D810 and light metering

Hi I have a question about the d810 and light metering which I hope won't try the patience of absolute Nikon d810 experts! I come from a background of using spot meters and film cameras, so one thing which wore out my patience a little bit with the modern digital camera was the approach to metering - essentially, I would like to choose what is the mid tone I require in a picture, get a reading off that, reframe and then take the picture. Of course it is possible to do that using the camera in a Manual mode and that worked pretty well for me, until I discovered Auto ISO. I've got to admit I love auto ISO. It's so fast! Of course, using Auto ISO means you're not fully manual and so therefore the metering is done after you've framed the picture. Hmmm.

 

I expect most photogs use back button focussing on the d810 - I do. It's great. Find the thing you definitely want to get in focus, focus, take your thumb off the button, then reframe. I love it. What I'd really like to do is the same thing with metering... find the thing I think is the mid-tone (or choose it by whatever method you will) Take your finger off (whatever button you used to set the auto ISO) reframe, then shoot.

 

Is there any way of achieving this? I know I can do it by using the camera in Manual, but there are so many circumstances in which the auto ISO is better than fumbling with buttons.

 

Thanks for reading and if any more experienced Nikon guy or gal can suggest a solution, I'd be grateful

Seems to me the easiest way would be to set c1: Shutter-Release Button AE-L to ON. Then, as long as you keep the shutter release pressed halfway, the exposure will remain locked. May not be what you are looking for, since you need to keep a button pressed rather than release one. <br><br> The option that comes closer to that involved the AE-L/AF-L button that can be customized via f6 to AE-Lock/Hold which will lock the exposure upon pressing the button and needs a second button push to release the exposure lock. Definitely the better option when working off a tripod.

Speaking from the other end of the photo spectrum, as a long time user of the lowly D3200, and now a D7100, I've always done it as Djschaefer does, and found it just fine. Focus first, then find your spot metering spot, get the reading and use the shutter button for AE hold (since on these cameras the AE/AF lock button is turned over to AF ON). Then recompose and shoot. You have to enable shutter button AE hold in the menu for this.

 

But this is on a camera that does not have a separate AF ON button, and takes over the AE/AF hold button for focusing. If you have a separate AF-On button, then I think you should be able in the menu to set the AE lock to do only AE and "hold," whereupon it will lock AE on one push, and unlock it on the second. AE hold is available even on the D3200, so I am guessing it's available everywhere.

Yes that AE lock thing works. Thank you both very much for replying. By the way Matthew, you are not at the other end of the photo spectrum - I have a Brownie on a shelf here. Now that's the other end of the spectrum.

Pegging mid-tones doesn't work well with digital Tom. You have to reverse the film mindset, which was "expose for the shadows". With digital you have to get the highlights under control.

 

The reason being that once highlights have "blown" there's no way you can recover them, whereas there's a lot you can do to bring up shadow detail in post-processing.

 

I'd recommend using the technique called "expose to the right" or ETTR for short. This means making sure that the histogram the camera shows doesn't bunch up at the right hand-side, but extends just to the right-hand edge, or maybe a tad to the left of that. This ensures you get the maximum amount of tonal detail retained in your images.

 

There's too much to explain in a single post. I suggest you do a search for ETTR and digital exposure techniques, as well as using the RAW file format. But basically digital ain't film. However once you get the hang of it, especially post processing, it can give you far more tonal control than film ever could.

Edited February 12, 2017 by rodeo_joe|1

Thanks for that Joe. I was aware of it and actually do make some allowances (ie in what I rate as mid tone, which isn't what I would rate as mid tone for film). You'll see I do a lot of shooting into the light in the pictures I have already uploaded to this site. I'm not sure about 'getting the highlights under control' and ETTR though if it is to be a portmanteau rule. The photographer still has to find a balance and there are still decisions to be made. According to DXO the camera has almost 15 stops of dynamic range - but I'm not in the least convinced all of that is useful. It's a point well made, could be an interesting debate and I am very grateful to you for bringing it up.

When I am shooting a scene that is a pretty wide dynamic range and I don't want to do hdr, I use a meter that is calibrated to my camera and it shows the clipping points for the camera. It enables me to expose as far to the left as possible and still get detail in the highlights, ie keeping the highlights from crossing the clipping point. I expect your camera has at least 6 stops of detail so you could spot meter the brightest highlight in which you want detail. A reflective reading places it at zone 5, 18% gray, and you can increase exposure 3 stops to get it to zone 8. That should place the highlight as close to actual clipping as possible. The lcd, blinkies and histogram are based on conversion to a jpeg so aren't precisely accurate. They are a ball park. If I am set up and am watching the light for a land/seascape. I have time to meter and it saves messing around in post or sorting through bracketed shots.

Zone V isn't 18% reflectance. Not if you count zone viii as pure white. Ansel's maths doesn't add up.

<p>

If each zone is one stop, and you count zone viii as pure white (i.e.100% Lambertian reflectance), then that places zone v at 12.5% reflectance. That's a half-stop discrepency.

