I am not arguing at all about tile size. My suggestion only holds if tiles ever get a fixed size, which I hope very much happens some day. If they never do, then they never do. In that case, DF won't be the game I would like for it to eventually be, and farming can remain as it is. You're right: it ain't broken. It's just unrealistic and that's ok to some people, and to me it's pretty much a meaningless, contrived game mechanic that I begrudgingly tolerate. Fortunately, there are many other things about DF that I love.
I think you misunderstand. I agree that farming needs to be overhauled. I do not, however, think that that needs fixed-size tiles and I was simply pointing out that fixed-size tiles will cause more trouble than they are worth. For a slight amount of added realism, Toady would need to handle so many more things. Also, maps would likely need to be larger as not as much would fit in the gridsizes we have now. Larger maps means more rendering time, which means more calculations for pathing as well... This means a huge FPS reduction. It'd be nice to say, "I made a 200 meter tall statue of Urist McDumbass, my mayor that went berserk and got tossed into my HFS Express Tunnel" but it doesn't really -improve- the game in any conceivable way, except for satisfying people who want everything to conform to very rigid and exacting units.
Me being against fixed-size tiles has nothing to do with my feelings on farming. It is a nice thing to think about, but when you consider it from the standpoint that we already have FPS, CPU, and memory problems that we are just now getting over then you realize that more tiles to show the same amount of stuff is just not necessary. It's sort of along the lines of giving DF polygonal graphics. It would be cool to see, but the amount of power and processing required is simply unrealistic. The tile system in DF is modeled off of Nethack and Rogue and it isn't broken. In fact, "tile" is pretty much an accepted unit around here and around Nethack and Rogue communities. "One tile" means something very concrete, even if it isn't exactly clear what. It's like bad handwriting. It says something, we just aren't sure what.
IF YOU DO NOT READ ANYTHING ELSE, PLEASE READ THE BELOW SECTION!What I am saying is that we can remove the unnecessary complexity and simply tweak how farming works, which is supposed to be what this is about. If a Dwarf eats 10 Urists of food each eating cycle and your farm produces 50 Urists of food in roughly one Dwarven eating cycle, then that farm can feed 5 Dwarves. Let's figure that each tile produces roughly two Urists per cycle. That's a 5x5 farm to feed 5 Dwarves per cycle. One dwarf could manage a 5x5 farm. That's a 20% producers-to-consumers ratio. Voila, now farms are successfully balanced to where you WILL need to upsize your farms with your population. If a 5x5 farm feeds dwarves with 1 of those dwarves producing, then you have a 1/5 ratio per 25 t^2 (that's tiles squared. Yes, I use them as a unit.) That's 2/10 for a 5x10 farm (read: 2 producers / 10 total). That's 4/20 for a 10x10 farm. You would need 5 full 10x10 farms to feed 100 dwarves. 10 full 10x10 farms to feed 200 dwarves. That's a MUCH more manageable number, while still maintaining the "realism" ratio of food production. Of course, this also means that Quarry Bushes would require nerfing. Honestly, that's fine with me as I only need one season of them in a 6x5 farm to feed my dwarves for almost a year. That's absurd. We can keep the FPS problem managed, though, by having each "tile" of farm able to produce a set maximum of food overall in Urists (or a Dwarven measure of weight) that is settable in the inits. Or we could be more invasive and make it so each "tile" of a planted <plant> is able to produce a set maximum of food overall. Food should also have some waste that cooks can reduce with skill. Quarry bushes have waste in their useless plant matter and roots, like other plants, while plump helmets should have little waste if any. The maximum is how much a legendary <relevant skill> can create from the plant or foodstuff in question. Everything else is lowered by efficiency percentages, which could be set via raws or inits. Legendary is 100% efficiency, obviously.
If you skimmed here, please go up to the bolded message and read what I posted in its entirety. I feel as if it is an excellent compromise and has been compiled from several ideas I have read around the threads. It is not, by any means, an entirely original idea yet I feel as if it has overwhelming merit in addressing this problem using the new features introduced in 0.31.##.