Grad filters with slide film - what strength (situation)

Occasionally for enjoyment I shoot some landscapes with Fuji Velvia (sunsets).

 

Situation:

I spot meter the sky and the foreground (with adjustment), I found there was a 4 stop difference. I only have a max 3 stop grad filter. Because slide film is said to have 2 stops from midtone to the highlights and 2 stops from midtone to the shadow. Am I safe to assume that the dynamic range of the film would be OK in this situation?

 

 

Thanks.

You should be able to achieve a 6 stop range of capture. However Velvia has a steep contrast curve. This means a small change in exposure results in a dramatic change in density on the film.

 

For sunsets, I expose for the sky abut 45 deg from the sun, and let the foreground go to silhouette.

 

You don't want to render the sky as dark as the foreground. I think a 1 or 2 stop ND grad would suffice. Expose for the sky +2, then open up further by the strength of the filter. The sky should remain bright, but the foreground will open up 1-2 stops. Even so, the foreground will have a purple cast. The color temperature of an overcast sky is very high, over 7000K, and Velvia is not very forgiving of deviations from clear sunlight.

 

Green seems to come through well on cloudy days (green = - magenta). Look for subjects with grass and foliage, and not much sky.

Edited July 19, 2019 by Ed_Ingold

You don't want to render the sky as dark as the foreground. I think a 1 or 2 stop ND grad would suffice.

Two stops I think. I'm not sure a puny one-stop would make a lot of difference.

 

All that spot metering and filter-faffing sounds like a lot of trouble. I shot this RAW and combined two layers with different exposure settings to get the result I liked.

You don't need a spot meter. Just point the camera so that the sun isn't in the field of view. If you use a split ND, orient it to cover the sky and open up from that setting by the filter factor.

 

Photography is 80% craft and 20% art. Matrix metering could probably do okay, but someone had to "teach" the meter. Fundamentals are important if you want to know why certain things work and others don't. Before digital, before auto focus and before auto exposure, there was Hasselblad, a Sekonic L-508 meter, and a Florida sunset.

 

The Florida sunset was taken with Velvia film.

Raymond, with film I used to carry a 2, 3 and 4 stop set of resin SinghRay and Hitech soft and hard grads using a holder and adapter and a tripod. Now that I shoot with a digital camera, even those with low dynamic range I just travel with a 3 stop hard and a 3 stop soft and do any correction in post with a holder . It’s been many years but when I shot film (Provia for wild life and Velvia for landscapes) Art Wolfe, I think, told me to spot meter the gray not too close but adjacent the sun for my mid tone for sunsets. If I were shooting film again I would definitely be carrying a more complete set for multiple circumstances. Also know that even the SinghRay filters gave me a fringe color cast at times. My understanding is that the NiSi filters that are optical glass and more expensive are better over all. About 15 years ago I was at Red Reck Canyon outside of Las Vegas at the rest top on the highway. There was a young man standing under the awning next to the bathroom no doubt working on his college photography project hand holding his rebel camera in one hand and a square Cokin grad in the other and a cigarette hanging out of his mouth, while dangling the filter in front of the lens to take the shot. A budding Ansel Adams no doubt. Smart to use the awning for shade no doubt. Some how the scene it’s etched in my mind. I truly hope I never get to that place. My vote is for three stops but then with film I might have a two stop to try as well.. Good hunting.

For landscapes with digital, it is far simpler and more effective to shoot a 3-5 frame bracketed exposure, then blend the results in post. There is no line of demarkation as with a split grad, hard or soft, and you are not limited to a linear horizon. There's no reason this technique would not work with film, other than the high incremental cost.

 

Another technique for film is unsharp masking, whereby a negative dopy is used to mask highlights and reveal shadows. It is contingent on an original image with enough dynamic range of capture to retain detail in all parts of the image. In the simplest form of "unsharp masking," you can dodge and burn while printing.

 

The orange base of color negative film is, in fact, a variable mask. The color comes from the unused leuco base dye, and is lighter where the image is darker.

Edited July 23, 2019 by Ed_Ingold

Since you ask about slide film specifically, I'll just note that the "flat square filters" of the Cokin sort (link) are more easily moved to put the darker area where you want it for composing in the camera.

 

Here's a short little piece about ND filters on film:

 

 

Popular Photography 1985-03

WRT sticking grads in front of a digital camera, there may be consequences not yet mentioned.

 

For the sunset I posted above, I had to removed the multi-coated UV filter I normally keep on the lens, since there was a flare spot on an earlier attempt. I realised that it was a ghost image caused by reflection off the sensor returning through the lens and then reflecting from the flat surfaces of the filter again. This is quite a well-documented phenomenon.

 

Anyway, removal of the (very clean) UV filter resulted in a ghost-free picture. However, had I been using an uncoated grad filter, I'm pretty certain that it would have caused a much stronger ghost reflection in the system.

 

FWIW, by shooting RAW at base, or near base ISO, I've almost never had to used a grad or multiple exposure to capture all that was needed for detail in the deepest shadows. While the highlights are prevented from blowing by a judicious choice of exposure. 12 stops of dynamic range go a very long way when they're helped out by residual lens and camera-body flare.

 

And did nobody at 'Pop Photo' notice that the highlights are still blown and bereft of colour and detail in their second example above?

