Double checking - Df and stop-down metering

Hi all,

 

I'm actually finally putting my feature request list together. One thing I had was the option on the Df to use stop-down metering - that is, use Pv to stop the lens down, read the meter, and optionally use AE-L to lock the meter reading with the lens stopped down. Some older film bodies could do this.

 

On, say, the D850, the meter takes a reading immediately before depth of field preview, and the meter doesn't work while stopped down (as far as I can tell, experimenting with a modern lens in spot meter mode). The viewfinder reports "AE-L" while the aperture is stopped down.

 

The benefit to stop-down metering is that on a lens with no aperture coupling (pre-AI), you can set the aperture on the lens and the camera doesn't need to know. Ironically it's only the aperture lever that causes a problem - on an old tilt-shift lens with no aperture lever, you can just stop down manually and the camera doesn't know it's using anything but a wide-open aperture.

 

The disadvantage compared with the "normal" Df approach of separately setting the aperture on the camera to match the aperture ring on the lens is that your exposure is locked. One could imagine the camera automatically stopping down, then metering, then flipping up the mirror - which would solve one problem, but delay exposure and mean you didn't know what settings the camera would pick in advance. Also I'm not sure that the current Df aperture system can do it - unlike the D8x0 series and single digit bodies, I believe you can't change aperture (moving the aperture lever) while in depth of field preview mode.

 

I'd had this as a feature request for a while (only relevant to the Df, since there are few lenses with aperture levers but no AI coupling which wouldn't crunch the aperture ring on other bodies). I've just had a crisis of confidence because I notice Ken Rockwell's compatibility information and Df review suggest that you can use stop-down metering. But he might not have tried it when he wrote that, and just assumed it worked. I can't see any specific statement in the manual.

 

Anyone with Df experience... does it work? (That is, if you use depth of field preview, does the meter still record changes in light?) I'm not sure you'll need a pre-AI lens mounted to make it work, but it would be more definitive.

 

Just trying to do my research before bothering both the members of this forum and Nikon with my list. Thanks!

It is not there. Using Df + 35mm Pre-Ai lens. When pressing dof preview while metering, exposure reading does not change, AE-L appears on bottom of viewfinder. Pressing AE-L button does not release locked exposure value, menu items on AE-L button behavior do not support idea of stopped down metering.

Thanks, Hapien - much appreciated. I found an old thread where I said I thought Ken was wrong, but I'm rusty on the Df, and couldn't find a smoking gun. One for the feature list in the (unlikely, I suspect) event of a Df2! (I'm intending to give the group a view of my list and then do a doodle poll for which features people care about before passing the info on to Nikon - if this is something that nobody wants, I don't want to be responsible for Nikon wasting time on it!)

Ive never understood the difference between the FTB light meter and whats used on DSLR cameras.

 

The differences aren't all that dramatic really. For one thing, almost every digital camera now lets you choose between matrix, centerweighted or spot metering patterns. Your FTb has a semi-spot or what Canon used to call "selective area measurement": 10% rectangle middle of screen that is halfway between spot and centerweighted. Current cameras in spot mode are about half the FTb rectangle, approx 5%-7% spot in center of screen. Choose that meter pattern, and you shouldn't notice very much difference at all from your FTb readings.

 

Centerweighted mode measures the whole frame, but bases most of its reading from the center third of the screen. This was the most common metering pattern among film SLRs prior to the development of matrix metering. Matrix is a very complicated computerized meter: hardly anyone really grasps how it works, they just use it because it tends to gives reliable auto-exposure results. Matrix takes multiple readings from all over the frame, constantly comparing what you're shooting to lighting examples stored in its microchip memory. Kind of a crude artificial intelligence: the meter "recognizes" common variations of backlit, side lit, and high contrast scenes, then automatically applies exposure compensation to the standard reading. With AF lenses I often leave matrix running, when using manual lenses I alternate between spot and centerweighted.

