9 with 14-140 and would like to test the lens sharpness (I know, the GX10 could be coming, but I expect it to take a while and not have major IQ upgrades and the GAS was too strong).

What would be a good distance from the camera to the test subject (e.g. a brick wall) to test? I looked at a few video's, but the distances used were not discussed, but seemed relatively close, like a few meters. Wouldn't that impact the corner sharpness to much because the distance from the corners (left and right of the subject) to the camera vs middle of the subject to the camera differs somewhat? (relatively speaking and especially at the shortest FL)?

Thanks for the help!

Chris

Click to expand...

I would find distant panoramic cityscape and take shots with horizon running through the image corners diagonally at a few focal lengths at different apertures. For me it works better than a brick wall.

Click to expand...

Wht I have seen people do is use something like a row of buildings about 100m away

ChrisFB said:

I just bought a GX9 with 14-140 and would like to test the lens sharpness (I know, the GX10 could be coming, but I expect it to take a while and not have major IQ upgrades and the GAS was too strong).

What would be a good distance from the camera to the test subject (e.g. a brick wall) to test? I looked at a few video's, but the distances used were not discussed, but seemed relatively close, like a few meters. Wouldn't that impact the corner sharpness to much because the distance from the corners (left and right of the subject) to the camera vs middle of the subject to the camera differs somewhat? (relatively speaking and especially at the shortest FL)?

Click to expand...

That is not necessarily a problem because camera lenses are almost always designed to have as near as possible a "flat field". The subject distance in that case means the distance from the camera to the plane of best focus. The distance between a point and a plane means the distance between that point and the nearest point of the plane. In other words, the shortest distance between the point and the plane.

However, you also need to be aware that many lenses do not perform equally well at all distances. Many lenses (particularly very large aperture lenses) are designed to perform best at large distances and the image quality deteriorates slightly as you focus on very close objects. Macro lenses are designed to perform well at all distances, but they generally have much smaller apertures than other prime lenses of the same focal length.

Eric Nepean said:

Wht I have seen people do is use something like a row of buildings about 100m away

Click to expand...

I see that too, and question if it's the best way for a boring-but-repeatable lens test due to image degradation from haze and heat. 100m is about the length of a soccer pitch and there are times I can see conditions softening and distorting subjects at that distance.

For a zoom, I would tape a newspaper to a wall and fill the frame, moving in or out as FL is changed in order to keep the framing. Be vigilant about keeping the camera parallel to the target, which can be surprisingly tricky.

Have fun, then compare it with your other lenses.

So much spare time.

Rick

Skeeterbytes said:

Eric Nepean said:

Wht I have seen people do is use something like a row of buildings about 100m away

Click to expand...

I see that too, and question if it's the best way for a boring-but-repeatable lens test due to image degradation from haze and heat. 100m is about the length of a soccer pitch and there are times I can see conditions softening and distorting subjects at that distance.

For a zoom, I would tape a newspaper to a wall and fill the frame, moving in or out as FL is changed in order to keep the framing. Be vigilant about keeping the camera parallel to the target, which can be surprisingly tricky.

Have fun, then compare it with your other lenses.

So much spare time.

Rick

Click to expand...

What I have done in the past is created some high resolution targets (in areally old version of photoshop) then printed these out on 8 1/2 x 11 paper, taped these to a wall in the center and the corners of a pretty big subject plane (as big as my home will allow)

Then I move the tripod back and forth so that the lens's field of view matches the frame.

Its really handy, especially with a manual focus lens, to put a post it note in the middle with the lens Name and Aperture.

For a wide angle lens that is still only a few meters, but its enough to serve my purposes.

--
Cheers
Eric

ChrisFB said:

I just bought a GX9 with 14-140 and would like to test the lens sharpness (I know, the GX10 could be coming, but I expect it to take a while and not have major IQ upgrades and the GAS was too strong).

What would be a good distance from the camera to the test subject (e.g. a brick wall) to test?

Click to expand...

I have a standard test which I use to check how well fine detail is resolved and to check border to border sharpness. From this I can determine what apertures give optimal results for my use case. I think it's a good idea to relate the tests to your use case if possible.


Standard test scene

ChrisFB said:

I looked at a few video's, but the distances used were not discussed, but seemed relatively close, like a few meters. Wouldn't that impact the corner sharpness to much because the distance from the corners (left and right of the subject) to the camera vs middle of the subject to the camera differs somewhat? (relatively speaking and especially at the shortest FL)?

Click to expand...

This scene has a lot of detail and is shot from what I would call a moderate distance. In my experience fine detail shot at a close distance is not representative of fine detail shot from 20 metres away. Shooting only close up subjects can be deceiving - things tend to look sharper when shot close up.

