That was utterly fantastic! Thank you! They've recently realized that the powder for their fireworks is rather useful for blowing things to smithereens (following history there lol), but I hadn't yet gotten to the nitty gritty of comparing the components of black powder to the records I've found of old Chinese firework components.
Well then, here's some more historical info:
As guns became more common on the battlefield, tactics changed to reflect their deployment. Firearms were needed to defeat the enemy, but soldiers armed with them were vulnerable to shock action. To stand and fight, they required protection. The best defense against enemy cavalry was a close formation of pike-armed infantry. Early tactics combined firearms- and pike-armed troops. The pikes provided shelter from enemy cavalry while the guns killed the opposition from afar. This is commonly referred to as pike and shot. Later, the bayonet was developed. This allowed every soldier to serve as his own pikeman. Instead of a portion of the formation being dedicated to shooting and a portion to melee, every man could do both. This dramatically increased the formation’s lethality at each task, and eliminated the need for specialized melee and fire troops.
The barrels of early handcannon (gonnes) were often made of bronze, but occasionally of wrought iron. The Chinese developed cast iron comparatively early and sometimes used it in gonnes. When Europe developed more sophisticated personal weapons, some early models had wrought-iron barrels, as did Japanese firearms inspired by them. The later matchlock weapons, and wheellock and flintlock firearms, had cast-iron barrels, which gave better strength for the same weight. Most barrels were smoothbore. These were favored for military use because they took less time to reload and could thus achieve a greater volume of fire. However, balls from smoothbores come out spinning on a random axis, which often deflects them from the target. Rifled barrels, which have spiral grooves that give the ball a controlled spin about an axis parallel to the barrel, date to the 1500s. Rifled guns were used in hunting, where accuracy was more important than reloading speed. Hunting guns often had two or more barrels, to give the hunter a second shot if the first one missed. This was especially important when hunting dangerous game such as bear or boar.
Matchlock firearms were standard military weapons through most of the 17th century, long after wheellocks and flintlocks were invented. So why didn't armies immediately adopt better weapons? Simple military conservatism played apart, but the main reason was that the new weapons were more expensive. Kings and generals preferred a large army with matchlocks to a small one with flintlocks. Still, armies did invest in more advanced weapons when they really needed them. Longarms were awkward for mounted men–cavalry men needed pistols. But matchlock pistols were also inconvenient on horseback. Thus, the cavalry got wheellocks by the end of the 16th century. Artillery forces worked with large amounts of powder. Burning slow matches risked setting it off! Soldiers guarding artillery were therefore issued the fusil – a flintlock long arm – in the late 17th century, long before ordinary infantry men got it. Rifles were more accurate than muskets, but took longer to reload. In mass combat, firing 50% more shots in the same time improved combat effectiveness more than did superior aim. Hunters, however, often worked alone, pursuing wary game; making the first shot count, and taking it from farther away, improved their chances of a kill. So hunters adopted rifles long before soldiers did–and early military sharpshooters were often called “hunters”(
chasseurs in French or
Jäger in German).
Matchlock: Developed by 1411 and standard after 1500, this lock placed the burning match in a mechanical holder attached to the gun. Pulling a trigger moved the match into the touchhole. Matchlocks had limitations – they were vulnerable to rain, the smoking match made them hard to conceal, and the glow was visible at night – but they were sufficiently easy to use to become the standard infantry weapons of their era.
Wheellock: This design came into use around 1500. Such locks were complicated and expensive. A clockwork spring in the lock turned a serrated steel wheel against a piece of iron pyrites, striking sparks that set off the powder. Pulling the trigger released the wheel. Then the gunner had to rewind the spring with a small lever, the
spanner; reload the gun; and lower the cock, which held the pyrites, onto the firing pan. A gunman who loses his spanner can improvise one.
Flintlock: Developed in the 1500s and perfected by 1620, a flintlock has a cock that holds a piece of flint, which is pulled back against spring tension. Pulling the trigger releases the spring, which drives the flint against a steel
frizzen, discharging the gun. There were similar designs known as
snaphaunce,
Miquelet lock, and
doglock.
Loading multi-part ammo – powder, shot, and wadding (A.K.A. patch) – is a rather involved process. Briefly, you must pour in the propellant, usually from the muzzle; then insert and ram down the projectile, followed by the wadding; and finally fit the igniter to the action. The number of seconds required varies by weapon type, but some general rules apply (all times approximate):
A self-measuring powder flask subtracts five seconds from the final loose-ammunition loading time.
A self-priming pan subtracts 10 seconds, but this benefit isn't cumulative with that of a self-measuring flask.
Pre-measured paper cartridges halve the basic time required (round up), and supersede both of the above measures.
Muzzleloading musket or shotgun with flintlock action. Takes 30-40 seconds.
Muzzleloading rifle with flintlock action. Takes 50-60 seconds. Reduce by 10-15 seconds if using a greased patch.
Muzzleloading smoothbore revolver with flintlock action. Takes 15-20 seconds per chamber.
Muzzleloading rifled revolver with flintlock action. Takes 25-30 seconds per chamber.
Black powder produces a solid residue. In a long fight, carbon and lead buildup can have adverse effects on accuracy, loading speed, and reliability. After every five shots, the firer must take a two-minute break to clean his gun
thoroughly. Any kind of watery liquid, including urine, will do. Saliva works well – shooters commonly hold a patch in their mouth until it’s saturated. Battlefield swabbing can’t remove all the fouling. Black powder is corrosive and attracts moisture. Cleaning and oiling the gun is a necessary after-battle chore. Disassembling the lock, cleaning it, reassembling it, and swabbing the barrel takes about an hour.
Gun calibers
aren’t standardized, and balls can’t be mass-produced. Gun owners need equipment for casting their own balls: an iron or bronze crucible for melting lead, tongs for handling it, and a set of bullet molds. The latter resemble a large pair of needle-nose pliers with several round hollows of the proper size; they can be closed, clamped shut, and then opened when the balls have cooled. Even a single-cavity mold lets one make bullets quickly – a bullet per minute is a leisurely pace. It’s an easy job, too; a black-powder bullet is either good or it isn’t, and anybody can tell the difference. A campfire produces enough heat to melt lead; casting half a dozen balls takes 4 minutes. Armies carry lead ingots with them.
I'm a Civil War junkie, so I'm more familiar with the later strategies and how they organized then, but that's three hundred years past where my book is at. Good to know the basic unit is essentially the same. I've always wondered that they didn't work on ear protection, since the pain and ringing had to last for days.
That...that...is an abomination. I don't have any magic in my fantasy world. (I'm sure they'll revoke my Fantasy license any second now.) So everything has to follow a logical and plausible progression. No mana boosts for my engineers. They're at the technological cusp where firearms would be the next natural innovation, but pushing them over it and into more modern warfare is a tremendous amount of work. 
