Hooray, D7100 focusing screens!

Discussion in 'Nikon' started by PapaTango, Sep 4, 2019.

  1. PapaTango

    PapaTango Itinerant Philosopher

    It was a tiny bug. Crawling about inside the of the optical pentaprism chain. How it got there, I am clueless--but the extraction job involved looking at everything. One of the first things I noticed was that there was an easily accessible metal bail holding the focusing screen in place. Frankly, I have never liked a solid matte screen--and research, when I got the camera several years ago, did not indicate that there were any choices.

    Imagine my surprise when I found that a quick push with the roach clip (err, I mean hemostat :rolleyes:) dropped this little gem in my hand. A quick bit of research on the innernutz showed me that there were MANY possible replacements--ranging from actual ground glass, to split and microprism ring configurations!

    I note that there are ordering options available--ranging from fleabay to Taiwan direct. Have any of you had experience in adding one of these--and recommendations? I think that I want the split center and a microprism ring.
  2. Bear in mind that Nikon's metering is housed in the prism and above the screen. Do you really want the matrix or spot metering completely messed up? For the dubious advantage of a split dark ring image screen and arguably better manual focus.
  3. PapaTango

    PapaTango Itinerant Philosopher

    Bummer... :(

    I had not extrapolated that older camera's using this sort of focus were configured to compensate for the screen. But how then in systems such as the Nikon F (Photomic for instance) could one interchange screens and maintain metering accuracy? I used one such camera and different screens for macro, slide duplication, and general shooting. Perhaps again, it was the sort of metering sensor/circuit that allowed that.

    The split/microprism would still be good for me doing digitization of film media--as I do every setting including focus manually there.
  4. The classic manual film-era Nikons like F/F2 Photomic did not have sophisticated, computerized, self-configuring, SkyNet-level matrix metering. They used a simple, centerweighted-always metering area, and the presence or absence of brighter central focus aids did not significantly impact the reading. A very few specialized screens that were much brighter overall (clear glass crosshair, full-screen microprism surface, etc) could be compensated easily by using the special ISO offsets on the film speed dial.

    Modern electronic DSLRs (and some late-period film SLRs) place exaggerated metering calculation directly over the area where a split image aid would go. So any irregularity dead center on the screen drives the meter bananas. Perverse, really: for all their cleverness with 144,000 pre-programmed data points, they can't immediately recognize "oh, theres a distinct perfectly circular anomaly center screen which is probably an optional focus aid: let me factor that in". SMH: this has been a needless obstacle to getting decent focus screens for years and years.

    OTOH, each camera and photographer reacts differently: the meter deviation isn't consistent. You might not experience any issues at all, or be able to compensate fairly easily. Most optional screens are cheap nowadays: the premium Katz Eye that sold for over $100 is long gone. Try one of the screens and see how it works for you: with some cameras, you need to disable matrix metering and/or certain program-AE modes, others (even the identical model) will meter perfectly fine in all modes. Its a crapshoot, depending on camera, lens, and shooting conditions. Most photographers can at least get centerweighted mode to work reasonably well.

    The bigger problem is the screens available now just aren't very good. Mirrorless cameras with EVF decimated the premium manual-focus optical screen industry overnight, so all thats left are Chinese cut-rate screens. These can be way off spec, requiring a lot of fitting and shimming. The more you fuss with the screen, the greater chance you will damage it (or worse, the base of the camera prism or bail wire). You should know within minutes of first install whether the screen will be worth adjusting: if its murky and hard to use, optimizing its fit won't make it any better. If this is a large part of your work, consider getting a used Sony A6000 (or similar DX-sensor mirrorles) as a backup body.
    PapaTango likes this.
  5. PapaTango

    PapaTango Itinerant Philosopher

    Thanks to @rodeo_joe|1 & @orsetto for responding--and kudos to the latter for a well-developed explanation. This all got me to looking at how the focus and metering chains work. I had a mistaken concept that the metering was accomplished through the sub-mirror, the same chain as autofocus. Now obviously, not so... o_O

    It also explains why given it is so easy to remove the focusing screen in the D7100 there are no OEM Nikon replacements hawked at us.

    In the final analysis, it is a crapshoot impacted by a number of variables that result in outcomes ranging from acceptable to rubbish. Another caveat emptor for those of us with GAS! :p
  6. One minor $0.02: I vaguely get wanting to use a split prism for manual focus lenses in the field if that's how you like to work, but for digitizing film, I'm going to say you really ought just to use live view. Even if you can get perfect 24MP accuracy out of your eyesight on the prism, it's still putting a mirror in the optical path; the film isn't going anywhere, so you may as well see what the pixels look like before you commit.
    PapaTango likes this.
  7. PapaTango

    PapaTango Itinerant Philosopher

    @Andrew Garrard I do use such a method when working with MF & 4x5 negatives. I have built a copy system for these comprised of an old Beseler 45MXT motor frame, an LED panel backlight, and a frame that holds Beseler negative carriers perfectly aligned to the light and the camera plane. When doing this the software is digiCamControl on my Surface. But for 35mm (and where I think the split/micro would be nice) is in full manual mode with an old Spiratone Vario-Dupliscope tacked on.