<p>

The old Weston meters had it about right. The "O" and "U" markings were 7 stops apart. With O(ver) being 3 stops above the C(orrect) reading, and U(nder) being 4 stops below. But of course a digital camera can dig out shadow detail much deeper than that nowadays.

<p>

I'd suggest calibrating the camera to see exactly where it clips whites and pegging mid-grey (zone v if you must) 3 stops below that. If you then meter for the brightest highlight where you want detail, you'll know your highlights won't be blown.

<p>

The human eye isn't good at judging reflectance values, especially brightly coloured ones. So "mid grey", as judged by the eye, is a very moveable feast to be pegging an exposure value on. It's debatable whether the zone system has any value at all outside of B&W sheet film use.

When the post-shot Histogram shows 'Blinkies,,could that not be from the RAW,? I'm sure the maths can't be that hard internally.

Mike, RAW files typically have between 1.5 to 2 stops more headroom than their JPEG version, and it's the JPEG that's used for the histogram and triggers the "blinkies".

<p><br>

Technically, that may be a bit of an oversimplification, but from a practical point of view it's pretty close.<p><br>

My view is that you should consider a RAW file as comparable to a film negative, and its post-processing as a combined development and printing process. Except unlike film you have the same "developing" flexibility in colour as you do in B&W.

FWIW, I absolutely expose for the highlights and then recover the exposure in post, because the dynamic range of the D810 (at low ISO) is good enough to let me do so. Hence I'm usually in highlight priority mode, then exposure compensated to deal with the fact that highlight priority doesn't always avoid blowing the highlights (for... reasons?) What I'd really like is a true ETTR for the raw (expose my raw file so that no channels blow, as best the meter can manage), and for the associated JPEG (I shoot JPEG+raw, but there's also an embedded JPEG in the raw file) to be push-processed to match the matrix meter's opinion. That way I'd usually have a usable JPEG along with the most flexible raw file.<br />

<br />

Of course, if I'm too fussy I'll actually spot meter, and if the scene isn't too contrasty I'll just use the matrix. But I can dream about everything working perfectly.

Been waiting for Nikon (or any company for that matter) to bring a histogram that's based on the RAW file and not the JPEG. Naturally, I'd actually want to see three histograms, one for each of the R,G, and B channels. Shouldn't be too hard to bin the RAW luminosity values that way - yet such a histogram remains elusive and I have to resort to "guessing" how much I can push the exposure to the right before things can't be recovered in post. Of course, I could spot meter the brightest portion of the scene to get an idea, but in many scenarios that's not really a viable option.

+1 Dieter, RJ and Andrew. Looks like there's something useful Nikon could implement in the next FX body.

 

Maybe I'll do a little experiment and shoot the same still life and keep increasing the exposure until I get 'Blinkies' and keep going and see when there's nothing left to recover in DxO. Is there any reason, apart from the channels issue, that the 'headroom' (thanks RJ!) is not a finite amount?

One of my proposals in recent idea discussions (yes, poll coming, very busy) was to split the in-finder meter into three, then use the intensity of the segments as a histogram. Probably easier with the OLED viewfinder on the D810 than on the LCD versions. I do like - and only recently realised that this worked - that the RGB histogram view when you chimp an image reflects the area you're currently zooming to. Basing the exposure off the JPEG does mean that it depends on what the camera has done with white balance (I usually shoot AWB for the most likely usable JPEG, but I fix up white balance in post if needed; don't get me started on Nikon scaling their raw values). My bigger problem with the D810 is that highlight priority seems too keen to selectively ignore some highlights when metering; I understand the need to allow the occasional spotlight to blow, but "ETTR and I really mean it" would be a very useful mode when I get sick of blown clouds. And, in recent shooting, blown bald eagle heads, blown mountain goats, and blown otters. I learnt in my most recent trip and exposure compensated everything after chimping a lot more, but it shouldn't have been necessary. I swear I had less trouble just using the matrix on my D700...

Mike, sometimes it's possible that very saturated colours "blow" one or more of the RGB channels irrecoverably before the other(s).<p>

In such a case the headroom may be much lower than with a more neutral "white". The same happens with unusually extreme white balance settings.<p>

I'm not convinced that matrix metering copes adequately with that circumstance. In fact, because non-LiveView metering is done after the viewing screen, I'm not convinced it's entirely accurate full stop. I'm pretty sure that metering with lens apertures wider than f/1.8 is fudged, since the viewfinder optics are incapable of showing any brightness change at apertures wider than f/1.8.<p>

In evidence of the above. If you fit an f/1.4 or f/1.2 non-cpu lens, but tell the camera it has a smaller aperture, you'll get overexposure. I believe this is because the camera simply subtracts any aperture above 1.8 from the metered exposure to compensate for the fact it can't actually read the screen properly. If you follow.

Edited February 14, 2017 by rodeo_joe|1