Wyrm. GreatWyrmGold. A worm is a little crawling thing. A wyrm is a dragon.
Err... sorry. I kind of remember the names phonetically, and it's been a while since I typed them. (And yes, I've kind of been gone for half a year... during which time, I've actually played a wide range of sandbox games - DF may well have been leaving its mark.)
As to the rest of what you said: I really have no clue how RL metallurgy works, so why don't we look up the rough ratios for that?
Well, everyone's favorite lazy research tool to the rescue:
Wikipedia says hematite and magnetite are "direct shipping ore" with over 60% iron content, which is relatively rare nowadays what with the easy stuff already stripped and depleted.
The largest amount of relatively pure stuff comes from deposits in the
Banded Iron Formations, which was a result of Earth's ancient oceans having huge amounts of iron dissolved into the seawater that all oxidized and turned into rust at around the same geologic instant when
stromatolites came into play. They oxidized the water, causing all the iron in the oceans to start rusting, and they fell into the ocean floor to form what are some of the oldest sedimentary layers on Earth in the areas near those shallow ancient oceans. This is why those formations are relatively thick with iron - nature conspired to concentrate all the iron into that geological formation.
Banded iron formations, notably, are generally only a few centimeters thick. It's like mining out a giant pancake - you have to dig very broad and very shallow. (Note, a z-level is roughly 3 meters in this game, which you are excavating for only the .1 meters at most that are actually the hematite, of which only about 2/3rds is actually iron.)
The other major way to get concentrations of iron ore (magnetite) is through the formation of
felsic magma, which usually means "you find it in granite" (the igneous intrusive felsic magma stone). The distinction in magma types comes from the fact that "mafic" magma contains high concentrations of magnesium and iron ("mafic" being a concatonation of "magnesium" and "ferric", the Latin word for iron), but because the iron and magnesium cools and solidifies faster, magma trapped in chambers will eventually have the chambers have a shell of magnetite around them while the remaining magma turns "felsic" as the chemical composition of the magma changes more towards oxides of silicon, aluminum, potassium, and sodium.
Hence, these two sources would be very concentrated.
Beyond that, all other sources are either going to be made of, basically, large puddles of rust and a whole lot of other impurities, and will require much more effort to get anything remotely pure out of the metal. This is where things like
bog iron come in, where you need to process tons of peat to get even low-quality iron.
This, by and large, means
limonite, but it can also mean hematite if it's relatively low in impurities. Limonite is an inferior ore compared to hematite, and takes much more processing to remove impurities. If you wanted to get truly realistic, you'd make the process for making pig iron from limonite different from the process for making pig iron from hematite.
Without blast furnaces, the only way to work limonite was to heat it to extreme temperatures (I.E. magma forges alone may actually not be hot enough) and repeatedly beat the metal until most of the impurities were cast off as sparks. With blast furnaces, you have to understand the chemical composition of the ore you're working with, and cater amounts of fluxes to the amounts of impurities.
Note that NOT all impurities are easily removed.
See the smelting section. Lime (the flux) is actually there to help remove silicon impurities, but the level of those impurities (and hence, how much flux you need) will vary from source ore to source ore. There are different techniques for the removal of each element that might be an impurity.
To give an example from the article, sulfur is a major problem, as it causes iron to become brittle when hot (and shatter when you try to work it):
The importance attached to low sulfur iron is demonstrated by the consistently higher prices paid for the iron of Sweden, Russia, and Spain from the 16th to 18th centuries. Today sulfur is no longer a problem. The modern remedy is the addition of manganese. But, the operator must know how much sulfur is in the iron because at least five times as much manganese must be added to neutralize it. Some historic irons display manganese levels, but most are well below the level needed to neutralize sulfur (Rostoker & Bronson 1990, p. 21).
If you wanted a relatively-simple-but-realistic version, you could have differing iron ore qualities or purities (or just say that magnetite is best, hematite is second-best, and limonite is worst, but preserve our current types of metal ores without differentiating too thoroughly) and make the formula for turning them into iron or pig iron different.
Preferably, going for realism, the game should have different types of "iron" for each type of source ore. Current "iron" is just pure iron, as if someone wrought out every single impurity, which is fairly unlikely. Having different qualities of "iron" based upon a source ore would be more historically accurate. (This is why certain countries were famous for having better iron than others - Damascus steel was so wonderous because its ore just happened to already be nearly perfect steel, and sweedish steel was more valuable because it had such low sulfur contents that made it easier to work.)
Having every different region have iron in some form, but having the
quality of the iron be based upon the source ore is the most historically accurate way to portray the iron industry, and it also has the side-bonus of helping foster that whole trade thing.
As for the higher-quality dwarven steel process, you could then just cater the smelting process it takes to make the pig iron to have differing amounts of flux, coal, and ore in to pig iron out ratios based upon the quality of the ore.
... Or, you could just say screw realism, and keep the relatively simpler system we have now.
Simplicity alone, of course, has its virtues, but then, why not "1 + 1 + 1 = 1" of the materials, the way that the game started out, and just ignore the "keeping the ratio of tiles mined to bars produced" idea? There was no game balance purpose to the original formula, so saying that the game balance would be ruined if we had to mine more iron to achieve the same effect. (Which, if we have hematite deposits on our map, don't kid yourself, you'll have enough to be paving the streets in iron if the whim strikes you because you're not going to run out if you have it at all. There is no "balance" there to ruin.)
The primary reason I see most people arguing for the 1 boulder = 4 bars formula is that it is, basically, what they got used to, and they just don't like the change.
Likewise, when there was the version change in 2010 that temporarily took away the necessity for multiple steel bars to make a given piece of armor, when it was eventually put back in around 2012, all the people who had learned the game on 1 bar = 1 full torso's worth of armor complained about the change.
For what I'd personally like to see, it's making the different ores actually different, and having human nations invested in trading for better iron deposits. Dwarves, as well, might need to have different types of flux. (Manganese isn't in the game now, but pyrolusite, it's ore, is, and could be used as an anti-sulfur flux, for example.)
Other than that, I'd like to see proper stacking coding put into place, so that we don't have to worry about such low-integer values, and we get more dwarves capable of carrying 10 charcoal bars at a time so that the integers can be higher and more exact (and we don't have to mine 48 cubic meters of stone to produce the boulder that makes a single .000001 cubic meter large earring).
Failing THAT, however, I would simply argue that lower integers are better for as long as we're spending more time shipping the materials from the stockpile to the forge than actually working the material. If we want to talk game balance in general, it would take a much larger and longer-range look at the state of the game, and what in it has to be changed to create a real challenge before we can talk about game balance.