Re: Tripod Test - 50-200 SWD + EC-14
Tony Rogers wrote:
Anders W wrote:
...
Many thanks for redoing the test of the optical performance of the 100-300 versus the 50-200 with TC. Looks more evenly matched now and with the 100-300 up to about the level I think it is capable of. I am perfectly open to the idea that the practical keeper rate may be higher with one lens than the other in spite of that. I am certainly not saying that I am anywhere close to 100 percent with the 100-300 at the long end.
I have never noticed such a difference between tripod and handheld with the 100-300 before. Partly, because I very rarely use a tripod. Perhaps this shows the limit of OIS. Or perhaps the CDAF finds it harder to get a precise focus with handheld and OIS. Or maybe, when the optical elements are moved off centre by OIS to compensate for movement, the lens resolution goes down. The earlier 100-300 sample looked to be lower resolution rather than motion blurred anyway.
I wouldn't think it's the OIS or motion blur in this case (although this is sometimes hard to tell). Looks more like a focus problem to me. One possibility is that you swayed very slightly out of focus before taking the shot. I often use the 100-300 for close-ups, sometimes with extension tubes, and have noticed that this is a problem. Usually, I like to prefocus and only then release the shutter, but at close distance, there is a clear risk of swaying slightly out of focus if you do that. I have even experimented, with some success, with using AF-C instead of AF-S to get around this problem. Going back to the old MF strategy of just setting approximately right and then rocking to the right focus point with your body is of course another option.
Another thing. OIS vs IBIS. When I first got the E-M1 I did some comparisons between OIS and IBIS to see which was best on the 100-300 at 300mm. I found OIS to be more reliable.
I too found the (firmware-updated) OIS of the 100-300 to have a very slight edge over the E-M5 IBIS. But I haven't tested this extensively enough to be sure and the difference is slight anyway. The main problem here is that shutter shock spoils the fun if I try anything slower than 1/250. I can get passable sharpness below 1/250 but not tack-sharp ones.
This is a concern not only in hand-held shooting but on a tripod as well. With the 100-300 at the long end on a fairly sturdy tripod (Sirui T-1205X with Sirui G-10 head), I get blur due to shutter shock if I go below 1/250 even with long anti-shock delay. The only solution is to stay at 1/250 or higher or go down to really slow speeds (1/25, preferably even slower).
Because of this experience, when I got the 50-200, I was a bit concerned that IBIS might not be up to the job. However, I have found it to be excellent at 283mm on the new lens + EC-14.
How would you say the 50-200 with TC and IBIS compares to the 100-300 with OIS with regard to stabilization/shutter shock? Adding weight in the proper place helps against shutter shock and the 50-200 with TC beats the 100-300 in that regard. On the other hand, the proper place to add the weight is below the shutter, not in front of the body. Weight in front of the body moves the center of gravity outwards and increases the likelihood that the shutter forms part of a moment arm. This in turn increases the likelihood of significant blur.
My tests have shown that the angle of view on the 50-200 + EC-14 is smaller than the 100-300 @ 300mm implying that the Pany has a shorter focal length than the Oly combo. Perhaps the 100-300 is really a 275mm lens or even less given that some of the image is lost due to distortion correction.
It may well be that the 100-300 is not really 300 at the long end. But if so, it has nothing to do with distortion correction. It's only at non-tele FLs that this is a factor to reckon with.
Maybe this is why IBIS doesn't seem to work very well with it since the camera needs to know the focal length to stabilise properly. But the Pany reports 300mm even though it is probably a lot less.
I wouldn't think this is a factor for IBIS functionality either. In all likelihood, IBIS does get information about the real FL rather than the nominal one in case the two differ. And even if it doesn't, the difference in AoV between 275 and 300 wouldn't matter much.