best way to test a lens?

[takeru]

canon i have a 17-55 IS lens and a canon 40D. i recently got the lens (used) and i am trying to test it to see if everything is operating ok, but i am not sure how to properly test it thoroughly. any tips/suggestions? randomly shooting at things doesn't seem viable. i am most concerned about its focusing. i took a random shot and compared it to a similar shot i took with my canon 28-135 IS lens, for some reason the 28-135 seems sharper on the object i focused on? need to test this thing to be sure asap.

 

[cputeq]

Different lenses may have different performances - it may just be that your 17-55 has less sharpness.

Generally, for testing, I will do this --

Find a focus chart or just get a newspaper.

Set the chart/paper on a flat surface. Mount camera on tripod and look down at about a 45 degree angle.
Set aperture wide open, focus in on something.
Use 2sec timer, take picture.

View the image on the computer and see if your focus point is sharp -- then see if the text behind and in front of your focus point is sharp. Generally, you should have a bit of sharpness evenly distributed around the focal point (in front and behind), and then they should both blur at around the same time.

It's a bit tricky, so take it slowly and make sure you make no mistakes.
Generally, though, I don't even use a tripod - I just take a few angled shots at things with text, etc and mentally remember the focal point, then check on the PC.


Luckily my brand new 40D and 28-135 is spot on in the focus =D
What's more, I did a resolution chart test, shooting wide open to F/11 at 28mm, 70mm and 135mm, and I can't believe there's so little difference in sharpness between the apertures...not bad at all! I actually have to compare corners at 100% to see the difference, and at F11 I can barely see more paper texture than at f/5.6 Color me impressed.

 

[virtuamike]

If you're troubleshooting then shoot a ruler or a newspaper or a brick wall, but best way to really get a feel for what a lens will do is to just shoot with it.

 

[PowerRanger]

You can try this: http://focustestchart.com/chart.html

 

[takeru]

Originally posted by: PowerRanger
You can try this: http://focustestchart.com/chart.html

thanks, that looks good. i'll have to try that later when i get home.

mike:
what do you do with the ruler? yes, trying to test lens to make sure its 100% working correctly.

tried the 45 degree newspaper shot. seems ok, but wanted to see if any other tests to confirm.

 

[jpeyton]

100% crops are a tool of the devil.

I'm only somewhat joking. I don't think I even bothered being a measurebator until I started frequenting photographic forums.

Testing a prime lens is easy, but zoom lenses are much harder. Lenses have different focusing characteristics that change depending on their focal length and focusing distance. For example, testing my 80-200mm at 200mm at the closest focusing distance produces backfocus, but focusing at 10+ ft produces sharp focus.

Charts can be useful if you're noticing significant deficiencies in your AF, but if nothing shows up in day-to-day use, I say leave well enough alone.

 

[cputeq]

Originally posted by: takeru

Originally posted by: PowerRanger
You can try this: http://focustestchart.com/chart.html

thanks, that looks good. i'll have to try that later when i get home.

mike:
what do you do with the ruler? yes, trying to test lens to make sure its 100% working correctly.

tried the 45 degree newspaper shot. seems ok, but wanted to see if any other tests to confirm.

Ah yeah, I forgot the ruler. It's an easy one.


Just take the ruler, face it up/down from you and take a shot of a tick mark on the ruler.
View the picture and see the distance the sharp numbers are (forward and back) from where you focused.


You have to be careful though, and make sure the camera is focusing where you *think* it's focusing.

A good way to do this is draw a dark line or place a dark, slim object in the middle of the ruler, assuming the ruler is of a solid color. That way, the AF sensor will pick up on that easily (instead of a lot of little black lines on the side), but you can still notice the sharpness of the lines.


Again, though, you have to be really careful of your methodology or you'll think you have problems were none may actually exist. For instance

1) Make sure your ruler is perfectly perpendicular to the angle of your camera.
2) It helps to do this in natural daylight, as artificial light can mess with camera AF.
3) You have to make sure your camera is actually focusing where you think it is, hence the idea of focusing on a high contrast spot in the middle and then looking at the edge marks.