Long exposures can be noisy because the sensor starts to heat up. To compensate, cameras usually have an option to take a second frame with the shutter closed to then substract the “base noise” (this is called dark frame substraction IIRC). You can also do that manually by taking a frame with the lens cap on.
1 Like
I’ve been there on that nice festival in May, what a wonderful place…
In playraw I participated in that milky-way shot with those two tents. Maybe you like some of my settings in the xmp-file 
1 Like
Thanks for the hint with the PlayRaw Shot! I’ll take a lot and see what I can learn. 
I love that place, it has a very calm feel to it. Let’s hope the weather will be better again for next year’s Kirwa. 
Last year I spent a night up there and captured a timelapse of the milky way moving across the sky. However, I have some frames where the WB is not equal to the rest of the pictures, apparently the AWB algorithm failed. This leads to some color flicker which I unfortunately don’t know how to fix. I tried Rawtherapee and ImageMagick, but I can’t iron it out. I’d like to try LRTimelapse, but I don’t have LR nor Windows
Good idea!
However I’d worry about losing a stop of light or so. And I don’t have ND grads
But I might try bracketing the foreground!
Also, @pitbuster made a good suggestion with sensor heat. I took a darkframe on my desk and subtracted that which yielded better results. In addition bracketing might be also beneficial due to lower exposure times.
But overall I’ll just shoot the foreground at blue hour, that should give a better image overall.
Thanks for your ideas! I’ll definitively check my camera settings. I also tried noise reduction with an ad-hoc darkframe on my desk and got better results. I will try to employ those techniques in the field next time!
I’ll send you a PM later about Timelapse
1 Like
Since you are using raw files, consider setting one WB setting for all of your output. Flickering might be due to your brightness and contrast levels as well. There are a few scripts that may help you with that.
Can’t think of any ATM, save for one from the Magic Lantern community. I am sure someone can provide some links or do a web search for us. 
The problem is, that I did a Sunset-Milkyway-Rise Timelapse which means that the color temperature changes over time. A fixed WB results in either a blue-ish or red-ish tint.
@AxelG
The scripts I know about are timelapse-deflicker and autowhite.
For exposure flickering darktable’s exposure module offers an “automatic” mode which also helps a lot. Rawtherapee offers an AWB mode, similar to autowhite, however the algorithm fails just like the my camera’s AWB algorithm 
I don’t see a problem with that. Sunsets are supposed to change the temperature of the sky. Auto balancing everything would definitely result in the colours jumping all over the place, especially if the algorithm is already struggling.
Consider selecting the correct temperature for all of the images. Pick a few frames that balance well and see how that setting affects the rest. Or, if the temperature transition is too dramatic (i.e., has too wide of a gamut), you might consider ramping the colour balancing, transitioning from one white balance to another. I am sure that is possible. I haven’t done time lapses myself. 
Yes, I thought if I use AWB it’ll be good enough for doing the ramping. For most pictures it performs reasonably well, but every 20th or 30th picture (don’t quote me on the exact numbers) is an outlier.
I was hoping for a software based ramping solution but I couldn’t find something for Linux. Implementing this might not be that hard but I’d need time to dive deeper into the topic. Time that I actually don’t have.
That sounds like a camera issue. Open question: would different settings or hardware help? Perhaps, consider skipping the outlier frames and interpolate them.
Sorry for the late reply!
I don’t think it’s a hardware issue, cause under regular conditions the AWB algo works fine. Also, rawtherapee’s AWB algo and darktable’s spot method both produce a color shift on the outlier frames.
However, there are some (low) clouds moving through the frames. Those clouds reflect city lights, which results in an orange glow on the clouds. With the clouds moving through the frame, occupying sometimes a smaller, sometimes a larger fraction of the frame are my guess for tripping the AWB algo off.
I tried to use libraw to read the temperature coefficients from my RAW files. That works fine and it should be easy to find the outlier frames. My idea would be to either ramp the WB for a subset of the frames: Calculate the median, not the mean (!) value for, let’s say 10 frames, apply it to those 10 frames and move on to the next 10 frames. Or, if one could reliably detect the outlier frames, just fix them.
However, what I don’t know is how to use that information to fix the White Balance. libraw does not allow modification of RAW files (which is a good thing), so I need to devise a mechanism to apply my corrections. I could, for example use ImageMagick to apply the corrected colors on lossless files (e.g. TIFF). Or, and that’s what I would prefer, modify darktable's XMP’s. The problem with that is, though, that the temperature coefficients in darktable's XMP files are stored as a binary dump of the internal data structure, so I need (de-)serialization code.
So, as far as I can see there’s no easy way right now. Given my time constraints for (private) software projects I’m likely not going to make fast progress here, but I’ll try to stick to it.
And for the primary topic: I had the chance to shot the foreground again at blue hour yesterday. I hope I can construct a better image now!
Re: WB. dt supports Lua scripting.
1 Like
@Tobias Too bad.
@jchnkl I suggest that you do minimal processing on the raw files (I prefer linear no-clip float output at this stage) and copy the tags over to the *.tif. TIFF tags can be modified, so you could write a script to smooth out the outliers on the WB ramp. Then open them in dt for further post-processing.
1 Like
Yeah, that’s really a shame. It could be so powerful for batch processing pictures, but right now it’s rather limited! 
That’s a really good suggestion! I will definitively look into that! 
Quick update: I finally wrote a python script which
- reads the WB coeffs from the RAW files
- smoothes out the outliers (still wip, not that easy unfortunately)
- packs the new coeffs as binary C struct & serializes them
- writes the coeffs back as temperature module to darktable XMPs
This works pretty well already, however, I need to find a good smoothing algorithm. I’ll keep you posted