Lens testing chart?

Nifkin

Nifkin

Suspended / Banned
Messages
1,267
Name
Simon
Edit My Images
No
Where would I find a lens testing chart, like I've seen mentioned/illustrated in various forums, to test the focusing properties of my lenses? I am able to spot CA and vignetting when I see it, but there's always so much talk of lenses being soft, back or front focusing, etc, that as a newbie it's hard to tell if the bunch of lenses I've got are any good or a pile of crap! Any suggestions would be great ;)
 
Nice one Ger, exactly what I was looking for!

Nif :clap:
 
Kerioak said:
There is one mentioned in this month's PhotoPlus magazine - around the £70 mark if I remember correctly
Click to expand...

Huh? £70 for a test chart? :eek: Don't most people use the free one listed abover or a brick wall?

I would love to see what you get for £70!! :thinking:
 
there are free ones, which i downloaded a while back, its the official ones you see people use in their reviews all the time. called ISO 12560000 chart...

However it looks really complicated to use, and i can't even understand what all the lines are for. If you are just after AF accuracy like I do when i buy second hand lens this is the place and it's chart is dirt easy to use! http://www.dphotojournal.com/focus-test-chart.pdf

Vignetting, i think go for full frame lens then one your dx body, it won;t have problems what so ever, and also since you have lens the lens to test, u might as well go out and take a few shots and see what happens. I notice vignetting on my tokina most prominently when there is poor lighting. so maybe that's something you can recreate and test for vignetting as field/visible rather than boring and useless stats. with regard to CA, i used to think its a big issue, but shoot in RAW and use ACR CA can be really easily corrected.

so basically its a problem of the IQ, Build Quality (BQ) and AF issues. IQ you can open you test image and have a look at sharpness, AF you can test with the chart (linked) and BQ well u can see what you have bough so...
 
I would say DON'T use test charts like the use linked to by Gerods. They're really not reliable, for two reasons:
(1) The AF sensors in the camera are bigger than you think, so you can't be sure what they're focussing on.
(2) You're typically working near the minimum focussing distance of the lens, which makes the AF particularly vulnerable to small aberrations in the lens.

Instead, use a test chart with lots of fine detail like this one which is designed for micro-focus adjustment. Set it up perpendicular to the axis of the lens, and at a distance of at least 50 times the focal length. Have a look at the picture you get using the AF and then see whether you can make it any sharper by manual tweaks. If you can't, the AF is spot-on.
 
Don't get me started on test charts :eek:

Over the years as a photo magazine technical editor I've tried everything. No test is perfect, all can give misleading results, and with some it is actually quite hard to get anything meaningful at all.

The best lens test procedure I know of today is, on balance, that used by DPReview but even that has its shortcomings despite the huge amount of resources they put into it. Fortunately, they know what they're doing so any anomolies are picked up and commented on separately.

But for most people, and I really don't mean to put anyone down here even though it probably sounds like it, a little knowledge is a dangerous thing. In particular, shooting any flat test target at relatively close distance will show 'apparent' shortcomings in some lenses that are simply irrelevant. Sorry Stewart, but even 50x focal length is not enough for some lenses, eg fast wide zoom ;) And if you test the AF on your 70-300mm with that A4 sheet of paper at 45deg, your focus is more likely to appear to be out than in, when it's actually fine for normal photography.

Test targets are only necessary when you need to take comparitive measurements, that are repeatable day to day and month to month. Most people don't need that - they just want to know if their gear is working properly. In which case, the best test target you will ever find is right outside the door in everyday subjects.

It's easy to devise real world tests that are 100% relaible and relevant. I've gone on too long now, but if anyone wants to check anything specific I'll try and post something up. Right now I want my Sunday lunch, but I've posted stuff before if anyone care to search my profile.

Cheers,

Richard.
 
You must log in or register to reply here.
Back
Top