 

As far as those youTube videos: there might be factors that aren't obvious. Those old bazooka Vivitar super-teles often had completely manual preset diaphragms (no automatic stop down linkage). With such lenses, modern auto exposure cameras will work fine because they're measuring the actual light coming in thru the already stopped down aperture (or thru the varying max aperture as the lens is zoomed). Since you aren't invoking the depth-of-field lever or any dedicated stop down meter function, the camera doesn't pitch a hissy fit or try to lock exposure: it just reads the damned light and sets an auto shutter speed. A feat they seem incapable of when shooting electronic or auto-diaphragm lenses sopped down.

 

Auto ISO is a feature thats become increasingly sophisticated in newer cameras. What that does is add a third leg to possible automated meter settings: along with the traditional shutter speed and aperture, the camera can change its own ISO up and down the range. Modern sensors no longer really have a distinct "ISO" like film or older sensors did: its sort of vague. You can input a fixed ISO based on how you like to work, or allow the camera the flexibility to adjust its own ISO continually as it deems necessary. Eliminates the old "gotcha" from the film days when your AE camera would run out of apertures or shutter speeds in extreme situations.

 

This can be a valuable trick when dealing with a setup like those old long tele zooms: the camera can boost or reduce its ISO to suit the variable apertures along the zoom range, balanced against lighting conditions and whatever custom input you choose. Some cameras on auto ISO take over completely, others allow you to set range limits on the ISO variable (so it doesn't drop too low and miss a shot, or go so high the shot could look terrible from noise).

Edited February 8, 2020 by orsetto

my rebel 35mm had all three metering options, but only one ever made a decent photo inside buildings. Asking felt like a good idea.

 

But on the vivitar super telezoom lens ive gotten to play with, the aperture is just a fixed hole in a disk they put inside the body, lengthening the zoom length makes the aperture changed based upon the "f stop is a ratio between focal length and size of hole in aperture."

 

Yes - there are several lenses (notably the reflex Nikkors, but also the Petzval I own) that don't have apertures. The variable aperture thing applies to lenses with an aperture too, of course - which is why some are f/3.5-5.6, etc. This is another reason the camera controlling the aperture sometimes works better, unless you have a weird lens that moves the aperture ring around based on focal length and focus distance.

When I use Stop Down Metering on my Df with manual-aperture lenses, such as my adapted Schneider Retina lenses and the 35/2.8 Perspective-Control Nikkor: I set the focal length and maximum aperture of the lens into the camera with the non-CPU lens screen of the Menu, set the F-Stop wheel of the Camera to the widest aperture -or closest possible-, then set the aperture manually on the lens. For example, I code in the 50/1.9 Xenon as a 50/2, set the aperture as F2, then stop the lens down manually. The same is true of useing the 35/2.8 PC-Nikkor. The same should work for non-Ai lenses. I tend to use the exposure meter in un-coupled mode, I got used to doing this on my "Bullseye" Photomic F.

Brian - the issue is that non-AI F-mount lenses normally have the aperture held wide open by the aperture lever, and this is where the camera meters. If you use lenses that don't have an aperture lever (so you focus and compose at the stopped-down aperture, unless you manually change it), the camera will be looking through the lens at the shooting aperture when it meters.

 

Unless the lenses you mention do have aperture levers? But then I'm not sure what you mean by "stop down manually". I have a couple of (third-party tilt/shift) lenses that have no interconnect at all (in addition to a telescope adaptor), and have ways to toggle the aperture by hand; an old fish-eye I owned for Canon worked the same way.

You are right: the only way to meter with the Df with non-Ai lenses with an aperture coupling is to dial in the working F-Stop to the camera. The camera appears to latch the meter reading before the DOF preview is engaged. The only way I could use "stop down metering" with my non-Ai 55/1.2 Nikkor-SC was to use the lens release and partially rotate the lens for the aperture to stop down. I used my Ai 55/1.2 on the Df, set it to F4, to determine the reading, then put the 55/1.2 Nikkor SC on the camera- changed the setting to non-Ai. Setting the lens to F4, left the exporsure wheel to F1.2, then stopped down the lens using the DOF preview- reading was for the lens at F1.2, not F4 as expected and that I would get with my F2AS. Once I used the wheel on the camera to set to "F4", the reading matched that of the Ai lens. The only way I could get the correct reading with the camera set to F1.2 was to use the lens release and partial dismount the lens to close the aperture to F4 without using the DOF preview.

