135mm SHOOTOUT! Detailed Analysis of the Hand-Holdability and General Sharpness of the RF135/1.8, the RF100-500/4.5-7.1, and the EF135/2

[AlanF]

Reikan Focal looks interesting... I'll bookmark it. What's AFMA?

From my 90s lens collection, sharpest aperture always seemed to be two stops down. But now, the 135/1.8 and 100-500 (at least at 100-135) are so sharp wide-open that I think a 45MP sensor can't even make use of any more resolution. That is to say, they're already delivering just about as good a result as the sensor can sense wide-open. Maybe on 65MP or 100MP you could detect them getting sharper yet, but on an R5 possibly not.

= AutoFocusMicroAdjustment. Used for DSLRs as the focussing detector is not on the sensor and as often as not gives slightly the wrong focus that needs micro adjustment. Sometimes the focus can be radically out. The AFMA dials back in the right focus. One of the great advantages of mirrorless is that focus is detected on the image sensor and so no correction required.

 

[Viggo]

I remember finding it funny that the sharpest aperture of both my 200 f2’s was 2.2, and the difference between f2.0 and 2.2 was hardly visible

 

[Michael Clark]

How does mode 3 of IS whereby IS comes on only when you press the shutter manage to work?

I'm guessing IS spin-up time is a lens-by-lens thing. Most of the lag when IS is first turned on is due to the time required to spin the gyros up to nominal speed.

Those lenses with IS Mode 3 tend to be more expensive, and thus probably have faster activation times. The only one I have with that kind of option is a third party lens, though.

I'd be very surprised if Canon lenses with Mode 3 didn't go ahead and spin up their gyros when AF or metering is activated. They just wait to actually move the IS element based on the information gathered by the gyros until the instant before the shutter opens. The main advantage of IS Mode 3 is that the IS element/group is centered when the exposure begins, and has equal travel available in all directions during the exposure. If IS is already active before the exposure begins, it may already be at or near the limit of travel in a direction that the lens needs to keep moving during the exposure to maintain the benefit of IS.

 

[SwissFrank]

One important thing to know about IS systems is that they takes ~0.5 s to achieve full stabilization, and during that period between IS activation (e.g. with a half shutter button press) and full stabilization, the IS system actually makes the resulting image worse. So in a situation where the lens is pre-focused and you're just pushing the shutter fully down in one motion to take the shot, the IS system is working against you. If you're going to 'mash the shutter button' to take a shot, you're better off with IS turned off.

OK, in my tests I was using the camera like I always do: hold R5 to face, half-depress shutter then gently squeeze. I didn't time how long I had the shutter half-pressed though. Just to experiment, based on my understanding of your meaning, I tried a few 2 sec shots with the 50/1.8 across the room. This is just a trial of one shot each, turning the camera off 5 sec, then turning it on and doing a test shot:

turn on, bring to eye, let the sensor know I'm here, press shutter normally: pretty sharp

turn on, bring to eye, let the sensor know I'm here, and go from no shutter to pressed shutter in one movement (hard to overcome decades of habit on that!): pretty sharp

turn on, hold at arms length so eye sensor doesn't see me, press shutter normally: unusable image

turn on, hold at arms length so eye sensor doesn't see me, and go from no shutter to pressed shutter in one movement: unusable image

So if I understand your point correctly (not sure about that) it seems the R5 at least doesn't care how long you press the shutter button. Either 1) maybe RF no longer needs the half-second you refer to, 2) maybe it does but the eye sensor at the viewfinder initiates it, or 3) the difference is mainly that from decades of practice I hold the camera well-braced against my face and arms, but don't do so when holding it in front of me. I can't determine which, but I don't think I can agree that "~0.5 s to achieve full stabilization, and during that period between IS activation (e.g. with a half shutter button press) and full stabilization, the IS system actually makes the resulting image worse" is true, at least with the R5 with RF lenses, and at least if understanding your point.

For the record I wasn't testing with IS Mode 3. Also, the above tests have the menu Camera-7-IS mode set to "on" and "always." I suspect IS could do an even better job set to "only for shot," because the IS wouldn't have used up some of its leash stabilizing the viewfinder. But I haven't tested that. (Maybe I should.)

 

[neuroanatomist]

I tried a few 2 sec shots with the 50/1.8 across the room.

I should have been clear that I was referring to lens IS systems, which the 50/1.8 lacks. IBIS likely behaves differently.

I haven’t tested this on a R body, I did so some time back with my 1D X and two lenses (100/2.8L and 600/4L II).