Lens testing and performance

Blasted

Blasted

Suspended / Banned
Messages
927
Edit My Images
Yes
So how do you test or know if a lens isn’t quite on the ball?

From everything I have read, some lenses are soft in the center, some on the outside and some at different focal lengths. Third party lenses are renowned for being hit and miss.

How do you know if the lens is performing at an acceptable level?

I’m looking for a fast zoom for my 5D. A lot of people say the 24-70 F2.8 Mk II is the best, but it’s a huge stack of money. If I were to buy one how would I know that its performing as it should, same goes for other third party lenses, also not forgetting primes.

Do you test?
 
Blasted said:
So how do you test or know if a lens isn’t quite on the ball?

From everything I have read, some lenses are soft in the center, some on the outside and some at different focal lengths. Third party lenses are renowned for being hit and miss.

How do you know if the lens is performing at an acceptable level?

I’m looking for a fast zoom for my 5D. A lot of people say the 24-70 F2.8 Mk II is the best, but it’s a huge stack of money. If I were to buy one how would I know that its performing as it should, same goes for other third party lenses, also not forgetting primes.

Do you test?
Click to expand...

If the lens is a good performer you'll cut yourself on it :)
Seriously, yes I test mine, usually against a brick wall on a tripod, sometimes I photograph an inclined 1m ruler to check near/far focus adjustment requirements.
I test different lengths and apertures for a zoom, usually at one Iso setting, mirror locked up and self-timed.
Generally though I just look through my snaps and if they are sharp enough then the lens is performing as far as I am concerned.
DONT pixel peep (although I dont take my own advice here :) ), complete waste of time.
The 5d3 will show up any defects in your lens(es).

Matt
 
I use them, if they appear faulty I test them to make sure it's not just perception.
 
Pixel peeping and reading around whilst considering my next purchase haven't helped at all.

In the past I have had spot on shots and some that are not quite spot on, Maybe its all the lenses fault as I didnt test them. :bonk:
 
Do you want to know how sharp a lens is, compared to other lenses, or whether or not you have a good sharp copy?

The former is very difficult to do in a reliable and properly comparable manner, but the latter is easy. By far the greatest variable with lenses is centering of the glass elements, both with a new lens, and one that has taken a knock. This shows as one side or corner being less sharp than the others.

Brick walls are popular subjects, but are often very unreliable. The problems are than unless the camera is absolutely square to the wall, depth of field will throw out one side at low f/numbers. Also, very few lenses have a perfectly flat field and that also throws the edges out, whereas in practise a little field curvature is completely invisible and irrelevant (unless you shoot brick walls a lot).

So, choose a fairly distant subject to reduce focusing and field curvature errors, say a car up the road, or a street sign. Focus very carefully, lock everything in manual, shoot at lowest f/number. Use IS if you have it, and keep the shutter speed up - push ISO if needs be. Position the target (eg number plate, sign) near the corner of the frame and take a shot. Then without changing anything, take three more shots with the target positioned in the other corners. Make sure the target is positioned exactly the same distance from the corners each time.

Check results - if your camera has a decent LCD, zoom in on that and compare each corner. What you are looking for is not stunning sharpness (you probably won't get that, it's a tough test) but a consistent standard of sharpness in each corner. You don't need to look too hard as any significant problem will be immediately obvious. If you do look closely, even with the best lenses you'll probably see slight differences.
 
Phil V said:
I use them, if they appear faulty I test them to make sure it's not just perception.
Click to expand...

^^^ :)

I've never 'tested' a lens as such. When my 135L arrived from eBay, I gave it a good clean [covered in fingerprints etc which gave me instant doubts] popped it on the camera to make sure it focused & all was moving smoothly, went to the park with the kids & took some photos. All were acceptable sharp but not quite where they should have been so then a week or so later I tripod mounted it & gave it an MA once over. Don't know what it needed or even in which direction now.

Go take some photos & see what you come back with ;)
 
HoppyUK said:
Do you want to know how sharp a lens is, compared to other lenses, or whether or not you have a good sharp copy?

The former is very difficult to do in a reliable and properly comparable manner, but the latter is easy. By far the greatest variable with lenses is centering of the glass elements, both with a new lens, and one that has taken a knock. This shows as one side or corner being less sharp than the others.

Brick walls are popular subjects, but are often very unreliable. The problems are than unless the camera is absolutely square to the wall, depth of field will throw out one side at low f/numbers. Also, very few lenses have a perfectly flat field and that also throws the edges out, whereas in practise a little field curvature is completely invisible and irrelevant (unless you shoot brick walls a lot).

So, choose a fairly distant subject to reduce focusing and field curvature errors, say a car up the road, or a street sign. Focus very carefully, lock everything in manual, shoot at lowest f/number. Use IS if you have it, and keep the shutter speed up - push ISO if needs be. Position the target (eg number plate, sign) near the corner of the frame and take a shot. Then without changing anything, take three more shots with the target positioned in the other corners. Make sure the target is positioned exactly the same distance from the corners each time.