Edited July 24, 2019 by rodeo_joe|1

And did nobody at 'Pop Photo' notice that the highlights are still blown and bereft of colour and detail in their second example above?

 

The article points out that the "photographer wanted" the detail in the mountains.

 

I'd shoot it in digital myself, but the OP wanted FILM

JDWvW - thanks for the article. It makes sense.

 

The article exposes the sky as it is and then makes the foreground 1 stop darker. Originally a 3 stop grad became a 2 stop grad.

 

Any other views on how you guys tackle this? Making the sky 1 stop brighter than a metered foreground wouldn't be as nice right ...

Raymond, I understand your point but I think it depends on the situation with film. That is why I would consider having more than one density. With digital you can use light room or NIK digital filters in PS to brighten or bring down the sky as long as you’re within range and can capture all the information with the dynamic range that is available or just combine multiple different exposures in Photoshop. With film as you pointed out your more restricted so you need to bring more options when you’re shooting. There is nothing wrong with that and the Hitech grads are a bargain at $30 for the 84mm version. I have never used a round screw in grad on land. It’s always been for split images between water and the skyline. The rectangular filters with a holder are better than square because they give you more room to place your transition zone. I would also say that with landscape, the way I visual the shot is not necessarily how it appears in real life. So my goals with graduated filters may well be different. Good hunting. Edited July 25, 2019 by 2Oceans

A light meter of a certain area will give a value which will render that area as though it had neutral reflectance - 13-18 % (depending on the meter calibration). This is the "brigthness" of green grass, clear north sky, barn red, and (of course) a grey card. If you want that area to be darker or lighter than "neutral," you adjust te exposure down or up respectively from the recommended value.

 

If you wish to balance bright and dark areas, you can measure both areas and use a split ND filter to reduce the brighter area by the difference. There are several scenarios, depending on your photographic intent.

• To lighten the foreground, expose for the sky, then increase that exposure by the value of the ND filter (typical sunset photo)
  
• Expose for the foreground, and use an ND filter to darken the sky (typical bright cloudy sky strategy)
  
• Measure the darker and brighter areas. Adjust the camera up or down for the darker area, if you want it darker or lighter than "neutral." Select an ND filter which will then reduce the brighter area according to your rendering intent.
  

Even with digital, the span of brightness, in which you want to retain detail, seldom exceeds 4 stops. This excludes the sun itself, which along with specular highlights, are not expected to show detail in most cases.

 

In the following example, exposure of the sky and fountain, shooting directly into the sun, was +2 stops from average. The dark foreground, in shadow and backlighted, was - 2 stops from average. It is a digital composite of 0, +2 and -2 exposures, but it illustrates how you could use ND filters to balance a photo, with a scene with about as much range of exposures as you would typically encounter.

 

To get about the same results with an ND filter, you would expose the foreground at the reading - 1 stop, and use a 3 stop grad on the sky.

 

Attempting to implement a closer emulation, I used the "0" exposure frame, increased the overall exposure +1 stop, and applied a 3x split ND filter in Lightroom. Colors and highlight details are better in the bracketed version above, as they would be using film and a physical ND grad filter, but the exposure levels are pretty close to real life.

 

The article points out that the "photographer wanted" the detail in the mountains.

Yes, but it doesn't say he wanted to blow out the sky.

 

The exposure details given just don't add up. The foreground exposure was supposedly 4 stops greater than the sky, and a 16x - 4 stop - split grad was used. 1/60th to 1/4 second is indeed 4 stops. Therefore, with the nd grad in place, the sky should have received the exact same exposure as in the first example, which it obviously didn't.

 

That's unless the scanner guys at Pop Photo simply tweaked up the slide exposure for the second 'example', and no grad filter, or alternative shot, was involved at all.

Yes, but it doesn't say he wanted to blow out the sky.

 

why do you assume they/he cared about the sky? You are so suspicious:)

You are so suspicious.

The world is so deceiving!

An average exposure will yield an underexposed foreground and overexposed sky. The following images were taken with a +\- 2 stop bracket. I show the image (+2) exposed for the foreground, and the composite image. The final image closely emulates use of a 4x ND grad with a +2 exposure. As I read it, that is precisely what PopPhoto did in that tutorial.

 

Do you care about the sky? If not, try to keep it out of the image. Otherwise it is part of the image, and you should make the best of it. That's where an ND grad (or bracketed exposure) comes into play.

1. In daylight, and a clear blue sky, use the average exposure. Nothing special is required, except for a polarizing filter. If the latter, use it with discretion, or the sky can turn dark purple in a high, dry environment.
   
2. On a bright cloudy day you will get a very dark foreground, or a completely washed out sky. Reducing exposure by about 2 stops will bring out some detail in even the heaviest overcast. Using and ND grad (or bracketed exposure), you can get detail in both.
   
3. Treat a colorful sunset like version #1 and let the foregound go, or like version #2, whichever gives you the desired effect.
   

Real clouds are never white. The light on an overcast day is 7000K or higher (7600K is not unusual). Film like Velvia will assume a strong blue or purple tone. Summer (cumulous) clouds reflect the ground below, but with a bluish cast due to atmospheric scattering.

 

Anyone care to guess why my ND grads are in storage, not touched in 15 years?