Chip for MF Lenses

Part 2!

The MF chip is of great interest to me as I have a Pentax 500/4.5 K mount which ,although all manual has been converted to Canon EF. This works fine, no problem at all.
Except, on a 10D I can use Extenders with it if I tape the lens contacts. On the 1DMk2 this isn't possible because that body has a microswitch in the lens mount which 'tells' the body that a lens is mounted and to expect signals from it. Now all the adapters for other lenses to EOS and the T2 mounts have a slightly shorter flange on the mount which doesn't trip this switch, so the camera doesn't expect any signals from the lens.
Putting on an Extender 'tells' the camera to expect signals from the lens, no signal = ERROR.
This chip just might fool the body into thinking that there is an EF lens on the other side of the Extender and let me have my 1000mm f9! As well as giving the focus confirmation of course........
I shall attempt to contact the guy who is making these.

condyk wrote:
Time will tell

What a pessimist you are!
It's lasted 18 years (is it really?) and a Canon EF lens made back in 1987 for the EOS 600 and EOS 1 still works on a !DsMk2 - whilst some Nikon lenses much newer than that will not work at all on some of their modern DSLRs and some will even cause damage..............
I'm not saying that Canon will NEVER change the mount again but I don't see it happening for a while. What I do see happening is an overhaul of certain lenses in the Canon range to update them for digital use - possibly sooner than later - which is maybe why we have seen the demise of a couple of oft lamented lenses without replacements as yet (50/1.0L, 200/1.8L).

@PaulB:
I'm not sure will this chip work as expected with 10D, Pentax 50mm F4.5 and Extender. Light level may not be sufficient for focus confirmation - 10D will not activate AF with lenses slower than F5.6. It will with lenses that report incorrect value (e.g. Sigma 50-500mm F4-6.3) so actual light level needed must be a few f-stops smaller. This chip, as I understood, will report F2.8 but actual light level may be to low. Check that with the seller before buying.

This is not a silver bullet, OK the focus confirmation can help putting the subject in focus, but it'd still not be as precise as a split screen.
So if you *don't* have the split screen, it could be a useful replacement...

The AF mechanism is not dependent on aperture of the lenses at all. It's artificial limitation by software.

Let's say you have an f/5.6 and it can focus at EV 12. If you put an extender on it which makes it f/8, then the AF will work EXACTLY THE SAME WAY if you have EV 13.

I agree, but as light level goes down, reliability of AF decreases. That's probably reason why Canon restricted it.

The problem, as I stated in my earlier post, is NOT with the 10D.
I can fool the 10D, but not the 1DMk2! And that isn't limited by the apertures I'm using anyway as I DON'T have AF and the AF confirmation would be a bonus over and above been able to use the Extenders - the alternative is to shorten the flange on an existing Extender; then it wouldn't work on 'normal' lenses!

Also the maker of this chip DOES NOT say that the chip will report f2.8.
What he does say is that he used a 24-70/2.8L to obtain the chip protocol from; as I understand it all he needed was the circuit to 'fool' the body into thinking that an AF lens was mounted in MF mode. aperture doesn't come into the equation at all.
Think about it! If it did EVERY different maximum aperture lens (f1.4, 1.7, 1.8, 2.0, 2.8, 4,.........) fitted to an adapter would need a different chip.

I must say,this little gadget is SOOOO cool!
Sure it doesn't do a whole lot that a good focus screen can do but what stuns me is just how small and simple he made it!!

I've read that the answer to this little dillema is to carefully file back the flange so that it doesn't trip the switch

With all due respect, this statement is IMHO very wrong. The AF mechanism uses similar principle to split prism - it needs two spatially separated "rays of light" to do it's work, and when physical aperture is too small, these "rays" just don't exist. Sorry can't explain it better now, but there are very good white papers available somewhere in the net.

You can easily check it by combining lenses and TC-s with total aperture of say f/11, cover the contacts so the camera's "artificial limitation" does not activate and try to autofocus on anything. Even the most well lit, brightest, contrasty object.

Edit: Here is a very nice paper on the subject of focusing and aperture.

Definitely every Canon EF lens will have to give different information to your camera ... so it know the exact data of the lens. Don't know if they use a different chip (but assume so) or just a different EPROM with the same communication chip.

But if the EF 24-70/2.8 L lens was taken to get the communication data, then the camera should believe that the largest aperture of the lens is f/2.8.

Best regards,
Andy

It's a very good article and I appreciate you posting this. If you have a link to more of them, I'd really appreciate it.

But the article doesn't contradict what I'm saying.

Notice how the article keeps talking about "effective AF aperture". Meaning that it's not the actual aperture of the lens that determines AF capability, but the angle from the AF sensor to the "edges" of the aperture.

If we're talking about the central AF point, then it's the aperture diameter over the rear nodal point of the lens.

For all practical purposes, we as photographers are using telecoverters with telephoto lenses. Canon and other manufacturers have done their best to decrease the size of the lenses they're making and thus had to move the nodal point much closer to the film plane than the focal length is.

So, in the end it depends on how the rear nodal point shifts upon attachment of the teleconverter. So the AF will work with some of the lenses and will only be hindered by because less light is available.

Not sure that you have this right - I don't think that ANY information on the aperture is used.
The chip must act as a feedback device to inform the camera that the signal being passed from camera to 'lens' (in reality just the circuit on the chip) comes back into the camera as indicating that the 'lens' has the MF/AF switch in the MF position. Therefore the AF circuit in the camera does not try to send AF signals to the 'lens' but keeps the focus confirmation circuits live.
ANY lens could have been used to establish the protocol for the MF/AF On/Off signals.
The aperture circuitry isn't live and thus no information as to Av is transmitted or requested.

Hope that first field results will be posted soon. By analyzing EXIF data from pictures taken with the assistance of this device we shall see which data it sends to camera. If data contains aperture and focal length value (obviously faked by the device) it will be clear why Canon didn't support AF confirmation with MF lenses.

Why ? They could put "zero" in these fields; they already do with the aperture to start with. There are no reasons not to have focus confirmation in the EOS bodies, and that little adapter thingy proves it. I've been iching to do something just like that for some time!

Camera decision when to lit focus confirmation LED may be critical. Maybe it needs to "know" lens data to determine if focus will be achieved - lens focal length and max. aperture similarly to AF algorithm. It is not irrelevant if lens used is 400mm F2.8 with shallow DOF or zoom at 24mm and F4 with deep DOF. In this case I think that this device will be even less accurate than AF lens in MF mode because lens data are faked.
Or maybe camera works much simpler, in which case there is really no reason for Canon to disable it (except market reasons of course ).