I think this is your most reasonable argument for adding rubble to the game, but it'd require a bit of balancing. As it is now, it is fairly easy when starting out to just dig a 1-tile tunnel into a mountain, stick a wall there once you're all inside, and never have to worry about anything until a building destroyer shows up. I agree that rubble could help making things harder starting out, which is good, but if it made it impossible to create a large fortress within a year or two, it might not be worth it. Think of the bedrooms/halls for 40-80 dwarves, stockpile storage for food, workshops, workshop storage, barracks, training rooms, etc. Those are all basic things which still require space. Another thing is that it looks like as if most of your argument for rubble could just be replaced by making mining take longer.
You say:
The reason why we need rubble is, counter to your expectations, it should reduce the number of junk items being generated because it slows down the pace of excavation considerably, as each time you mine a set amount of stone, you have to deal with the excess material generated. By the time that you are mining again, that rubble should be dealt with and out of memory in some manner or another.
If rubble can't be used, how is it nothing more than junk? You would be creating the same amount of junk, but it would have the name of "granite rubble" instead of "granite stone", and possibly be even more worthless (assuming that you can't do anything with it).
I'm sure that rubble would add complexity to the game (sometimes a good thing by itself), but would it add a corresponding amount of depth? I would be in favor of it if you had specific ideas for what it would add. As the idea seems now, what new things does rubble add to the game besides slowing down mining? Can we do stuff with it? If it has !!SCIENTIFIC!! applications, like being ballast for trebuchets (as of yet unadded 
) or being able to be combined into cement for pillars for supports with new cave-in mechanics, I'm all for it. If it is just for adding more things for us to dump while slowing things down, why not just decrease mining speed via the skill, instead of adding something that has no use at all?
You use the example of beds and sleeping a few times, so I'll try to see if I can explain while I think beds add depth. If a dwarf sleeps on the ground, they get an unhappy thought. One of the whole management goals is to keep a happy fortress. While this is easy enough to do with making statues and awesome food, when/if it gets balanced, it can actually have an impact. Also, sleeping can be influential in terms of increasing variability and depth in the game. It allows vampires to feed, and your militia to not be ready for that siege that comes in the middle of the night, when you only have 10 dwarves awake instead of 20. This is an example of adding both complexity and depth. Rubble seems to only add complexity, with no depth (influence of other game mechanics) that couldn't be achieved just by decreasing mining speed.
Wyatt Cheng had a good blog post about the sort of thing I'm trying to convey here on his blog. Sorry for any misinterpretations of your posts or points I don't explain clearly.
I think you are coming close, but misunderstanding a little.
To start from the bottom up, that blog post has a decent point, and one that is very applicable to DF in general.
It is not applicable here, however.
Adding rubble to the game (and specifically only rubble) adds one aspect of complexity to the game - that there is rubble that needs to be cleared before mining can continue. That isn't a terribly complex issue for most people to wrap their heads around.
The only things rubble actually do to change how a player plays is to functionally add a small punishment (a delay) on trying to expand their fortress length-wise. It does less to punish width-wise expansion of the fortress. It gives players the opportunity to build more complex cart tracks if they choose, but wheelbarrows are sufficient, and I would expect most players to just have cart stops that expand only very infrequently, while wheelbarrowing their rubble to the nearest cart stop to be shipped to an automated dumping.
For a slight amount of complexity (simply learning that you need to dump a rubble item), you get a fair degree of depth to the game - you get the same "speedbump" that I was talking about to Slink before.
As I was saying then, in the 2d version of the game, which was much more game and less simulation, the underground river (and the magma and pits as well) provided a speedbump in progress - the enemies it spawned and periodic heightened difficulty, and crossing took time and effort, and they most importantly just stopped progress until you could confront them. At the same time, the game progressed in materials - copper was in the shallow pre-river areas, but offered little proteciton, while iron was deeper, and more useful, but took higher risks and further development to obtain.
That gave the game a more clear sense of progression without being terribly complex - you were progressing into a "harder stage" when you cleared each roadblock on your way into the mountain.
Rubble serves the gameplay purpose of just being a roadblock to slow your game down. It doesn't add snakemen or fire imps, but it does at least let other aspects of gameplay, like early ambushes and sieges and harsh environments and a possible scarcity of stone actually add more to the game than they currently do. If it slows down the rate at which the player can discover or explore every ore vein, map every cavern, and exploit every resource, it adds depth to the game by making the player figure out how to make do with what they have at the time rather than assuming they will always have unlimited resources.