    I figure at the rate my library of film images are being digitized, I should finish about two years after I am dead... :eek:
    Andrew Garrard likes this.
  8. PapaTango

    PapaTango Itinerant Philosopher

    The verdict is in--at least on the D7100. I received my China Post packet today. The first order was to establish a small series of test exposures under differing light conditions. Shots were made, and the speed & stops chosen by the camera recorded. The screen was then swapped out. In well-lit examples, there was about a stop difference, and if the stop remained the same, a decrease in speed. As illumination levels (such as those taken indoors) decreased, the lens jumped by a full stop, and the speed decreased by half or more. The result of all was varying levels of overexposure. :(

    Now I own just one more perfectly useless gizmo...
    ericphelps likes this.
  9. Oh dear. Quick suggestion: if you spot meter, with the spot off to one side of any split prism, do you get any consistency? I wonder whether dodging the "problem area" might still give you some functionality.
  10. Just for the record, as I recall, some of the focusing screens available for the old F did result in TTL metering errors, and warnings were made for this. One of the other things I recall is that the split prism focusing aid was tailored to lens speed, and the OEM center prism not only went dark but was not entirely accurate with lenses slower than F4. There was a special split prism screen for slower lenses. I had one for the 400/5.6.

    Nowadays, with the slower DX lenses we tend to see on a D7100, I'd be hesitant to use a prismatic screen even if it does expose correctly.
    rodeo_joe|1 likes this.
  11. Only magnified LiveView is accurate enough to get a grain-sharp copy IME. Even then, you have to watch out for focus-shift on stopping down. So focussing at the working aperture is advisable.

    I'm actually in the middle of digitising my old film. Mainly colour negative, which is a bit of a pain to get colour accurate in the positive.

    FWIW, I'm using an enlarging lens for copying on the end of a bellows. Enlarging lenses were computed for ratios in the region of 1:1, and most of the decent 6 element ones have proven ideal for the task. I did try a couple of macro/Micro lenses, but they showed a surprising amount of field curvature and needed a lot of stopping down to overcome it. No such issue with any of the enlarging lenses I tried.

    Plus you can pick up a top quality enlarging lens for a pittance!
    Sorry to hear that.
    I suppose it would be very churlish to say 'I told you so'?
    Last edited: Sep 13, 2019
    PapaTango likes this.
  12. That's disappointing - my understanding was that one of the design goals of macro lenses, especially the shorter focal length ones, was a flat field, specifically for copy work.

    I agree that at least live view should still work. :)
  13. I used replacement focus screens from Katz Eye Optics (they closed) on the D200 and D700 and was very happy with the result. No obvious metering issues and excellent manual focusing. It is however possible, that shimming may be required since the original screen doesn't give good contrast for accurate manual focus, its position may be a little off, depending on the camera. I think it is quite possible that you may find a good solution for manual focus by replacing the screen, but I haven't used other brands of focusing screens so I can't tell what they are like.
  14. I was not sure, but thinking the same thing.
  15. Well, there are degrees of field curvature. On any 3D subject, even coins and such, I doubt you'd notice anything, but when the few microns bend in a piece of film gets significant you do start to get a bit picky about field flatness. Plus it varies quite a lot with distance.

    I would never contemplate getting rid of my old 55mm f/3.5 Micro-Nikkor..... but when you find that an 80mm Durst Neonon, bought for £10, does the job you need even better, well it then frees up the Micro-Nikkor for tasks more suited to its talents.
    Andrew Garrard likes this.
  16. Just for the record, since there seems to be some confusion and mixed memories on this point: the metering system in the F and F2 prisms was never flummoxed by any of the optional focus screens. Being a completely analog, fixed, centerweighted circuit, there was no issue with the metering pattern being disrupted by focus aids or brighter/darker matte. With a select few exotic screens that were significantly brighter or dimmer than the usual standard, overall exposure readings needed to be compensated by biasing the ISO film speed setting. Nikon provided dedicated engraved index marks on the rim just for this purpose. Once set to match the installed screen, metering could be performed as usual with no additional calculations or corrections required.

    Below are the instruction manual pages for adjusting the F and F2 meter compensation for various screens: dead simple. The chart can be reduced to a single-sentence advisory: only the oddball, rarely seen or used microprism-covered G and H screens, and clear crosshair screens, require compensating the film speed setting. All other plain, checker grid, split image and/or microprism screens can be swapped at will with no compensation needed.

    The color-coded caveats re some of the (now-ancient) extreme tele, fisheye and shift lenses flag issues that were inherent to those lenses, not the meter (i.e., some lenses were preset-aperture so never meter-coupled at all, the long slow super teles needed special split-image screens to even see a split image, shift lenses were not compatible with split image, circular image fisheyes need separate meter compensation for their blacked out frame borders, etc).

    Nikon Screen Exp Comp 1.jpg Nikon Screen Exp Comp 2.jpg

    This is separate and distinct from the issues with Nikon DSLRs, the FA, and later AF film SLRs, with computerized matrix metering. Those circuits do get unpredictably thrown way off by different screens, esp those with central focus aids. The brighter or darker focus aids interfere with the programming of the matrix and spot features of the elaborate, more recent meters (center mode may or may not be more compatible).
    Last edited: Nov 25, 2019

Share This Page