The photographer is looking for generalized feedback about the aesthetic and technical qualities of their image.
Description
From a night photography workshop led by Paul Dileanis, with special access to the State Park at night. Skies were very dark and clear. I would have preferred to be a few feet to the left for a slightly different perspective on the structure but there were 8 of us and I was delighted with whatever I could get. The structure is quite large – maybe 30 ft high. Low level lights.
Specific Feedback
All comments welcome!
Technical Details
Single exposure, darker than I would like thanks to an f/2.4 lens (Rokinon 14mm f/2.4 SP) and an exposure time of 8 sec to minimize star trails. Minimal adjustments with Curves in LR. Into PS for star removal with RC-Astro StarXTerminator. On starless layer: Topaz Denoise, Severe Noise with modified parameters, with a lot of attention to low-frequency noise in the midtones, then Nik Tonal Contrast. On star layer, stars made smaller with Annie’s Astro Actions MiniStars 3. Stars blurred slightly to hide the hideous wide angle artifacts in the corners (and this is one of the “better” lenses for stars) and the blue CA minimized. Burning on FG with masked Curves. Crop from the left.
Critique Template
Use of the template is optional, but it can help spark ideas.
You may have seen this comment before, but this is a type of photo that really draws Robbo in like a moth to a candle. I also appreciate your ability to move around or change the lighting was somewhat restricted on the night.
I do wonder how the photo would have looked with the old diesel winch illuminated instead of the headgear - or perhaps both (which might have looked too busy).
Is it possible to see a version of this photo without the Topas denoise? I would really like to compare the before and after.
Thanks, @Rob_Sykes – I have the same sort of reaction to things like this! I had no control over my position or the lighting. I would have loved to be a little to my left to get past that large sagebrush, and and lower and closer, but there were 8 tripods with legs overlapping. I wouldn’t have cared for the winch to be more prominent due to the degree of wide-angle distortion at the edge of the frame, but that’s just a pet peeve of mine.
After I did denoise I did two layers of Topaz Tonal Contrast and Pro Contrast, which were pixel layers incorporating the denoise but maybe it answers your question about noise if I present 100% screenshots of the galactic center area with the denoise layers hidden?
I’m not delighted with the softening and may revisit the denoise step.
I need an f/1 lens that is optically corrected for stars in the corners if I’m going to do much of this.
Your two sample photos were great - exactly what I wanted. Topaz has achieved a great result with few stars dimmed out. I thought about alternate methods to ascertain if I could use my existing software to produce similar results. I used your unfiltered sample , opening it within ACR and going crazy with noise filtering which then wiped out most of the stars. I then took a second copy of the original unfiltered sample and inserted it as an under layer, blending the two by means of lighten. I adjusted the amount of blending by changing the luminance of the filtered sample. Hey presto, stars back.
I used Photoshop Elements to complete the above.
Thanks for giving me plenty to think about when processing starscape environments.
I think composition and denoising has been addressed but I am somewhat baffled at the lowered shutter over the higher ISO. To reduce the noise, I probably would have opted for ISO 1600 and a shutter of 15 seconds and, even with the crowded condition, given what I can see in your shot, would have shot a three to four shot vertical composition to offset the extreme foreground distortion. Since you did not, my critique is then based on three criteria:
the distortion of the foreground elements
the foreground color balance
the Milky Way
Wide angle lenses can be a PITA which is why PS invented split warping. I did a general warp correction for perspective, then applied two horizontal split warps to correct the foreground verticality. The tool will make you crazy at first, but once you get used to it, it’s hard to not use.
Fixing the warp (which by using the splits I did not affect the sky “abnormally,” ) drew more attention to the foreground and thus the brightness and fairly intense “orangish” coloration which I suspect was caused by a tungsten light source. To offset this, I selected the foreground and added a bit of color balance moving to the blue side and away from the red and later added a touch of blur to the foreground grasses that were sharper than the MW, so rather disconcerting.
Then, there is the sky, which against the foreground elements as presented, is rather dull and lifeless. It’s there and the audience can see where you have adequately denoised, but it doesn’t jump out and sing its own praise. For that to happen, I did two things, one in PS and one using the TK9 panel for frequency separation, though I think TOPAZ also offers a similar feature - and can be accomplished in PS but requires more work.
In PS, I selected the sky and added a color balance adjustment layer, +21 on the blue, -3 on the red. On the same selection, I did the frequency separation and adjusted to a -80 and then dodged the texture sub-layer to pull down the sharpness on the foreground grasses.
Thanks for the ideas, @Chris_Calohan. This is definitely a WIP.
The SS is an easy answer – at even 8 sec with 14mm, stars are showing movement. And the best lenses I have found for stars show more corner and edge star shape distortion than I want, so I try to minimize the pesky shapes. (I’m spoiled by the gorgeous round soft stars in tracked images shot with a corrected flat-field refractor and processed in PixInsight.)
A pano in vertical orientation with the camera a little more level would have allowed for better perspective correction, but wouldn’t have given me many more pixels on the subject with this lens. That is a strategy I use with the 28mm but it was set up elsewhere doing star trails with my backup body. I have been pleased enough with the distortion correction I can do, and I already did quite a bit. I liked the lean of the minehead but a little more correction could be good, and I definitely should do more on the winch.
I would have liked more exposure but the compromises left me needing to dig out dark detail. I played quite a bit with both the lighting and sky colors, but I will probably wind up at a different point when I revisit the image. I like your bluer sky. The lighting was LED low light panels with gels and they were too warm for my taste, but I wasn’t doing the setups. I have done a lot of burning and softening of the sagebrush, but more is probably good.
If I ever get through the bulk of processing from the workshop (hundreds of daytime images in addition) I’ll post a revision.
