Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

The new channeling is covered in awesome sauce, the old channeling smelled real bad.

113 (19.3%)

The old channeling was the best, we don't need two ways to make ramps it is just silly.

245 (41.8%)

Old channeling was the best, new channeling is also the best.  Can't we all just get along?

132 (22.5%)

You people need to get on with your lives, it's not a big deal either way.

96 (16.4%)

Total Members Voted: 583

[CppThis]

I'm not overly familiar with the old system but from what I've read I see the utility of both.  Pits are good (and trust me, pit defense works just fine in .31 only it takes more micro) but auto ramps are definitely more newbie-friendly.  How about a compromise, new system stays in but with two new features:

-Construction menu lets you build a pit, which doesnt 'construct' anything but eats one stone and works like an old pit.  Good for doing small areas where constructing one-off stairs isn't really an option, while adding a cost for doing so.
-When deconstructing a constructed ramp or stairs, the dwarf straddles both Z-levels and can path his way out of either.  It's possible to disable a pit stairs by removing just the upper part, but the remaining lower part clutters the pit unless you're willing to sacrifice a dwarf (in which case the pit is now cluttered with a dwarf corpse).

[Shades]

I'm not overly familiar with the old system but from what I've read I see the utility of both.  Pits are good (and trust me, pit defense works just fine in .31 only it takes more micro) but auto ramps are definitely more newbie-friendly.  How about a compromise, new system stays in but with two new features:

-Construction menu lets you build a pit, which doesnt 'construct' anything but eats one stone and works like an old pit.  Good for doing small areas where constructing one-off stairs isn't really an option, while adding a cost for doing so.
-When deconstructing a constructed ramp or stairs, the dwarf straddles both Z-levels and can path his way out of either.  It's possible to disable a pit stairs by removing just the upper part, but the remaining lower part clutters the pit unless you're willing to sacrifice a dwarf (in which case the pit is now cluttered with a dwarf corpse).

Add in the ability to queue multiple orders on a tile (so I can mark a tile for both dig and pit) and you have me sold. Dislike having to order some stuff then wait till they do it before ordering the next stuff.

[Sean Mirrsen]

I did, rather sarcastically, suggest limiting the "dig pit" option to 1 tile per designation... Generally, just limiting the designation size to some obscure value (like the current 8x8 maximum construction) would serve the purpose of making channel defences more difficult rather... nicely. It wouldn't hurt most of the liquid breaching or local engineering efforts, but any sort of large-scale digging would require some work from the user.
What's more, we can already build stairs down - extrapolating that and "constructed pits" to more options and structures seems like a logical move. Making empty space in the ground is one thing, but then you could also automagically embed floodgates into the ground - an even safer way of handling liquid breaches. I'm not sure if there are more uses for this, but there probably are.

Also, correct me if I'm wrong, but didn't channels in .23.xx.xx.xxx and subsequent pre-3d versions have to be constructed patch-by-patch instead of designating like now? Because that might be a rather ironic twist for my points on "not returning to an old system". Except of course, in this case it's more of an evolution of the old system.

[Shades]

What's more, we can already build stairs down - extrapolating that and "constructed pits" to more options and structures seems like a logical move. Making empty space in the ground is one thing, but then you could also automagically embed floodgates into the ground - an even safer way of handling liquid breaches. I'm not sure if there are more uses for this, but there probably are.

I still think it should be treated as two tasks by the game so, from your example, they dig the tile and then the mark for the floodgate placement is activated and added to the job queue. But from the players perspective they've just said 'Dig here, with floodgates here, go'. Which means less effort and, more importantly, you can issue a whole design as your thinking about it rather than keep having to come back to it.

Also, correct me if I'm wrong, but didn't channels in .23.xx.xx.xxx and subsequent pre-3d versions have to be constructed patch-by-patch instead of designating like now? Because that might be a rather ironic twist for my points on "not returning to an old system". Except of course, in this case it's more of an evolution of the old system.

Patch by patch is silly, as well as being terrible design, everything should be through the designate as much as you like system. Or something simpler if you can come up with something.

[zagibu]

I like it. If enemies can't cross it, the dwarf can't dig it from above or climb out of it. I'd actually even prefer the old system with the dwarf sitting down in the channel afterwards.

Then we still couldn't cut skylights, without dropping dwarves through the holes.

Yes, you can, but you have to do it before you hollow out the room below it. Or you build some stairs inside to catch the dwarf falling through the roof.

The real problem is that layers are 1 "Z" thick. If they were split in the 8 parts that water is already using, you could have your skylight, if the ceiling was only 1/8 "Z" thick. Such a shallow pit could surely be dug from above.

[Sean Mirrsen]

The problem with having "embedding floodgates" as a double-designation is that you'd still have to have a "construct floodgate" order that would work from above. Would it not be simpler to just say "build me a floodgate in this spot", and then the game would automatically assign such labors as are needed for the task - first mining to "clear the ground", then installing the floodgate, all without allowing any adjacent flows through the tile. It can be extended to normal constructions as well, for example you could order a bridge built, and miners would arrive to clear away any obstructions on either end, simulating the use of any tools or temporary constructions they'd otherwise use to get the task done.

Or not, as that would probably pave way for various exploits as well. It all depends on implementation, again.

