Sigma calibration

Messages
2,065
Edit My Images
No
I have a Sigma 24-60 which, whilst razor sharp at f/4-5.6 and above, is slightly soft at f/2.8. I know most lenses are softer wide open but, having seen 100% crops from samples of this lens at f/2.8 which are as sharp as mine is at smaller apertures, I'm wondering if it can be improved by sending it to Sigma for calibration.

The softness is purely a function of the aperture, not the focus. I've tested the AF and it's spot on at all apertures. Now I know that calibration can correct issues with the AF "missing" but can it actually solve my problem or would I be paying for nothing and likely get the lens back in the same condition (or even worse at smaller apertures!).

Anyone here had a similar issue to this and had their lens fixed/improved by calibration?
 
Get in touch with Sigma, it won't cost too much to calibrate it; the 24-60mm should be very sharp wide open IME.
 
Thanks. Is this the type of thing that can usually be fixed by calibration though? I've always thought that calibration was simply for correcting front or back focusing rather than "softness".
 
It can be rectified....best to get in touch - their support team is very good and will be able to advise accordingly.
 
The softness is purely a function of the aperture, not the focus. I've tested the AF and it's spot on at all apertures. Now I know that calibration can correct issues with the AF "missing" but can it actually solve my problem or would I be paying for nothing and likely get the lens back in the same condition (or even worse at smaller apertures!).

The only way to test that it's not a focusing issue is by manually testing the focus - AF working correctly does not prove anything if the camera backfocuses slightly. I had an experience with a few Sigmas (all 24-70 F/2.8 DG) and Nikkor 50mm f/1.4 AFD that had exactly this. The front/backfocus slightly (a few mm) will result in an appearance of lens being soft wide open and will disappear at smaller apertures (well not disappear but will be hidden by increasing DOF). The only way to test it is with a focus test chart - this one for example. If you already done exactly that - then I'm sorry and please ignore my post ;)
 
I have tested the focus, not using a focus chart but using manual focusing via Live View at maximum magnification and couldn't get it any more accurate or sharper using manual than the AF managed so I believe the focus itself is spot on.

The effect itself at f/2.8 is a bit strange - not just a lack of sharpness but a very slight "fringing" or "ghosting" effect. Will sort out some 100% crops as examples and post them up (guys at Sigma UK have asked to have a look too).
 
Sigma UK have suggested that I might want to send my body in with the lens so it can be calibrated more accurately to the body. Presumably no changes would be made to the actual body itself.

Has anyone done this and/or would advise it? Not sure about the hassle of losing the body for 4-5 weeks but also not sure how much better calibration I could expect if I did so.
 
Well to get it 'properly' calibrated, it is common for the body to go with the lens. I very much doubt it'll be as long as you mention. Typically Sigma have a turnaround time of under two weeks.
 
Well they said that, as the 24-60 is an older lens, it's "non-RW". I think the RW refers to Read/Write and means they can't program the circuitry inside the lens themselves as they said they'd take test pics with it and send them to Sigma Japan who would send back a new circuit board to go in the lens. They'd then re-test and repeat as necessary.

Seems like a lot of hassle and this is why it can take 4-5 weeks but impressive that all this is covered under their standard £33 calibration fee tbh.
 
The effect itself at f/2.8 is a bit strange - not just a lack of sharpness but a very slight "fringing" or "ghosting" effect. Will sort out some 100% crops as examples and post them up (guys at Sigma UK have asked to have a look too).

Ah, that looks like one lens element may be slightly misaligned inside - definitely needs to be sent to Sigma to fix then.
 
Interesting, would such a misaligned element only affect images at certain apertures? I'd have thought something like that would affect the performance regardless of aperture (not that I know anything about optics :) )
 
Did some tests against an ISO12233 test chart. 100% centre crops, tripod with remote release, conversion from RAW with no sharpening whatsoever.

Whilst I appreciate that most lenses are softer wide open, this one seems to suffer more than it should. It's not too bad at 24mm but at 60mm, f/2.8 is bordering on unusable unless you're going to be resizing it right down.

It's not the AF missing either as I also tried manually focusing via live view at maximum magnification and couldn't get it any sharper than the AF.

It's also underexposing slightly at f/2.8, the histograms show this very clearly.

60mm f/2.8:
60mm_f2.8.jpg


60mm f/5.6:
60mm_f5.6.jpg


60mm f/8:
60mm_f8.jpg


24mm f/2.8:
24mm_f2.8.jpg


24mm f/5.6:
24mm_f5.6.jpg


24mm f/8:
24mm_f8.jpg
 
Back
Top