I would like to know if you calibrated your camera's focussing screen for optimum MF accuracy. If you choose Option 4 or 5, please share how you performed the calibration.
I cannot choose any option, because I have had different experiences with different cameras.
Sometimes the precision matte or split prism screen just works. Other times it needs to be calibrated. My Ee-S in my 5D worked perfectly out of the box. My split-screens for my 10D, XTs, 20D and E-330 all had to be calibrated.
For the 10D, XTs and 20D I had to add some shimming material (slices of scotch tape or other tapes of various thickness along the stepped edges). With my E-330, I needed to remove some material from all four edges of the screen (there was no shim to remove).
The testing procedure is simply to take some manually focused shots of a ruler on an angle, focusing by eye precisely on one number (I like "14"). The resulting image will either be "bang on", back-focused, or front focused. Add shims or remove shims (or material) and use trial and error until accurate focus is attained.
For Canons with replaceable screens, a set of properly-sized shims of various thickness can be purchased. I've never used them, but I can see how handy they'd be.
I focused with a 1.4 50mm Planar c/y lens on a ruler at 70 cm with and without the shim in place and took two pictures. From the two focus points (with shim 2cm front focus without shim 8cm back focus) I calculated that my shim was 20% too thick. I had a .25mm shim which I exchanged for a .20mm shim. Now focus is spot on. I use an EE=S screen for my 5D.
I presume when you are discussing calibrating a MF screen, you are going to compare it to the downloaded file or to live view 10X? I personally find the EG-S screen to be much harder to accurately and precisely focus than Live view 5-10X. thus when on a tripod and MF, Live view it is. When hand held with a MF lens (Zeiss ZE or Canon TSE) I usually use focus confirm, which is calibrated Vs Live view using the focus MA feature.
Mike K
Ah yes, always forgot to cover for some situations in these polls. I should've listed LV and AF confirm at least under the first option (i.e. don't use the focussing screen for MF).
The trial and error method is what puts me off of calibrating my screen, but I have confirmed I need to adjust it. hjwhoub's method sounds good, I'll give it a try.
Had a 5D spot on out of the box. 5D2 had to have a shim change, then was zeroed in. Now I have a 1Ds3 that doesn't even have shims (still trying to figure out how to calibrate this one).
5D1 + Ee-S. I used a 3-D focus chart, with fast telephoto, normal, and wide angle lenses. Wide angle lens was best indicator for me -- I'd front focused it and at first thought I'd be shimming; luckily I decided to test it again (twice, on different days) before going ahead with shimming and screwing it up.
I calibrated my filed F2 screen with two layers of scotch tape at the ends, to get focus right on my D700. It was very close with the original shim, but not good enough for me.
jan_haidn wrote:
I can not explain it. It was the same with the Km. It might have to do with our glases. I am using varifocals, she fixed.
The calibration of the focus screen in relation to the sensor has nothing to do with eyeglasses or diopter settings, so this situation is very puzzling. If you said that "it looks sharper" for your wife than it does for you, then that would have something to do with your glasses or the diopter setting, but for it to front-focus for one person and not the other, that is simply bizarre. I am not saying you are intentionally trying to deceive anyone, simply that I cannot explain it myself and I am not sure if there is any physical principle(s) and corresponding rational explanation as to why this would happen.
The only explanation I can come up with is that they are using different focussing "algorithms". One may always focus coming from infinity and stop as soon as it appears sharp (and maybe always overshoot slightly) while the other comes from the front etc...
Ok I put together a little set of charts to see what effect a miscalibrated screen would have for MF. Siriusdogstar was right, wide angle lenses produce a more pronounced effect, which makes sense when you think about it. Landscape (i.e. infinity focus) is also more affected than macro.
My 5D, 1DsIII and 1DIII bodies all do manual focus very well with their standard screens. When circumstances permit, I often use LiveView. I rarely use MF lenses on my 40D. My 20D has a Haoda split-prism focus screen that works quite well; otherwise, it blows for MF.
I'm really surprised that so may people were able to manually focus with their cameras without having to re-shim the focus screen.
I have only tried manually focusing on 3 of the bodies I've owned, a 5D, a 5D2 and an XTI, and none of them could be focused accurately using the view finder withouit re-shimming. Not surprising on the XTI, and nothing you can do about it, but I thought my 5D and 5D2 bodies would be better. I do wear corrective lenses (progressive), so maybe that has something to do with it.
Just realized you asked for those of us who voted 4 or 5 to share how we performed the calibration.
The 5D was a bit tricky. I put the camera on a tripod and used auto focus first, then turned the focus ring one way or the other until the view finder looked sharp. I honestly don't remember the relationship between adding or removing shims (washers) and which way I was twisting the focus ring to get things clear. In the end it was trial and error using different shim thicknesses until I got it right.
The 5D2 was much easier. I did the opposite with it, I focused using the view finder first, then put the camera in live view 10x and twisted the focus ring one way or the other to get the image sharp. If I recall correctly a thinner shim was required when I had to turn the focus ring toward infinity, and a thicker shim if I had to turn the focus ring the opposite direction. Again there was a certain degree of trial and error to it, but once I got close I knew exactly which way to go.
I didn't use a ruler at an angle or a 3d focus chart. I put the camera on a tripod and focused on a street sign across the street and 2 houses down from me. I found it a perfect object to focus on. Once I got consistent results between view finder and live view I did more tests focusing on other targets closer and nearer to confirm I had it right.
One thing I will add is that it is very difficult if not impossible to focus as accurately with the view finder as you can with live view. Once I got close I had to keep checking back and forth between live view and view finder before making any further shim adjustments.
