Sooner or later we come to a point where we question our methods and wonder if there is a better way. As with many things, the usual answer is yes. However it is never an easy crossover to get to the other side. It takes will take a lot of work before seeing useful results. I am referring to 3D rendering in this instance.
My journey into Blender 3D started about a year ago. For me Blender represented an amazing tool, and a solution looking for a problem. Over the course of the year, the skill build up took up a crazy amount of accumulated hours. With a good amount of determination it was like solving a puzzle, and it started to make more sense as the various bits fell into place. Recently I opted to set up a beverate can model. This seems simple enough but it took a bit of a run up before I could produce a useful model. This is the point where product photography and digital illustration blur the line.
Here are some initial results -- two rendered can models, and an earlier example of typical product photography.
In theory the huge win from 3D is that you can get creative with composition as if you were at the shooting table, but with more precision. In practice however, there is also an unknown. It is the random quantity of required effort that is expanded on trial and error, with no guarantees of getting anything usable. Certainty and predicatabity comes over time.
From the general to the very specific, it is worth mentioning a hint that Blender material parameters include Index of Refraction (IOR). This is useful for modeling anything transparent, or in this case condensation droplets. Water should have a value of 1.333. If you miss this detail then you may just end up with hollow air bubbles.