Fuji M adapter and wideangle infinity focus?

Dante_Stella

Rex canum cattorumque
Local time
11:05 AM
Joined
Apr 28, 2007
Messages
1,862
I just received one of these Fuji adapters this morning. My lenses are all in good trim, having been collimated against neutral references and validated on an M8 for correct focus.

One thing that is apparent is that on the copy of the adapter I have, the register distance is shorter than 27.95mm - and the way I figured this out is that a 21/3.4 hits infinity at the 10m mark, the 35/1.4 hits at between 10m and ∞, and the 90/2.8 hits at ∞ (my target is a mile away, so it is very easy). Given the relative focal lengths, this seems to point to a constant error that has to be coming from the adapter.

It would not be surprising to me if the adapter were designed that way - just as many cheaper adapters err on the side of being thinner to allow all lenses to focus to infinity. Otherwise, faster lenses with focus shift (whose normal tendency is to front-focus wide-open), particularly telephotos (whose shift is worst), would not reach infinity by the time the lens reached its hard stop.

Has anyone with a 21/2.8 or 35/1.4 lens tested the Fuji adapter for infinity focus? I suspect that the type of error I am seeing would not be visible on an f/4.5 lens due to the greater depth of focus (and in my testing, it also tends to disappear by stopping down to about there). And in the real world, it's not an error that is going to make much difference to zone-focusing wideangles (because lens extension changes are much larger at close range) or EVF-ing longer ones (since you're TTL anyway).

But I do want to make sure I am not dealing with an abnormal sample of this adapter.

Thanks
Dante
 
Yikes. Every cheap M and m39 adapter for m4/3 I tried had that problem, but I would hope Fuji got it right! The m39 adapters were easy to fix, I would simply add shims until I got it spot on, then I would leave the adapter in place. I need to know about this before I buy into Fuji X....
 
I just received one of these Fuji adapters this morning. My lenses are all in good trim, having been collimated against neutral references and validated on an M8 for correct focus.

One thing that is apparent is that on the copy of the adapter I have, the register distance is shorter than 27.95mm - and the way I figured this out is that a 21/3.4 hits infinity at the 10m mark, the 35/1.4 hits at between 10m and ∞, and the 90/2.8 hits at ∞ (my target is a mile away, so it is very easy). Given the relative focal lengths, this seems to point to a constant error that has to be coming from the adapter.

It would not be surprising to me if the adapter were designed that way - just as many cheaper adapters err on the side of being thinner to allow all lenses to focus to infinity. Otherwise, faster lenses with focus shift (whose normal tendency is to front-focus wide-open), particularly telephotos (whose shift is worst), would not reach infinity by the time the lens reached its hard stop.

Has anyone with a 21/2.8 or 35/1.4 lens tested the Fuji adapter for infinity focus? I suspect that the type of error I am seeing would not be visible on an f/4.5 lens due to the greater depth of focus (and in my testing, it also tends to disappear by stopping down to about there). And in the real world, it's not an error that is going to make much difference to zone-focusing wideangles (because lens extension changes are much larger at close range) or EVF-ing longer ones (since you're TTL anyway).

But I do want to make sure I am not dealing with an abnormal sample of this adapter.

Thanks
Dante

Sean Reid reported the same thing.

He also found the Rayqual Leica M to FujiX adapter spot on at infinity.

Stephen
 
Thanks Stephen.

So I measured the adapter. I'm not sure how it sits on the lens mount, but taking the thickest part of the adapter (front surface to the "lip" around the back), I get .396", or 10.02mm. Let's assume that the Leica spec is 27.80mm (that's what Leica lenses are calibrated to hit). The difference is 17.78. The Fuji XF flange is 17.7mm (and who knows, it may be 17.74 or something). That means that by the numbers, the adapter is at least 0.04mm short and possibly 0.08mm short, right? That's quite a bit for a relatively fast 21mm lens or a really fast 35mm lens, enough to render the focusing scale inaccurate and make the infinity stop noticeably inaccurate.

There are a lot of arbitrary decisions you could make in making an adapter. Since fast lenses tend to front-focus wide-open, you might pick one that's slightly shorter. If you make it dead-on, then a 90/2.8, for example, would never get to infinity at full aperture. Fuji probably counted on people zone focusing slow wide-angles or EVF focusing telephoto lenses. With those you might never detect a problem.

Of course, the adapter is made in China and looks like some other Chinese adapters (big surprise). The electronics package is absurdly big; a 35 Summilux ASPH barely fits in the throat of it. The function button, however, is pretty cool.

Dante
 
Does the Fuji adapter put the lens information in the image file metadata? If so, doe it only work with specific Leica lenses or can the information for any lens be entered? That would be the only reason I would choose it over another adapter with accurate register depth.
 
Yes; you select the lens when you mount it; the focal length goes into the Exif. The adapter does not read Leica codes; you just preselect your lens' focal lengths ahead of time. Works with any make.

And I suspect that when the Zeiss 12mm and Fuji 14mm hit the market, there won't be a lot of excitement surrounding adapted wides.

Dante
 
Back
Top Bottom