EOS R6 focus guide inaccuracy

Started 6 months ago | Questions thread
rkc Forum Member • Posts: 68
Re: EOS R6 focus guide inaccuracy

ProfHankD wrote:

rkc wrote:

Yes, it seems to be focusing to infinity correctly.

I just double checked and it looks like I had misremembered slightly - the chipped adaptor I have that is behaving oddly is a C/Y adaptor not a PK one. I just tried it with a few different lenses and it reports "correct" focus in all cases when focused behind where it should be. It's not just a bit out either - it's enough to be very obvious that it has got it wrong.

My adaptor reports that the lens is f/1.4 - which none of the lenses I have tried actually are. I don't know whether that is is any part of the problem. It does seem as though the error is greater the more the lens is stopped down.

I checked in exiftool and it does seem to be reporting some rather weird values for the lens type:

Lens Type : Canon EF 35-80mm f/4-5.6

Max Focal Length : 65535 mm

Min Focal Length : 1 mm

Focal Units : 1/mm

Max Aperture : 1.4

Min Aperture : 1.4

Focal Length : 50.0 mm

One of the lenses I tried WAS 50mm, but other than that all the info is wrong.

Most main-sensor-based PD algorithms I know would use the aperture as a parameter, and it's quite possible they are tuning for f/1.4 marginal rays that don't exist when using an f/1.7 lens. If so, it would probably get worse when stopped down, and that's what you're seeing, so that's my guess. That actually bodes poorly for chipped use of adapted lenses in general, because chips generally report a fixed (maximum) aperture, so stopping the lens down would just make the PD data garbage.

The Canon EF 35-80mm f/4-5.6 is NOT a good lens. In fact, it's one of the poorer kit zooms with fully plastic build, a slow limited zoom range, and various optical issues. It is possible they'd be trying to correct for it, but if so, raw captures shouldn't be affected when processed by straight dcraw (or any other raw tool that doesn't apply corrections).

There is a slim possibility that the disagreement about the 1-65535mm with the 35-80mm causes the lens to be flagged as "fake" and is deliberately messed-up as an effort by Canon to discourage fakes -- sort-of like some Sonys refuse to work with batteries that don't ID themself as Sony made. However, I'm not aware of Canon doing that.

As luck would have it, I just got another chipped adaptor - M42 this time, which means I was able to test it with a F/1.4 lens.

I still see the exact same behaviour as with the prior adapter / slower lenses. Even with the lens wide open, the confirmation is considerably out from where it should be.

This chip is reporting the same weird settings via exiftool as the C/Y one. Both are unbranded and of unknown origin via eBay, free with lenses that I bought.

The other downside of using (these) chipped adaptors on the R5 is that it restricts the burst rate when using them, compared to when using an unchipped adaptor - presumably because the identified 35-80mm does not have a fast aperture.

-- hide signature --


Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow