How best to use spot meter in lf colour landscape

[brijphoto]

Hi,

I would like some advice on how to use a spot meter when shooting landscapes in color. Where are the best parts of the scene to make a reading from and how best to follow on from there?
Is there a method that can work in most cases for shooting landscapes in the early morning or late evening magic hour light?

Thank you for any advice

[David Evans]

Hi,

I would like some advice on how to use a spot meter when shooting landscapes in color. Where are the best parts of the scene to make a reading from and how best to follow on from there?
Is there a method that can work in most cases for shooting landscapes in the early morning or late evening magic hour light?

Thank you for any advice

I don't count myself as anything like an expert in this area, but here goes.
First I take a few meter readings, and specifically of the brightest part (sky) and the darkest part (shadow or dark water), and see what range I have in the scene.
I will also try and take what I think is an average reading (generally mid-green grass, or where I live some limestone out of direct sunlight). I use the average reading for guidance rather than using specifically in calculating the exposure to use.
I then work out how I can reduce the exposure range with ND grad filter(s) - usually on the sky.
At this stage I used to (but don't any more) meter about two stops down from the highlight (as adjusted by the ND grad filter) but that resulted in a lot of underexposed images.
After discussing with Joanna, I now meter about two stops up from the lowest reading, and this has resulted in much more consistent exposures.
With transparency film I work on getting the range of exposure within five stops maximum, and preferably four stops.
I rarely used to use a polarizing filter, but do increasingly now, and adjust the exposure for loss of light from polarizing.
If there is snow or ice that needs adjusting for also.
Be very careful early morning and late evening, as light levels change very rapidly.
Finally if I think the shot might be special, I will shoot two or three shots at slightly different exposures (not necessarily bracketed either side of the main exposure).

[dave_whatever]

I would like some advice on how to use a spot meter when shooting landscapes in color.

Are we talking transparency film or negative?

What David says above is all good advice. Generally I establish if the scene (or rather, how I wish to render the scene) is within the range of the film, then worry about the actual exposure and grads, filters etc.

The only key bit of advice I could give is get your meter setup to read out through the viewfinder in EVs rather than aperture/shutter speed. That way you can work in simple numeric values when you're assessing the contrast range of a scene, and just let the meter pop out the aperture/shutter speed for you at the end.

I would suggest taking time to establish a way of working that works for you though rather than just applying someone else's method blindly, and always work to the same method, then it becomes second nature.

[timparkin]

My first recommendation is to use a lightmeter that reads in EV's and has some form of mechanical calculator. This allows you to see and tweak the range of exposure values visually..

The first thing I do is to meter the scene and pick out areas that I want to appear at a particular tone. For instance, I might have some rocks in the foreground where I want the highlights at zone 7 (+1 stop) and the general body of the rock at zone V (a mid tone +0). For the record, I don't use the ansel adams equivalent zone values anymore, I just use number of stops above or below midtone.

I would read the value from the midtone of the rock and then apply it to the mechanical calculator. An example on the Pentax meter is on this page..

here is a picture of the pentax meter

http://bp2.blogger.com/_hAVtpP4knOY/R7g ... e-zoom.jpg

and here is the updated sticker I now use which is calibrated for velvia.

http://www.timparkin.co.uk/blog/926815673944184070

Once I have set my EV reading against the zone I want it to appear at, I can then check other parts of the picture and see what tone they would be.. e.g. I then take a reading of the sky and work out what tone it would appear at.. e.g. lets say the sky appears at +3 stops, I can quickly see that I actually want the sky to be at +2 stops.. This means I need a 1 stop grad..

I can also check the shadows in the foreground and if I find out that they are -4, I can place a one stop grad over the whole picture and increase the exposure by a stop..

It's easy using the scale to take a quick EV reading and see what tone it corresponds to..

I'm writing a blog post about taking exposure readings so hopefully this may help ...

[Charles Twist]

Hello Brijesh et al,
I think you'll find that most people go through similar routines as outlined above. It's important to know the limitations of your film and how the emulsion will render highlights and shadows: these are gained by experience alone. Then you can decide how you would like an element of your scene to appear.
If you don't want to go the EV route (or if your meter can't handle them), you needn't worry. Set the light-meter to one shutter speed and measure the scene for f-stop values in different areas. Although the absolute values are a bit of a pain to work with, within a scene you're actually thinking in terms of relative f-stops, ie +/-1 stop, +/-2 stops, etc. Then you can think as with EV's. You'll soon see what's bright and needs toning down with filters. And for 80% of scenes, your experience will tell you what needs to be done.
A couple level 2 considerations:
(i) If you apply a lot of bellows extension relative to focal length (more than a ratio of 1.1), you need to compensate for the loss of light. Divide bellows extension by focal length, square it: that's your exposure compensation.
(ii) If you intend to use large movements relative to the size of your lens' image circle, you will also need to compensate for light fall-off. This is noticible with transparency film and shouldn't be ignored. Every lens is a bit different, so study carefully whether you need to compensate by 1 or 2 stops and then you'll know for the next time. ND0.3 grad is a good starting point. But it is a bit hit and miss at first.
I hope that helps.
Best regards,
Charles

[patawauke]

I found Jack Dykinga's description of spot metering on an average value for LF colour landscapes to be very clear and helpful in his book 'Large Format Nature Photography'. Worth a read.

The above posts from our LF practitioners are really helpful too, to someone like me who grew too reliant on the very accurate in-camera meter of the Pentax 67II in medium format. Made me lazy about exposure techniques. The Dykinga book, but even more the Hicks/Schultz book 'Perfect Exposure' was my wake up call about trying to understand light and exposure, once I waded into LF.

A question for Tim please, to help me understand the statement ....

"I can also check the shadows in the foreground and if I find out that they are -4, I can place a one stop grad over the whole picture and increase the exposure by a stop.."

...what is the concept here, how does the filtration reduction of the overall scene brightness help compress the tonal range? Is it because the shadows compress less than the highlights? Help!

And a question for Charles over the compensation for excesssive movements vs lens IC - I suppose this is unlikely to affect most landscape shooting, but where's the best place to consult the rules on this? Suggestions welcome.

Thankyou

[timparkin]

A question for Tim please, to help me understand the statement ....

"I can also check the shadows in the foreground and if I find out that they are -4, I can place a one stop grad over the whole picture and increase the exposure by a stop.."

...what is the concept here, how does the filtration reduction of the overall scene brightness help compress the tonal range? Is it because the shadows compress less than the highlights? Help!

The idea is that if the very bottom of the picture is metering at -4, I can cover everything but this strip with a one strop graduated filter. This means the rest of the picture is now one stop darker. To compensate for this, I increase exposure by a stop. This brings the rest of the picture back to where it should be but makes the strip at the bottom one stop lighter.

And a question for Charles over the compensation for excesssive movements vs lens IC - I suppose this is unlikely to affect most landscape shooting, but where's the best place to consult the rules on this? Suggestions welcome.

I'll just answer from my own knowledge. If I were using a 110XL which has a very wide image circle and I was using the very edge of the circle I would have to compensate slightly. I know that the Centre Filter for the 110XL is 1.5 stops but I could work out from the cos^4 light fall off rule for typical lenses that at 30ish degrees away from the centre, you get about 1 stop of light fall off..

This means that at the edge of the frame, I would be getting about 1 stop less light across the frame and it would be graduated across the frame by about half a stop.

Lets say I had applied a lot of rise to my front standard. the bottom of my picture would now be using the centre of the lens and the top of my picture would be using nearly the edge of the lens. I would have the equivalent of a one stop grad across my picture, darkening the sky.

Interestingly, when talking to Joe Cornish about this, he said he occasionally uses this effect to cancel out the bellows factor he can get from having objects very close in the foreground (if an object in the foreground is getting on for half life size or more, you need about 1/2 to 2/3 of extra light to get the same exposure - this cancels out the light fall off effect discussed above, leaving you with just a 1 stop darker picture overall.

Bear in mind this are just general guidelines and in the field a good rule of thumb is to expect half a stop of reduced illumination if you are using the edge of a normal image circle lens and a stop if you are using the edge of a wide image circle lens. (super angulons are better with light fall off so 1/3 stop for normal - 1/2 stop for big image circle). As for bellows factor on near far compositins, don't forget that this will only affect a small strip at the bottom of the picture..

