Sigma 18-250mm f/3.5-6.3 DC OS HSM
OS in-lens stabilisation
A few casual tests revealed that on average the in-lens optical stabilisation works better than the in-body SSS when shooting between 150mm and 250mm with shutter speeds around 1/30-1/60th. Neither of us liked the floating effect of this stabilisation, or the sleep-mode parking off centre, and the sudden return to centered position. But it works just a little better than the Alpha 700 SSS at the long end where it counts most.
The floating OS group parks itself ‘up’ when the camera is held for a normal landscape scene. Viewing through the lens with the camera turned off, or when the OS has gone into sleep mode, the image appears as it would if the entire lens was shifted about 2.5mm vertically (like a 2.5mm rising front) at the 18mm setting. At longer focal lengths, the offset is greater, but normally you would keep the lens with the zoom locked at 18mm when walking around. When you turn the camera on, or touch the shutter button and wake the system up, the image shifts to the ‘centre’ position and stabilises, taking half a second or so. If you release the shutter during this period, the result will be blurred ; unlike in-body SSS, you must wait for the image to settle. This is a key difference between in-lens OS and in-body SSS/SS, and it has practical implications. For the two cricket shots below, in-lens OS was turned off. There was a possibility of the lens parking the OS between shots, as cricket can be a leisurely game, but the timing of a single shot is critical.
I did not want to have this small delay in safe shooting caused by the OS. In-body SSS/SS would have been perfectly usable as it only operates during exposure and no warm-up or settling time is involved. At the speeds used here, faster than 1/1000th of a second, I felt no need to use any form of stabilisation.
On the other hand, you may ask why Sigma decided to leave OS in a lens intended for the Alpha mount. They have omitted it from some earlier designs, which are OS for Nikon and Canon only, plain non-OS for Sony and Pentax. They reasoned that users can make the choice. Perhaps they also did some tests and found that in-lens OS can work a little better under some conditions, and that users may like the stability of view offered especially at the long end. There is no doubt that hand-holding can be a better experience with a stabilised lens when you are working at 375mm full-frame equivalent.
I won’t show shake tests, as I have no equipment to repeat them exactly. I just took a load of pictures at difficult or impossible settings using both systems and compared them. Neither achieved a perfect hit rate, but with a shutter speed of 1/40th at 250mm one in six shots was visually shake-free at 100% with SSS, compared to two out of six for Sigma OS. At shorter focal lengths I can only say that Sigma OS appeared better able to deal with unreasonably long exposures like 1/8th or 1/4. Sony SSS seems to have the most benefit in making marginal shots (like 1/60th at 100mm) critically sharp every time.
You can turn OS on or off, but when it is on, you must turn SSS off on the body. Using both together is a disaster, it only took a few test snaps to show why. You end up with amplified shake, not corrected shake!
OK – time has come to decide between new lenses for the A77 that is on order –
Sigma 18-250 OS HSM or Tamron 18-270 Pz motor?
I managed to take some more shots today, did some experimenting with
aperture and ISO settings. Double checked that OS was off and SSS on.
I took a number of pictures in the f8 – 11 range. 18mm, 250mm and a few macro shots. For the macro shots, I tried both manual and Scene/Macro settings, didn’t make any difference that I noticed. I tried the f11-16 range.
I hate to say it, but I am not impressed.
The image quality at low ISO (800 and below) is about the same as my P&S up to about ISO 200; but going to higher f numbers forced me as high as 1600 ISO …in bright daylight. So stopping down the aperture starts to cut into the advantage of the better ISO noise.
Macro IQ is definitely better on the P&S.
I do see some stunning images on the web from this setup, but right now I’m thinking there may be a lot of PP involved. I certainly never came close to “stunning”.
This being a lens review article, I feel a bit guilty bringing the camera body into it, but I can’t really separate out whether it is a limitation of the lens or the body. I concede it may be me, but I sure don’t see what I’m doing wrong.