And it happens to be realistic, as well, and that adds to the depth of the verisimilitude of the game.
As for sleeping, you get part of what I meant, but not the other part.
You see, the reason I was talking about sleeping was because sleeping doesn't actually
give the player anything, it's a problem for players to solve. It's a speedbump to building a fortress, as well, as it requires you build a system of bedrooms for your dwarves and it takes up valuable wood resources to make those beds. It is also an opportunity for players to be creative in their bedroom designs, as making fractal or other decorative bedroom patterns are one of the favorite ways to customize a fortress.
The miscommunication occurs where you aren't seeing the parallels between these two systems.
Rubble slows the digging down without adding significant complications to the game by simply making a miner have to wait for a hauler to take rubble away before mining can continue. Rubble does not take up valuable resources, but it slows the rate at which resources can be gained down, and that has a massive effect on the depth of the game, as exploring for and finding metal ore deposits becomes much more tricky and less likely to be assumed that you will have found huge hematite veins that are just waiting to be mined out as soon as you are done with the last giant hematite vein you depleted. That forces more conservancy of resources as you are unsure of when your next metal supply will come in.
It also offers you a choice - the more haulers you have working, the more quickly you will mine, as your haulers are the key to opening the bottleneck. The more labor you devote to hauling, however, the less labor you have free for other uses. Is mining to expand now more important than building mechanisms for traps or putting up a better wall to keep out invaders?
It also gives meaning to the new carts system we have, by giving you plenty of reason to want to try making a labor-saving highly sophisticated cart path for hauling your goods around without taking up dwarf labor in actually hauling your goods around. It allows the game to reward players who are clever and creative in their use of the game's tools by making their other strategic choices easier.
This is also a core element of some of my other suggestions, including the farming ones - where it is easy to set up a basic farm, and possibly just trade for further goods like wood or cloth or more food, but advanced farming will give you extra resources that can be used to make up for shortfalls in other areas of the game. Hence, it lets players specialize in the mechanics they enjoy, while others can simply go without the benefits of specialization in one area and try to make it up by specializing in another.
Hence, I compare rubble to sleep - they aren't terribly complex in and of themselves, but the changes they cause to the rest of the extant systems in the game give them a great deal of depth.
This is why I say that we don't need a "use" for rubble in just the same way we don't need a "use" for sleeping - it's a problem to be solved, not a resource or a benefit. The problem itself is the use it brings to the game. If there are uses for rubble, that's a side matter, but rubble does not need to have some sort of benefit to the player to give its existence meaning, in just the same way having dwarves take sleeping breaks certainly doesn't give the player any real benefits, and gives them obvious drawbacks.
Or, in other words, "Not being anything more than junk" is exactly a goal state of this idea.
Further, you don't "build walls" with rubble - the idea is that Toady's "amalgam" code would take care of that. The amalgam idea was to prevent quantum stockpiles by simply deleting items that hit a critical mass of density in a tile and cause the entire tile to become a wall of junk. When mined, you get nothing but junk back, and that means you don't have to store the items that went into the quantum dump in memory anymore, and let the sorely abused STL vectors contract again, which results in FPS recovery.
All a player has to do is let rubble be dumped in a large pit, and eventually, that pit will fill on its own without much player activity.
Dumping could occur off-map, dumping could just have an "anthill" effect of building a giant mound of rubble in an out-of-the-way part of the map, or dumping could occur in some more exploiting manner, such as just dumping everything into a volcano or the magma sea or atom-smashing everything so that it doesn't take up any volume anymore at all.
Ultimately, though, unless you face a problem of having a landfill that is getting landfilled to capacity, all you need is to designate where you're dumping the excess, and letting the haulers do their thing. It's entirely simple and automated.
Incidentally, and to the rest of the people reading (mostly Slink, though, since she was who I was talking about arguing with), the benefit of arguing these things is that the more you argue a point like this, the more refined your arguments become. Obviously, the things that appeal most to me don't appeal most to everyone, and it's a process of elimination to find the best arguments to make.
Besides, there are always flaws in your ideas to be corrected, so having someone trying to find them helps you correct them.
Author
Topic: On Rubble: Treading on Unstable Ground... (Read 53948 times)