How effective is R6's IBIS with EF Lenses?

QSMcDraw wrote:

Right, the bug appears in a specific scenario. You can easily reproduce this, and see it without pixel peeping, by simply handholding at a slow shutter speed, say 1/10th to 1/20th sec., take several shots. Compare the first shot to subsequent shots in the series. It isn't only the edges, but that's where the blur shows most because of the twisting motion of the IBIS.

Ok, so perhaps I was having trouble reproducing the problem because I was taking pictures of a bookcase at too slow of a shutter speed, 1/2 second instead of 1/10th. I tried with an RF 50mm f/1.8 and an EF 135 f/2, both at 1/2 second and f/5.6 hand held. I turned the camera off after each shot. Then I looked at the book spines on the extreme left and right sides of the photo. Now, they were not sharp every single time, but they were reasonably sharp (not blurry even if perhaps a little softer than more central book spines) most of the time. These were all first shots after power on etc etc.

Maybe it only happens at wider apertures. Maybe I didn't wait long enough between power cycles, though I did wait at least a minute in each case. Maybe I am a fairly stable hand shooter most of the time. Maybe I have a slightly different standard for sharpness in the extreme edges of a hand held shot. Perhaps my standard is not a high enough standard. Perhaps Canon's standard is also lax.

As for using IBIS set to "Only for Shot," in my opinion, that setting sacrifices one of the best aspects of IBIS--keeping the EVF image stable while focusing and composing.

The motion blur caused by the problem does show up at higher shutter speeds, just not as much, eventually getting into pixel-peeping territory.

IBIS does not address motion blur. That comes from the subject. IBIS addresses camera shake.

I mentioned my workaround--just take a short burst of shots. And I've been doing that for years with the 5D series of bodies. Another workaround is to give up great non-IS primes, such as the very Rf lenses which prompted me to switch to Canon mirrorless.

Deal breaker? No. Identifiable bug that Canon should fix? Yes!

Agreed that Canon should fix the bug if there is a bug.

Not sure DIY handholding tests are really the way to eliminate the possibility that the problem is in the shooter rather than the camera. Perhaps that would require lab testing to eliminate the human element. Also, I don't really like repeatedly turning my camera on and off just to carry out a probably invalid experiment.