I really was hoping for better low light performance (got it), more features (got those), and better IQ both in good light(nope) and macro (nope).
I’m pretty disappointed. I did a lot of research and this combination came out as about the best all-round setup I could find. I really wanted it to knock my socks off … but I think it goes back in the box and back to the dealer.
I don’t know whether to bother trying one of the other manufacturers, or whether to give up on the superzoom idea and go to lens swapping.
Or maybe just forget DSLR’s altogether.
Anyway, thanks for your help David. I’m glad the lens/camera works for you, I guess maybe I just had unrealistic expectations.
Thanks for the quick reply and the tips David!
It’s raining here in southern NH/USA so I can’t get out and do test shots right now, but as soon as I can I will and will get back to you.
Your tips about aperture may be the issue; the quick shots I did last night were as wide open as the lens settings would allow; so maybe that’s the problem.
I was using SSS, with the lens OS off. I doubt I can be patient enough for that long wait for the OS to activate !
I hope I can get this IQ straightened out, other than that (which is key of course) I really like the setup.
I guess I’m looking for a sanity check. Never had aDSLR before, just bought a Sony a55 and Sigma 18-250 and am surprised at the relatively poor IQ compared to my Canon Sx120 P&S. I expected IQ to be better in all lighting, but so far (only a few test shots, can’t do many if I want to return it – which I think I do), disappointed.
At high ISO, the a55/Sigma is clearly superior.
At low ISO, outdoor shots in good light, both macro and wide, the Canon P&S IQ is better. Much unexpected!
Is this common, that a P&S IQ be better at low ISO?
Or maybe there is technique I’m missing … maybe just because I’m not at the ‘sweet spot’ aperture for the sigma?
Been shooting at the lowest f-stop possible. Maybe that’s bad?
Reviews for both the a55 and the Sigma are all good … so if this is as good as it gets I guess there’s no point in trying other DSLR / lens combinations 🙁
Clearly something is very wrong as the A55 image quality at low ISO is superb – and the Sigma is the best lens of its type. First of all, the Sigma is generally at its best around f/8 to f/11 on the A55. It will not be capable of good macro shots at 250mm and f/6.3 (wide open is never sensible with any long range zoom) and the depth of field will not be sufficient for many subjects, so for macro work I would always try to use f/11 or even f/16 and move to a higher ISO. It is noise free up to ISO 1600 if processed correctly from raw, and this is very useful for macro. If you try to stop down to say f/16 or f/22 for routine work you’ll just get soft images (diffraction) and probably subject movement at low ISOs.
One thing to be absolutely sure about is the SSS setting and OS on the lens. We use the lens OS, it really helps with viewing at longer lengths. This means the SSS in the camera MUST be set to OFF in the menu. If you leave Steady Shot set to ON, you will get very unsharp images when the Sigma lens is also set to OS, the two systems will overcompensate or even work against each other and blur the image badly.
David
Thanks a lot for the review, very detailed and very interesting indeed. Did you try it on an Alpha 55 which you reviewed as well?
Thanks again for what you do David for the Alpha users.
We’ve used the 18-250mm on the A55 and our sample works well enough except for one small issue, from time to time the lens can lose contact with the camera. It needs a factory fix to be perfect because it is an early example of the lens. Service is restored by simply pressing the lens release button and briefly shoogling the lens, it seems to be a mount-fit issue nothing else.
A very interesting review….. I have the Tamron 18-200 and really loved it when I first bought the lens, but now it is starting to feel a little long in the tooth, or my photographic skills have improved! The implementation of HSM is an obvious drawcard for this lens, and thanks to your review, I might have to start saving up for it……..
Your reviews always show the users’ perspective, not just the technical jargon that you get elsewhere. The important part of how a lens performs is how if feels on the camera, and the resulting images that it takes along with how much extra work you need to put in afterwards to get the image right! To me, the less that has to be done, the better!
Thanks again!