ChrisFB said:

Thanks for the help!

Chris

Click to expand...

Standard test scene



Looks too 3-dimensional for me: I would not use it as aa target. But I agree: that's me.

buratino said:

Standard test scene

Looks too 3-dimensional for me: I would not use it as aa target. But I agree: that's me.

Click to expand...

As I mentioned, it's related to my use case so it won't suit everyone.

ChrisFB said:

I just bought a GX9 with 14-140 and would like to test the lens sharpness (I know, the GX10 could be coming, but I expect it to take a while and not have major IQ upgrades and the GAS was too strong).

What would be a good distance from the camera to the test subject (e.g. a brick wall) to test? I looked at a few video's, but the distances used were not discussed, but seemed relatively close, like a few meters. Wouldn't that impact the corner sharpness to much because the distance from the corners (left and right of the subject) to the camera vs middle of the subject to the camera differs somewhat? (relatively speaking and especially at the shortest FL)?

Thanks for the help!

Chris

Click to expand...

Distance to the subject does have an effect on lens sharpness. Some lenses more than others. Ideally do one set close, say a meter or so and one set a longer distance, 10+ meters or so.

Also lenses often have field curvature where the focus point varies across the frame. This is most common at the wide end. At the wider focal lengths I'd take shots with the AF point in the center, left and right to see how the lens actually performs on the edges when properly focused. Often a lens seems 'soft' on the edges when it really has field curvature. This will let you know of that is the case.

The old lens test charts that I have floating about are about 3 or 4 feet wide and the idea was to fill the frame with the test chart.

A simpler way would be to tape an opened broadsheet newspaper to a wall in good light and back off until it fills the frame at whatever focal length you need to test.

Be warned you need some space as I remember testing some lenses in film days and needed to use nearly all the 90 feet width of my back yard for some tele combinations.

Be aware of field curvature with some wide angle lenses may give a wrong impression of things like corner sharpness.

Maybe better to use the lens at realistic apertures (for depth of field) on real scenes to see how they look. Such as I tend to use my 12-40/2.8 lens at f/5.6 as the best compromise between quality and useful depth of field.

Also be aware in lens tests if focusing close on a small target then the focal length will not be what you think it is, internal focus lenses tend to shorten their focal length as they focus closer, a classic case was my old film days Nikkor 105mm macro lens, at 1:1 closest focus it was a 79mm lens.

In my case I used to do too much testing and worrying, but now just use the lenses and accept what they deliver. We just need to understand that some lenses are better than others.

It is the subject and the composition and lighting that matters way more than any degree of sharpness.

Regards.... Guy

ChrisFB said:

I just bought a GX9 with 14-140 and would like to test the lens sharpness (I know, the GX10 could be coming, but I expect it to take a while and not have major IQ upgrades and the GAS was too strong).

What would be a good distance from the camera to the test subject (e.g. a brick wall) to test? I looked at a few video's, but the distances used were not discussed, but seemed relatively close, like a few meters. Wouldn't that impact the corner sharpness to much because the distance from the corners (left and right of the subject) to the camera vs middle of the subject to the camera differs somewhat? (relatively speaking and especially at the shortest FL)?

Thanks for the help!

Chris

Click to expand...

All,

Thanks!! Not sure if it is useful to answer you all individually -this forum is always amazingly active and I am happy and thankful for all the quick responses, which I think all added something.

I learned a thing or 2 (actually more) here, for instance about the focal plane, field curvature, changing focal length at close distance (some lenses) and some good testing advice (printing some high-res test stuff and put it in center/corners, using several distances, horizon/diagonal through the image, etc.) .

Bottom line I guess the best is to use a few different scenario's / test distances, and I like the idea too to use some printed high res stuff, but in addition, using a true landscape for a bit further distance is a good idea too.

Perhaps the best advice (thanks Guy) is to not test it at all, accept it for what it is and just use the lens -critical sharpness is indeed not the thing that makes out the memorable pictures. Not sure if I can manage to not test though , but I also enjoy the process I guess.



Thanks!

Chris

Guy:

I used to have a set of the old National Bureau of Standards lens test charts (Now NIST) and compulsively test every lens I got my hands on, wide open, and at several other f stops. Yes lenses vary, though much less these days. Sharpest lens I ever tested then was an old f2 Leica Summicron (coated) on a Leica M3. I was in LOVE with that lens (I was NOT the owner). I bet I worried much more about lenses than you ever did!

Now lenses have better glass available, as well as computer ray tracing design. The IQ is MUCH higher generally. I worry MUCH less these days! We are most fortunate.

