A couple factors can plug into how to write it. Era, setting, and the strategy of both sides.
Obviously one doesn't always put all their men in the center when in combat. Units are spread out into operating areas, whether these be flanks or key strategic positions according to command. The range of these may also depend on the era the RP is set in.
There may also be time for engagement to conclude after starting. Again this determined by where it's fought and era.
These basic things to approach in mind clearly every battle should not be noted as "X happened at Y and Z won over [whatever]". There is finesse to it and not charging in to slug out. Properly, one may have their perspective characters where the course of such situations is explored; whether from the top down or from the perspective of some guy actually in the field. So even in a short battle you could go into some physical and emotional detail on the process of killing; but a top-down perspective from a general or his commanding staff would also provide a more technical perspective on what's going on when it's important.
Now rounding to the previously stated points: how long this should carry out for comes down to environmental factors which should be considered before approaching battle. Battles may actually be short and brief, depending on how far back you go (per time). Battles during classical periods did not often reach great lengths of times and could conclude themselves in an hour or several; a day if it were to be really stretched out. But then if they were laying siege to a fortified position or a city then it could stretch out woefully to a week to the greater part of a year, or even well beyond a year with one side trying to starve out the other. So even if approached statistically a siege even should not conclude itself as soon as it begins where-in a battle in a less fortified setting could probably be calculated and ended then and there.
You might also have in these early eras a smaller space being fought over. Armies moved slower back then so the space fought over was often more intimate and there was hardly any real sweeping progress compared to the contemporary period. So the march between point A and B takes longer, the battles are on a comparatively small field, but a non-siege engagement can be started at brunch and finished by lunch.
As we move ahead then we come to a point where perhaps field-fortifications might come into play. And you can have forces actually turning non-defensible positions (that should have been destroyed in a day prior because of lack of concentrated fortification) and a concise understanding of putting up rapid fortifications can imbalance the time it takes to end an engagement (without straight-up giving up and breaking off from engagement). Sandbagging, trenching, etc is enough to slow things down. So I'mma take a moment to refer to
Rorke's Drift as an example of even a short battle for the late-modern era. And then post the final assault:
Even with guns the English were greatly (30-40x) outnumbered something that could have been mopped up even with spears and wooden-hide shields was turned into an about day-long fight because the fortified the mission, and at short-ish notice. But lacking mechanization at the time the battle is still confined locally.
But moving ahead we can see a marked shift in scale that further deteriorates the "Army X fight Army Y, after a long hard fight, Army X wins" method.
By the advent of mechanized warfare in the First World War with armored and mechanized infantry coming into the field by the Second World War and after there's a point in war where not only can an army move formerly slow heavy infantry at a light-infantry or even a cavalry pace, but the scale of war increases. I might even - arguably - suggest by this point we start to define the actions of war by less battles and more operations with events like the Normandy Landings being a part of a much larger battle across multiple locations because the pace at which both sides can move means a larger stretch of territory can be involved. Which means that if you define things as "X engages Y at Z" you're loosing out on a whole lot that both sides could have utilized or should have wrote. You're not fighting much in a single field or over a single mountain pass.
And I'd ramble about location but I feel I should close this now:
In closing, if you're going to do a battle then at its shortest I'd say at least two posts. One to establish the army and discuss its positioning and anything leading up to the first combat phase. You might even include minor skirmishing where the command would be looking for weaknesses in the enemy so they can abuse later. If against another player this would also give him a chance to explore his own army and for him to find his position, both should publicly say why their generals are doing what they're commanding to be done for posterity or so someone can call out on the bullshit (like-wise there should be some word of advisory dissent too so that there's nothing lost in potential character smugness).
After that the battle can begin proper and the second post is the battle itself. This'd fit in a more classical setting where the flanks of the sword or musket-bearing armies advance to engage and actually fight and things are decided. Although if against someone else this could go on for several posts with action-response moves; unless this was all established ahead of time.
If it's all set up ahead of time then you and your partner both know and agree that the battle should end in a certain way, at which point most of this post becomes less important and is pretty much the automobile to move the story along at a follow-able and plausible pace.
tl;dr - there are details in battle. Get yourself some perspective characters and use them.
Member
AliceAbigail
Scrawl Banditta
Wilted Rose
mickilennial
Ekreture
Supermaxx
erikh