I took this shot on a recent trip to southern Portugal. I love the original, and was very happy with the light and the end result, but there wasn't a lot of interest in the sky. I'm not big on sky replacements in general and usually work with single frames, but after I had to do some exposure blending on another shot, I figured I'd go back and try a time blend on this one--something that was originally suggested to me by another member of the fstoppers community. As I wasn't anticipating doing a time blend when I was on location, the night shot was actually taken from a slightly different perspective with more of the sky in the frame to capture the stars. That added an additional challenge to get things aligned properly to mask in the sky. I also used a third shorter exposure to mask in the waves and some texture in the water (and the bird if you can spot it!). I've posted the base layer and the final product here.
This is my first attempt at a time blend, so I'm very interested in what other people think and what suggestions you might have to improve. Is it too over the top? Is the transition in the sky too harsh? What works, and what doesn't?
I think that your actual blending job is well done as far as I can see, Josh. However, seeing such numbers of sparkling stars so close to the sun, while it is obviously above the horizon, looks very unnatural to me. Along with the convenient meteorite streak, it draws attention to the artifice, rather than the scene. The wave on the right dilutes the effect of the golden beams grazing the water. All in all, then, I prefer the first image
Thanks for the input Chris, really appreciate it!
Good to hear that the blending looks ok to you. Most of the edges were done by hand since the exposures I used don't align perfectly. I'm a Photoshop novice, and I couldn't figure out a good way to use luminosity masks with a gradient overlayed. My biggest area of concern on the masking was the top edge of the rock to the right of the meteor.
The placement of the meteor in the sky exposure seemed too fortutious not to include, but take your point that on a composite like this it might seem a bit farfetched and distracting. It's not exact since the images don't align perfectly, but it was as close to its original placement as I could get it.
I can definitely see what you mean about the stars closer to the sun. Do you think it would work better for you if the transition started further to the right of the frame away from the sun, or do you think the time blend just doesn't work in this case generally? I know it might be hard to say without seeing it done. I attached the exposure I used for the sky here for reference.
I initially only blended in the wave and spray around the rock in the middle of the frame, but the exposure on the water is generally longer than I would have liked and it lost almost all detail--I chose it as my base exposure because the light streaks were the best I had and that took precendence. I thought the wave on the right (and the one middle-left) might add a bit of detail and interest back into the water, but you may be right that it creates competing lines and points of interest. I'll fiddle with it a bit to see if reducing the opacity alleviates some of that, or else get rid of it altogether.
At the end of the day, I may agree with you that I prefer the original. Just figured I'd give time blending a try.
Thanks again for your comments!
I'm surprised that the stars are so visible even in this exposure, with the glow from the sun still so strong, Josh. I'd just drop the sky exposure in the first image and make it less cyan. I've introduced artifacts working from the posted jpeg, but I hope you get the idea. I do like that little wave splash at extreme left in your composite!
Thanks Chris! Color is one of the things I struggle the most with in my photography, so that's a helpful suggestion. I initially had the exposure on the sky lower, but was worried that the brightest part of the sky was starting to look a bit artificially dull without recovering any detail--it's not technically blown out according to the data, but with the sun in that vicinity there may just not be much there to bring back. Your edit doesn't seem to have that issue though, so maybe I can use a combination of exposure and blue HSL shifts to ease the cyan look.
It's funny, the night shot is pretty true to the original RAW file, but I actually had a similar thought to you when I first started editing it--but in reverse. I was surprised at how much the pre-dawn glow came through in the file considering how dark it was in real life--I suppose it's bound to show up more than to the naked eye shooting wide open with a 20 second exposure, but it is fairly pronounced in this shot. It was at least 1 1/2 to 2 hours before sunrise, and there wasn't much light to speak of.
Thanks again, your input has been super helpful putting some finishing touches on the first shot!
I find these discussions very interesting - always curious about the creative process, which I find fundamentally baffling!
I've noticed a cyan sky on many Fstoppers posts, for some reason, Josh. I thought it might be camera-brand-related, but I note you're a Nikon user like me.
I like the first one better. Not a fan of time blend in general though
Thanks for the comment Vincent! Time blending isn't my usual style either--just figured I'd give it a go here since my main gripe about the original was that there wasn't any clouds or interest in the sky. Might stick with the original for my portfolio either way.
I like the original image :)
Thanks Vincent!
The time blend is a unique style and works in some cases. I think you could certainly use this for a creative photography portfolio to showcase your talent in compositing photos. It is a little unnatural looking however and could be adjusted in some areas. But if just choosing one photo for a portfolio I would choose the first one, which is a very nice shot and great composition and lighting. The sky is a bit boring and you can tone down the cyan a little as mentioned, or of course manually add in some clouds in photoshop. I like to do some creative projects myself that are more for personal use, so overall if thats what you were intending it works well.
Thanks Eric, glad you like the original! I appreciate you guys pointing out the cyan issue too. Definitely helpful to have a few more sets of eyes on it.
The time blend was mostly to satisfy a personal curiosity--both from a technical standpoint since I'd never done it before, and because (as you said) the sky in the original is a little boring.