I think you can be satisfied with the overall result given the equipment you are using. You are fighting the effects that are unbeatable without traps and compromises. If you want substantially better, you need a better team or you need to "cheat" or both.
There are several apparent effects at work here.
As Matt says: there are substantial JPG compression artifacts. The original imgur file was 750 kB, which suggests that it had been substantially compressed somewhere along the line.
Consider the version marked below.
You can also look the imgur version marked here Y
As marked version here – click on the second icon in the photo at the top right to "download the original file"

- The target is too large for the available pixels.!!! – It's a magnificent view but
The 550D in the complete song produces a maximum image of 5184 x 3456 = 18 mp and the copy of the image was 3888 x 2592 = 10mp, so it represents approximately 45% of the resolution of the original area or approximately 75% linearly . Even given another 33%, it just doesn't have enough pixels for details. Here is an example where the swollen D800 sensor almost makes sense.
The camera must be set to produce everything to the fullest. RAW if RAW is used. The best possible jpg if not. EXIF or at least the central data for this particular image would be useful.
See the note G in the image: the lamp has been flown to show the details. It is an original of 30 x 30 pixels as it occurred to me. There are not enough details there to expect the light to be much sharper than it is.
-
You may be getting visible diffraction effects. Good prominence in street lights is often a sign of small opening and possibly diffraction. You say f16, which probably seems marginal for such effects.
-
The sensor is saturated with luminance (highlights "fade") and also in some or in the 3 color channels in many areas.
You cannot expect a "genuine approach" to the original material if the reflections are completely saturated and the details are destroyed.
This is an inevitable result if you want to render an image with such a wide dynamic range and achieve an eye-pleasing result without resorting to HDR, tone mapping or magic in general. Notes A. B. C, E show histograms of small areas.
The light in A has a pleasantly blown point of 100% white in the place of the light. The brain accepts that without much complaint.
The light in B also has its maximum level fully saturated (the red channel is shown here) but has a very large output component throughout the range. (What the histogram shows depends on the amount of surrounding darkness included, but it is clear that "tone mapping" or HDR or local camera contrast management (DLighting, etc.) would improve this if applied with the Enough violence
I included C to show that a line of lights with moderately lit surrounding areas has a lot of total saturation, although this is not necessarily obvious at a glance.
Where is Wally / D? ?
E is in the reflections of the river: the brightest points are saturated.
I looked at F to see if some sort of filtering could help the lights. Blue is well below but red and green are well saturated in the center.
In general it is a good image.
I could tolerate a somewhat reduced total exposure and this will help some of the highlights, but only a little.
This is an example for what HDR is done (one opinion, feel free to disagree :-)) – used with good taste, possibly with manual control, it will probably add a lot, but the effort can be severe and many will not notice .
Selective contrast or brightness control would also probably help, depending on how the camera was set up, more aggressive treatment may be available.
Playing with the opening will confirm or reject the idea that you are getting some noticeable diffraction effects in the spot lights.
In the days of the movie, they could be invited to dodge and burn badly.
How good is your tripod?
Have you tried the tripod / camera / timer settings on known perfect point sources with perfect focus and time delays of the same magnitude? In recent tests in extreme configurations and using a focus magnifying glass, I was a little surprised at how unsound and seemingly solid the tripod mount was and how long it took to settle after being touched or struck for the last time. That was extreme (focal length equivalent to 1275 mm: the orbital movement of the Moon in the plane means that it takes about 20 seconds for its edge to travel through the frame!) But still applicable to your situation and the focus magnifying glass Give an idea of the vibration that the standard viewfinder will never give you.
I have not mentioned the quality of the lens, as it is somewhat unknowable and there is enough above to explain what you are seeing (or not seeing) that you really need to address before an improvement of the lens is helpful. I think! The lens experts disagree at all.
I have not mentioned the atmosphere, which is also a variable party and is not easy to know. At night, when it is cool and quiet (or cooler and quieter) it is usually a good time in that regard.
Duplicate image to minimize displacement. Delete if you wish.