Jack Winberg

<<In my case I used to do too much testing and worrying, but now just use the lenses and accept what they deliver. We just need to understand that some lenses are better than others.>>

Nothing wrong with THAT! There has to be room for we obsessive types as well! ;-)

Jack Winberg

<<Not sure if I can manage to not test though , but I also enjoy the process I guess.>>

I concur. Test the lens at all distances you plan to use and inspect the results.



as an example, my sigma macro lens is great at short distances but kind of murky with low contrast and fuzzy at long distances.



test as you will use.



CDC

ChrisFB said:

Perhaps the best advice (thanks Guy) is to not test it at all, accept it for what it is and just use the lens -critical sharpness is indeed not the thing that makes out the memorable pictures. Not sure if I can manage to not test though , but I also enjoy the process I guess.

Thanks!

Chris

Click to expand...

That's a bit like I recently received back an old cathode ray oscilloscope (the kind of tool I sort of grew up with in my old job) fully intending to get stuck into researching shutter shock like CrisPhoto at https://www.dpreview.com/forums/post/53456685 - but sigh - life intervenes and it is gathering dust.....

You need to test it at the distances and apertures you intend to use it. Lenses are optimised for particular uses.

A brick wall test is a useful starting point, followed by some medium distance landscapes.

I'd also look at sample images from good reviews and test charts, to see what to expect. I've only once had a lens that didn't match expectations and that was from a known bad first batch. Mine wasn't as bad as some copies but much worse than others. 4th copy was acceptable but not as good as the manufacturer specially selected for reviewers.

Remember zooms have more copy variation than primes and almost all of them have good focal lengths and bad ones, just not the same for all copies.

Happiness is having zooms with overlapping FL, where their good FLs are at the opposite ends of the overall range.

Andrew

30x the FL is often quoted as the "right" distance to test and some dedicated test charts sold also advise this. Filling the frame with a sheet of newspaper is probably only useful for testing macro lenses, or maybe portrait lenses. Also, move the AF points to the corners or edges for testing there, unless you want to determine how bad/ good a lens is with field curvature issues... it also eliminates sensor plane alignment issues. Especially when wide open, take at least 10 shots and you will find small differences in AF accuracy or camera movement/ shutter shock etc (use electronic shutter preferably). Look at the best and sharpest samples (you are testing a lens and not the camera AF). Don't shoot faraway horizons as even on cold perfect weather days, heat shimmer wrecks sharpness. Don't view above 100% as everything looks bad then. Even at 50%, inspecting your PC screen close up, bear in mind that even at this level, you are probably exceeding any demands even a huge print will impose (unless you want to see how well images from a lens will stand up to cropping). Also, test at the tele end (100mm or longer) with and without any protective type filter... they can play havoc with tele lenses. Dont test smaller than f11... images will look the same or get worse... unless you want to see maximum DOF/ smallest aperture you can use when DOF is a priority. Use a m43 lens at an aperture that you know is a "benchmark" (highly well reviewed) as a reference as to what the potential of your system/ sensor is capable of... no point in rejecting a m43 lens if you are expecting medium format quality images.

Good luck.

--
Addicted To Glass
M43 equivalence: "Twice the fun with half the weight"
"You are a long time dead" -
Credit to whoever said that first and my wife for saying it to me... Make the best you can of every day!

Cafe Racer said:

ChrisFB said:

I just bought a GX9 with 14-140 and would like to test the lens sharpness (I know, the GX10 could be coming, but I expect it to take a while and not have major IQ upgrades and the GAS was too strong).

What would be a good distance from the camera to the test subject (e.g. a brick wall) to test?

Click to expand...

I have a standard test which I use to check how well fine detail is resolved and to check border to border sharpness. From this I can determine what apertures give optimal results for my use case. I think it's a good idea to relate the tests to your use case if possible.


Standard test scene

Cafe Racer said:

I looked at a few video's, but the distances used were not discussed, but seemed relatively close, like a few meters. Wouldn't that impact the corner sharpness to much because the distance from the corners (left and right of the subject) to the camera vs middle of the subject to the camera differs somewhat? (relatively speaking and especially at the shortest FL)?

Click to expand...

This scene has a lot of detail and is shot from what I would call a moderate distance. In my experience fine detail shot at a close distance is not representative of fine detail shot from 20 metres away. Shooting only close up subjects can be deceiving - things tend to look sharper when shot close up.

Cafe Racer said:

Thanks for the help!

Chris

Click to expand...

Click to expand...

A very good idea to use a familiar scene over and over for testing new lenses... You get familiar with what to expect.

--
Addicted To Glass
M43 equivalence: "Twice the fun with half the weight"
"You are a long time dead" -
Credit to whoever said that first and my wife for saying it to me... Make the best you can of every day!