Hope that doesn't just really confuse you

Tim

[Charles Twist]

Tim has given a fairly complete answer. In my experience, which is still being added to, I can use up to 2 stops of ND grad compensation, but then on some architectural subjects, I can end up with 10cm of shift on a 210mm lens or 5cm of shift on a 90mm, which is quite consequent. A couple examples:
This first picture of York Minster was taken without any filters: I used the fall-off on a 90mm to darken the sky. The top of the building was also darkened as a result and tentatively rescued in PhotoShop, resulting in some discoloration (not sure why?). Film = Velvia 50.


This next picture of the chuch in Roker used both rise and shift on a 75mm lens darkening the top left corner (which was already in a dark recess), while the top right corner was bright because of the windows. With some choice filtering, this was recorded on Velvia 100. I recommend a piece of paper, when trying to decide on the filtering to achieve the right tonality in different areas of the picture.



For landscape, I sometimes use a lot of drop on the front standard with a wide angle lens to give that looming foreground effect. That can result in the foreground being 1/2 to 1 stop darker. But it's rarely more than that. And therefore almost negligible in neg film and even provia to an extent.
At some stage, I'll sit down to work out the angle for a given shift for a given focal length, and work out from that what fall-off to expect with a bit more science and a bit less arm-waving.
I hope that helps.
Best regards,
Charles

[Joanna Carter]

... and work out from that what fall-off to expect with a bit more science and a bit less arm-waving.

Science! Dear boy, don't you know photography is an arcane art?

[patawauke]

Thankyou Tim and Charles for the expanded clarifications, I will now need merely a couple of months of brain workout to digest it all!

Coming back to the original poster's questions re colour exposure hints at sunrise and sunsets, the former usually has subtle lighting with a not extended brightness range, so spot metering on a mid value (distant grassy hill, the right colour cloud) or indeed an ambient light meter reading if the sun and subject are in a suitable position from camera (typically with sun over shoulder) should give reliable results.

However, glorious sunsets are high contrast affairs, and I have usually followed established advice and spot metered on things like blue sky or mid-grey cloud some 30 deg angle away from sun, or on the grey/yellow mass that surrounds the glowing sun. A digital camera with spotmeter function is useful to have with you, as you can then evaluate the result of these readings and adjust your choice of metered subject or your exposure settings to get the look you want, (eg slight underexposure, darker and more contrasty).

John

[brijphoto]

dear all,

Thank you for the postings...I have just bought a spot meter, and am planning to go out tomorrow and try out some of the techniques suggested in the postings.
If the rain gives it a rest...

Cheers

Brij

[Charles Twist]

For everyone's delectation (except Joanna), I have quickly looked through the numbers to give us an idea of theory. Firstly, I enjoin you to read this post which explains the background - the linked articles are quite interesting. With a few computations, one ends up with a curve like this:


The X and Y axes are the distance from the middle of the image circle on the ground glass. The Z axis shows how many stops of light (or EV's) you would lose relative to the centre of the image. This curve is for a 210mm lens focussed at infinity and shows the theoretical values based on a cos^4 falloff. The conclusion of looking at the numbers is that if shooting with 2-3 cm of shift, the fall-off is negligible (less than half a stop). Once the centre of the image circle is in a corner of the image, the f-stop drops fairly evenly at a rate of about 1 stop across the width of the 5x4 transparency. Let's say the transparency is a 10cm square: with a 5 cm shift, use a 1-stop ND grad filter; with a 10 cm shift, use a 1-stop ND grad and give the overall exposure an extra 2/3 stop.
With a wide angle lens, it seems that the design requires a cos^3 rule. The numbers are quite scary, but the image circle prevents us from straying too far. With a 90mm focussed at infinity, with the centre of the image circle concentric with the picture, fall-off at the corners is around 1 stop. Apply 3cm of shift or rise, and the light drops off by just over one stop across the picture.
I am beginning to think it might be useful to enter these equations into a phone or some small electronic gadget. Anyone with the programming skills?
These are theoretical values and there is indeed no harm in the arcane art to achieve similar results.
Best regards,
Charles