I made this image in early September of this year. I was using my new Pentax 645Z, and had a 35 mm lens and a 75 but nothing in between. I made a vertical composition and then wanted to include more to each side, without adding extra top and bottom, so it just made sense to do a stitch, even though at 51 megapixels, I didn’t really need the extra resolution. Switching to the 35 would have included a lot more.
It’s actually faster to do a stitch than it is to change lenses.
Normally the Kananaskis river is turbulent, fast flowing and deep, but at this time of year and at 8 AM, the dam was not letting much water through – an inch or two flowing over the rocks above the fall, and the pools quiet and reflecting sunlit clouds while the pools themselves were still shaded.
Literally 10 seconds after the last of three exposures (left, centre and right), the water started rising and within 30 seconds the scene here was gone, hidden by waves and deep water.
Stitching was done by using my ballhead to tilt the camera downwards (I was standing on a large rock), and then rotating the base of the ball head to make the three images. As this often makes the left and right edges tilt at quite an angle, you need to be sure that you have enough covered to be able to crop the angle, or else tilt the base of the ballhead forward as I did here so that, relative to the point of rotation of the ballhead, the head isn’t aimed downwards that much.
I’m able to change the angle of the bottom of the ballhead because I use a leveling platform under the ballhead. In my case this was an accessory for my RRS tripod, but in the past I have happily used a Manfrotto leveling base that could fit on any tripod. I know it seems strange to have both tilting under the ball, and also as part of the ball, but for horizontal row stitching, this is essential any time the camera is aimed up or down relative to the horizon. Much fancier panorama brackets can be purchased, but with modern cameras and lots of pixels I just don’t need to shoot rows and columns for stitching and haven’t since the time of my Canon 10D, a 6 megapixel camera. It’s just that now I can make 6 foot prints when I stitch. This image is some 13,000 pixels across, 8500 pixels high, without any uprezzing and after having been cropped.
Remember that when subject matter is both near and far from the camera, one must rotate the camera around what is commonly (but incorrectly) called the nodal point of the lens, a position part way between the front and back of the lens, and in a position that changes as you zoom, and not always the way you’d expect. In this case the near-far difference wasn’t much, and careful aligning of the lens was not essential. Have a branch come near the camera though, and you’d better be spot on with your rotation point. This requires some sort of rail. I’ve made them of oak with a routed groove to match my arca-type clamps, but they are readily available commercially, ready to be clamped anywhere along their length from below while the camera clamps on top at one end.
To find out where to set the rail for each lens, I just stuck a couple of garden canes into the soil, one near, one far, and kept swinging the camera and lens back and forth while adjusting the rail, until swinging the camera did not change the relative position of the canes, even as they moved from the left side of the viewfinder or lcd to the right.
In practice, stitching is a lot easier to do than to describe, and often I don’t even use a rail, just relying on all of the subject being at almost the same distance away from the camera, or taking advantage of my zoom lens mounting brackets to get the tripod rotating under the lens instead of the camera.
The images were brought into Lightroom for cataloguing and Camera Raw processing, then exported as TIFFs. I then brought the three images into PTGui for stitching. I have been happier with this software for stitching than Photoshop through several software iterations.
The stitched image was then brought into Photoshop for editing. Some wonder why I would use the far more expensive and complex Photoshop to do my editing when Lightroom is very powerful at a fraction of the price (and I started there anyway). The answer is three fold. One, I’m old – I’m used to Photoshop and frankly can’t be bothered to become good at Lightroom. Second, although Lightroom has hundreds of undo’s, after the first couple of dozen you can’t remember what you did when. I find the layers, masks and painting of masks in Photoshop to be the best for producing fine art images. The third advantage of Photoshop is that there are plug-ins for Photoshop not available for Lightroom and I use two on a regular basis.
In addition, I like the idea of being able to fade layers through adjusting their opacity, and to select luminosity instead of normal blend mode so colours don’t oversaturate as contrast is increased. I also like the ability to use the threshold layer to point out near white areas so I can both get sparkling highlights, but also not drive them to pure white. For help with editing in Photoshop, I direct you to a series of videos I produced on Youtube under georgebarrphoto. You can find the first video in the series here.
Oh, and the name of the picture – it comes from the river feature. A huge rock that juts out into the river on the near side, and is then matched by another rock, not quite aligned, coming from the opposite bank. This creates a sharp S-bend in the river and a narrow gap. Kayakers love to practice their skills on this feature (when the dam is flowing anyway). It’s located several hundred meters further into Kananaskis park from the information building on the right as one drives into the park after leaving Highway 1 for Highway 40.