I think it's Evans, Wallace and Sutherland's book on computer simulation that has an example of a tank combat model that purports to show the superiority of the event-scheduling approach over a "continuous" one; pardon the lack of an accurate reference, I'm sure I can find it if you like, but the book's been out of print for a few decades. One can also argue that "continuous" models are really discrete-event models with regular event scheduling.

I'm not sure that "artificiality" is quite the right word to describe the problems of the event-scheduling approach, so much as "awkwardness", at least for a manual simulation. However, there are certainly some solid arguments in favor of a "hybrid" approach (both for computer and manual sims). People familiar with only one of the approaches (and I confess I'm a discrete-event bigot, and am thinking of having a T-shirt printed saying so) tend to see the system they are modeling in terms of the approach they are used to -- it can be really refreshing to see how differently someone schooled in the other approach perceives the problem.

It would be good to think that wargames designers sat down and thought hard about which of the processes they are modeling are continuous, and which involve sudden state-changes. My perception of modern, tactical, terrestrial warfare is that most of the processes are of the "sudden state-change" kind -- the transition from "OK" to suppressed" or "panicked", or a change in mission, or the instant at which a hidden unit becomes detected, are all reasonably well modeled as occurring instantaneously. If you like, I see modern tactical combat as being "jerky". The same may not be true of other periods, nor of other scales -- loss of combat effectiveness, for example, is pretty much an on/off instantaneous thing for a single tank, but more like a steady process if viewed at, say, the division level. One can argue that movement is always best modeled continuously, as people and vehicles are not usually observed to jump suddenly from one location to another.

Thinking out loud, I'd be interested to see an event-stepped game whose event-scheduling mechanism was closely tied to the "decision points" in the command and control system of the game. It seems to me that the distinction between command and control could be nicely modeled if some way could be found to limit the number of events at which players could intervene. Good command would then be shown by players making good choices when their decision points permit them to make choices; good control and communications would be shown by having a greater frequency of decision points than your opponent. This would be especially relevant in showing the differences between large, cumbersome forces and agile ones -- which on the wargames table usually means a way of letting 1941 Germans beat 1941 Russians without resort to artificial "troop quality ratings".

Note that I tend to shy from pure modeling terms in this forum - mostly inappropriate (IMHO) for most of the discussions. In that I may well be incorrect. Given that "professional" modelers don't seem to have universal agreement on the meaning of the terms they use any more than wargamers do, I don;t think it matters either way! ;-)

I see very many game systems discussed and have great interest in the different reasons for the design methods and the background for the choices. Right now I'm trying to get a game together to check out my own and one other game system (written by DGaven). We hope to play the same scenario and see what happens under each system. If I can find the time to get to the games, it should be a rather interesting project. May also use one pure non-computer game (probably Empire III) as a base to compare from.

Ahhh, if only research grants were available for the discipline of "comparative wargaming"!

A long time ago, I did something roughly similar, playing out a standard British Army 1970s era section attack using the boardgames "Patrol!" and "Raid!", and the WRG Infantry Action 1925-75 rules, and, IIRC, Charles Grant's "Battle" rules. The main lesson was that each author obviously considered very different things to be important. The subsidiary lesson was that, if the results I obtained were credible, the British Army's minor tactics were not half as crazy as I thought they were when first introduced to them. It's always good to see reality validated by a simulation study.