 

So: I believe the only way to do this is to block the aperture actuating lever before you put the lens on the camera. As you look at the back of the lens you will see the lever in the stop-down position. You would have to block it in that position to essentially turn the lens into a manual-aperture mode. The alternative is to have the lens converted to Ai coupling.

 

A change to the Df firmware would be required to provide metering with the DOF preview engaged. Too bad it is not open source.

Edited February 9, 2020 by Brian

Thanks for confirming Brian - and I agree. It's on my feature request list. :-)

Just checked on my Df to satisfy my own curiosity, and indeed AE-L does engage with DOF preview.

 

Double-setting the aperture isn't my favorite way to handle things, but it's at least workable and gives some degree of metering/automation with Pre-AI lenses...

Andrew!

When use the PB-6 bellow on my Df, I can stop the lens down without using the Pv (and it doesn't work any way) so the meter would actually does stop down metering. However for some reason it's not accurate. I will have to do a controlled test to report how much error this technique would cause.

Thanks, BeBu. I gather there's an accessory that lets you simultaneously activate the stop-down on the lens and use a threaded shutter release on the camera (not for some reason hooking to the camera's aperture lever), but otherwise you'd be doing stop-down metering, as you say. So long as the Df has the aperture set to the maximum aperture you entered for the lens (I've not bothered putting "make a dSLR with an AI-s maximum aperture post reader" on my requests list) the camera shouldn't have a way to know the difference between a dim surrounding and a stopped-down lens.

 

Technically that's not quite true for matrix metering, which I believe behaves differently based on what it thinks the actual brightness of various parts of the scene is - so a different aperture would confuse it (as would an ND filter). Shouldn't affect spot or centre-weighted metering, though.

Thanks, BeBu. I gather there's an accessory that lets you simultaneously activate the stop-down on the lens and use a threaded shutter release on the camera (not for some reason hooking to the camera's aperture lever), but otherwise you'd be doing stop-down metering, as you say. So long as the Df has the aperture set to the maximum aperture you entered for the lens (I've not bothered putting "make a dSLR with an AI-s maximum aperture post reader" on my requests list) the camera shouldn't have a way to know the difference between a dim surrounding and a stopped-down lens.

 

Technically that's not quite true for matrix metering, which I believe behaves differently based on what it thinks the actual brightness of various parts of the scene is - so a different aperture would confuse it (as would an ND filter). Shouldn't affect spot or centre-weighted metering, though.

A couple of issues with the matrix meter.

Nikon published quite a lot of information on how the old FA matrix (or AMP) meter worked. It used the absolute scene brightness as a method of classifying the scene and it also ensured that parts of the scene brighter than LV 16 1/3 were given reduced weight or even ignored. Some of this persists to the present day since recent manuals state "values over 16 1/3 EV are reduced to 16 1/3 EV." Note that 16 1/3 LV is 16 1/3 EV at ISO 100.

 

Another point is that if you put a lens on a bellows the exit pupil moves forward and it's my belief that short focal length lenses need a correction for matrix metering since an exit pupil close to the film/sensor plane will result in a significant cos^4 falloff in illumination. My guess is that the camera uses the supplied focal length and aperture as an index into a database, retrieving the exit pupil position. Only a rough index, of course, since the identification is not unique. If the exit pupil moves forward then the falloff will be reduced and any correction may be excessive.

 

I believe that the above argument is supported by the statement in my manual that the accuracy of CW and spot metering is improved if both focal length and aperture are supplied.

 

Nikon used to supply a BR-4 ring that worked the aperture stop down lever via a cable release. There was also a double cable release that worked the camera release and also the BR-4 aperture control or the aperture control part of the PB-6 bellows. I've not seen any of these things but imagine that the lens is stopped down in a slow relaxed fashion prior to releasing the shutter on the camera - so this doesn't really make any difference to the Df stop down issues.