I once got totally consumed with the idea of an aircraft barrel (axial) engine. It has a lot going for it, starting with its cylindrical shape - you cannot get aa piston engine with less frontal area for the power. My idea would be this: 4 cylinders, opposed pistons, port valves, compression ignition. Picture the cylinders like the chambers of a revolver. Pistons go in from either end. Each connects to a wobble plate mounted on a straight shaft. This converts the reciprocating motion into rotary motion. It also eliminates about 93% of piston side loads. Front end drives the prop. Rear drives a vane-type blower (it's essential to make it run, it does not add power. It performs the scavenging.) Behind that the accessories. Injection could be as simple or sophisticated as you like, mechanical pumps and injectors or common rail injection. Most components can be made on a simple lathe. Cases can use sheet metal and the whole thing furnace brazed; no castings. Liquid cooled, since the shape does not lend itself to air-cooling. The advantages of the two stroke, opposed piston engine can be seen everywhere, including some modern iterations. The old problem of burnt out exhaust pistons of the Jumo can be solved with carbon pistons. In its simplest form, it needs no electrics. You could even go so far as fit a spring starter and don't even need a starter motor. For more power and efficiency, a turbo could be added - it's' almost a no brainer for a diesel, but it's optional. I once managed to get hold of an interesting diesel engine analysis/simulation program from a Russian university that was amazingly sophisticated, but very user friendly. I sat up many,may nights tweaking things like port angles and injection profiles and whatnot. It was addictive. It also proved to me that this wasn't some hare brained idea with a fatal flaw. I literally read every patent there is on this type of engine (most had at last one fatal flaw), and dug up whatever else there was on the subject (not much). Flight magazine archives had a series of articles on these engines in the 1940s and was as quite hopeful they would materialize. I think their chief merits are only really apparent in the opposed piston type. If you want to get deeper into these, read about Charles Redrup, a British engineer who was the guru of barrel engines. I almost fell off my chair when I read his biography and saw a drawing of his proposed 2000hp aircraft engine - it shared so many details with my own scribblings, it proves that given a certain problem, many people working on it independently will arrive at the same solution. Another advantage of this design is how scale-able it is. It should work from 100hp to some insane size. Anyway, here's just crude picture of something that illustrates barrel engines.