Check results - if your camera has a decent LCD, zoom in on that and compare each corner. What you are looking for is not stunning sharpness (you probably won't get that, it's a tough test) but a consistent standard of sharpness in each corner. You don't need to look too hard as any significant problem will be immediately obvious. If you do look closely, even with the best lenses you'll probably see slight differences.
Click to expand...

Thanks, I'm looking at the second option, that is if I had a lens is it a good sharp copy?

With some of the third party manufactures putting out very good lenses I just worry about the horror stories of lack of quality control.

I'm coming round to the idea the "L" lenses may not be the be all and end all in picture quality and that an alternative can perform just as well for a cheaper price. The sigma 35mm being a prime example.
 
I test too much, it becomes a habit though, one I am trying to kick but finding it impossible to kick, I'd suggest to only looking for issues if you notice something is off without pixel peeping
 
Ivan S said:
I test too much, it becomes a habit though, one I am trying to kick but finding it impossible to kick, I'd suggest to only looking for issues if you notice something is off without pixel peeping
Click to expand...

I can imagine you start chasing things that aren't there. All I would like to know is that if I spent between 500 and 1500 quid on a lens, it performs as designed. A lens could appear to be good to me but should be performing better, I just wouldn't know.
 
To test lenses and get meaningful results you need proper technique.
Here's something I wrote on the subject a while ago.
http://www.talkphotography.co.uk/forums/showpost.php?p=3949409&postcount=25

My original lens turned out to be a soft copy, but at that time my technique was barely up to illustrating the problem.
I only persisted because I was lucky enough to be able to compare it back to back with several other peoples copies of the lens and demonstrate mine was different.
My technique has come on leaps and bounds since then :)

As mentioned - when pixel peeping, not all lenses will give razor sharp images; it's an unrealistic expectation - some lenses are better than others.
For example, wide open my 300mm f4 is good (very good by most standards), but the difference is startling comparing the 100% zoom view against my 135 f2 wide open -the 135 f2 has stellar image quality and makes the 300mm look rubbish; but it isn't!
Also worth mentioning my 24-105 pretty much matches the results at sites like SLR Gear in that wide open some focal lengths are better than others, especially in the corners; you need to learn how to get the best out of your lens.
And worth mentioning my 50mm f1.4 is an unusually sharp copy; way better than the reviews would have you believe; soft copies works the other way too!

And the punch line is that in normal use you will never notice!
When a lens is used wide open, ultimate sharpness is missing the point of the image and isn't needed. And stopping the lens down pretty much sorts the problems out.
So most of the time lens sharpness is academic and will not affect your images.
 
Blasted said:
I can imagine you start chasing things that aren't there. All I would like to know is that if I spent between 500 and 1500 quid on a lens, it performs as designed. A lens could appear to be good to me but should be performing better, I just wouldn't know.
Click to expand...

You do, I can take a shot and think to myself, that looks fine, but then I get to thinking what if I adjust the AF tune some and then the image becomes sharper and i'm like :thinking: it was fine and now I am thinking the focus is off, so I keep testing until I am never happy<<<exactly :shake:, my own fault, I think I have a certain type of OCD:shrug:

Once you get it, you can ask here all you like and we will find out if it is sharp or not:thumbs:
 
Last edited:
I agree with Phil V and Duncan. Test it if you really feel you might have a problem. To read some of the forums you'd get the impression that photographers spend so much time testing their "copy" that they never have time to take photographs, and TBH I suspect that for many that is exactly the case. Just enjoy using it unless you spot an issue. Pixel peeping is a thief of your time and happiness ;)

As to the difference between the mk1 and mk2 versions of the 24-70L here's some food for thought:

http://www.slrlounge.com/canon-24-70-f2-8-l-mk1-vs-mk2-portrait-comparison

I have the mk1 and don't feel any need to upgrade anyway :)
 
Last edited:
wooster said:
... To read some of the forums you'd get the impression that photographers spend so much time testing their "copy" that they never have time to take photographs, and TBH I suspect that for many that is exactly the case. Just enjoy using it unless you spot an issue. Pixel peeping is a thief of your time and happiness ;)
...
Click to expand...

Photography has always been attractive to the kind of people who have as much interest in the process and the gear as they have in the art produced (or more so), as has music and other pastimes which mix art and science.

The web has made this very apparent, where the busiest parts of any forum are usually the gear discussions. I personally find it odd, but we have to accept people are different :love:.

How many weddings are shot professionally every year on Canon crop sensors?

How many forum debates have you read that constantly conclude that all Canon crop bodies are sub-par and unusable above 800iso? :lol::lol:
 
Agreed on all counts, Phil. The 40D was a stalwart of many wedding togs including very successful ones like David Ziser ( I believe he also uses the 7D now alongside his 5D mk3s ) yet this class of camera is often dismissed out of hand by gear - oriented posters.
 
Last edited:
You must log in or register to reply here.
Back
Top