For President’s Day weekend, 2019, I took a quick trip to Banff National Park in Alberta, Canada. It was cold with a capital C. On Monday, the car’s thermometer registered -16° F.
However, this photograph happened on the previous evening as the sun was setting.
I’d been out shooting all day starting at Rundle Forebay in Canmore. But the weather was a bit overcast and while I still shot, nothing was coming out very good. I then went to a place called the Grotto. It’s a shallow canyon with a frozen river running through it. There were some good things happening there, but it was still nothing really striking my fancy.
I then met up with a podcast listener at Johnston Canyon which is popular for tourists. I didn’t shoot much there due to the crowd and the narrow pathway that doesn’t provide much room for setting up a tripod. Plus, the weather had cleared out some and the sun was very harsh on the snowy landscape. A beautiful spot, one I really want to return to, but it wasn’t all that photogenic that day.
Then we decided to head to an overview of Mt. Rundle. This overview is highlighted here on the included map. It’s on the road up to the Norquay ski area.
We pulled off and had plenty of space to park and enjoy the scenery. I set up my Canon 5d4 with the 70-300L lens. I saw the small cloud being blown from the top of the mountain and I knew I had my shot. I wanted a long exposure which would cause the cloud to smooth out and turn quite wispy in nature. I also knew that I’d want to make it a B&W image due to the lack of color already in the scene and to enhance the mood, texture and forms of the mountain. Keeping the color would have distracted from all that.
But I had a problem. My 10-stop ND filter would not fit this lens which has 67mm filter threads and I only have an 82mm filter with reduction rings to make it fit on my 77mm lenses. So, I was out of luck. Or was I?
In my recently published episode of Latitude Photography Podcast, I talked about a technique called “exposure averaging.” With this technique you set up the shot, take several frames in rapid succession and then bring them together in Photoshop. So that’s just what I did and let’s talk about those details in the next section.
Setting up the Shot
The lens was set to 124mm. I chose an ISO of 100 to keep the noise down, but also, more importantly for this shot, to get a slower shutter speed. I wanted a relatively slow speed to help emphasize the cloud movement. But even with 100 ISO I knew I’d need to do more.
The aperture was set to 16 and the shutter was an acceptable 1/25 of a second. I was also using a standard circular polarizer which takes out about 2 stops of light. I used it to enhance contrast in the sky just a bit. It helped bring out the clouds a little better.
I got this shot a bit looser, and a bit tighter. And I’ll still process those to see how they come out. I really like this framing though because the mountain is so strong and it’s filled with contrasts.
Contrasts in light and dark tones to be sure, but there’s also contrasts in textures, shapes and expectations. Let me tackle that last one first. Contrast of expectations.
First off, I ask, how do you read this image? My guess is that you start at the upper left corner and go to the right along the top. You follow the cloud. But some of you start at the lower right corner. That’s the contrast of expectations I’m talking about. In our western society we read things from upper left to lower right. It’s how we’ve done things for centuries. It’s almost too cliché to mention really. And for many of us that’s where we start. It’s an easy place to start since that’s what we’re pre-programmed to do, but also it happens because the brightest subject happens to be there. The eye is drawn towards brighter subjects and those that have details as well. The cloud hardly has any detail, but it is bright and it does have a gesture. It directs you to the right. For the few of you that start in the lower right you’re drawn by the contrast in tonality there and the detail it provides. But you don’t linger. And for those if you that started in the upper left, your eye is cast to the right and there’s nothing of interest to look at there. It’s a great piece of negative space that allows the eye to rest but you also don’t really wander. You’re instantly drawn to the lower right and then again, you don’t linger. You’re instantly rocketed back up to the beginning and there you might enjoy some of the detail before you start the journey all over again. The eye travel is so well controlled in this framing and that’s why I like it.
With those camera settings, and mostly notably the shutter speed, I knew I’d betting a tiny amount of motion blur, but I really wanted more. A lot more. However, this is where things get a little tricky in understanding exactly what’s going on.
I set the camera up for 30 successive shots using the built in intervalometer. But the first frame was slightly out of alignment due to a tiny amount of camera movement caused by my hitting the button. So I got 29 usable frames. So if you did the math that would mean that 1/25sec X 29 frames would equal 1.16 seconds. However, the wind wasn’t blowing all that hard. I also had programmed a 2 second delay between each shot. So in effect, I got (1/25+2)sec X 29. That equals 59.16 seconds. So because of that built in delay between frames I actually extended out the “true” capture time to almost a full minute which created the amount of motion blur I was after.
With clouds and other soft subjects, this technique will work. With other hard edge subjects, likely not so much. At least not as easily since there will be hard edges that may be noticeable. This screen shot shows the individual frames in LR. The highlighted frame, 306, is the combined frame which I’ll talk about now.
Combining the Images Photoshop.
Once the processing is set in LR, and the settings are synchronized to each image, I opened them in Photoshop as layers. This brings each image into photoshop on its own layer. See screen capture above for the LR command.
Then I took those layers and turned them into a smart object, select the layers and click “Convert to Smart Object.” This creates an embedded .psb file that Photoshop treats as a single layer. If I were to double-click on that smart object layer it would open up the smart object as a separate file and it would be all the different pics in their own layers just like when I brought them over from LR initially.
With that smart object layer, I then went to menu item Layer, Smart Objects, Stack Mode, Mean and allowed the computer to crunch away at that for a bit. By doing that, you’re telling Photoshop to average out the pixels on each layer. Where there’s movement in pixels it simply averages them out and makes the cloud a smooth cloud.
Does this image look exactly the same as if I were able to attach my 10-stop and get a 1-minute exposure? Probably not. I’m sure there’s differences between the idea of getting it in one shot and digitally compiling the shot. But it’s close enough for me.
Plus there’s the added benefit of the ISO averaging. I was at 100 ISO. If you divide 100 by 29 you get 3.448. That’s an incredibly LOW ISO. I’m convinced that the camera has a noise pattern and so this low ISO isn’t perfect, but it’s still going to be better than if I shot it at 100 ISO at one minute.
I love the shot and folks I’ve shown it to also have really enjoyed it. My boss has even invited me to make a canvas out of it. So I’ll do that and add the process of making the canvas as a bonus to my print course I’m currently developing.
If you’d like to experiment with this technique and anything else photographic, please consider joining me on one of my workshops this summer. I’d love to help you elevate your photographic practice and expertise.