[Shades]

I was just using the example you had to illustrate how I meant the designation queueing to work. I would expect in that particular case to get a "can't path" error or something similar. Obviously some magic would have to be checking potential routes an the like.

On a side note but in a related manner it would be nice if when you ordered a construction it remembers the path to the material you selected so that other constructions won't block this path unless a new one can be calculated. This would make building walls and the like less annoying in certain situations.

Anyway one reason I didn't suggest the all as one build concept for the floodgate is simply easy of implementation, the designation queue would be a minimal change to the code based (and could already be done with a similar tool to the ore vein digger). Another reason is that I don't like it because it changes the flow of the game but that one is a personal objection so feel free to ignore it.

[Another]

Re: 1 tile per designation:

I already use Autohotkey program to designate digging of diagonal corridors and to speed up interaction with the trade screen interface. I know that some people use it to mark entire bedroom complexes with a single keystroke. Custom macroses inside DF are coming before long (to some extent already here).

What difference is forcing players to press multiple keys supposed to achieve? //mostly rhetoric question

[Sean Mirrsen]

Heh. That too. Though I suppose the need to use external programs would be difficulty enough for some people.

As much as I'd like a meaningful ingame cost for a plain rampless channel, I kinda don't see it working. Using up a stone or a wood log for a section of channel is not only expensive for the kind of usefulness it has (compared to a wall), but doesn't make too much sense in-game either. Just adding to the required keystrokes is also just a partial solution, as described by the previous post. Adding a tool requirement for channels, such as a rope that would be reused, would likely require a little too much coding, and would still imply that miners have Cutscene Power To The Max and can climb while they're mining, but not anytime else, at least until that particular ability is fully implemented. Nevertheless, this kind of thing would likely be the best long-term solution (if the concept of an omni-useful "dig down" designation is still too aggravating for people). The short-term effective and/or meaningful solutions, i.e. the ones we can see while the game's still a .31 version, are, in no particular order:

• Return the old channelling as-was - Though likely to upset the half-or-so people who like, or don't mind the new channelling, this would be the quickest "rollback" method, if it's indeed so quick in code.
• Return the old channelling as an additional designation - Requiring an unknown amount of additional time and effort, this is the "workaround to the workaround" solution, that would likely keep most of the people pleased, but would worsen the game's "patchwork" state as the core of the problem - the bugs - would be, if temporarily, overlooked. This option barely fits into "short-term" as it's uncertain how much would need to be changed to add a new designation.
  
  • Rollback once the system works could be an optional continuation of this, but very likely not in the short term as the bugs are likely numerous and adding required secondary functions is probably a sizeable task in itself. Nevertheless, this outcome could be the best for both the game and the players - serving for the versatility and adaptability of one, and for the intuitive and varied use by the others.
  • Ingame balancing can also take place instead of rollback, with ramped channels being somehow different from plain channels in the way they are constructed, thus serving the admittedly weak point about making defences and general differences in usefulness. While not a very likely outcome since finding meaningful ways of balancing is difficult, this is the other variant that would effectively leave all bases covered and everyone pleased.
• Ignore the problem altogether - You can't say it can't happen, even if we didn't yet get any major bugs continously overlooked. Keeping the system as-is without actively seeking to fix the bugs wouldn't require effort (surprise!), but wouldn't be much of a popular move.
• Patch holes in the system - As a short-term solution, several stopgap measures could surely be taken to prevent the most obvious bugs from surfacing. Without adding functionality like removing ramps from above, however, this wouldn't sit too well with the 44.4% (as of this post) people who like their channels clean. While not being the bestest solution by far, this could nevertheless still be implemented in a reasonably short timeframe, and whether or not it's actually worthwhile depends on how difficult option 2 is to carry out.

I believe these four, plus sub-points, are the only variants we can get - short-term. New poll, perhaps? Only this time, not which is "better" - as it's pretty much clear which is which - but which you think we're actually going to get? Only Toady knows which way to take the game, but we can have a game of our own in guessing.

(p.s.: I've revised the post several times; I think I got all the options right. If you can see anything else that can happen, feel free to share)

[Lord Darkstar]

Sean, that looks like the options we've been able to determine through all of this debate and discussion. Nicely done.

[Zombie]

Those do, indeed, look like all of the short-term solutions plausible. Good work!

[Andir]

Not sure why "balancing" is required though...

It's a freaking sandbox game.  Designating a tile to be channeled out isn't game breaking unless you think it is and you use it.  Just don't use it if you think it's "overpowered."

[Sean Mirrsen]

We've discussed that. For the game, it's better if the challenge is already there, and the player doesn't have to consciously limit himself. "Balancing" in game design, as far as I understand it, is making proportionate rewards. Therefore, if something is straightforwardly better than something else for most of their mutual uses, then the game can afford to make obtaining it harder. In this case, the difference is subtle, but apparent. There would have to be a mildly annoying "cost" to pay for using the better designation, but we couldn't so far find an appropriate "cost" that'd fit into a short-term solution. The long-term solution is improved AI and custom job tools, that'd render the differences in usefulness less noticeable and provide a non-intrusive way of differentiating the construction process of the two.

[Andir]

Careful where you stick your idea of challenge in the game, lest you be shoving your ideas down other people's throats.

I understand balance in game design... but we aren't talking about different weapon types here.  We're talking about a method of creating a pit in order remove the tedium of the process of digging a pit.

[Sean Mirrsen]

Different ways of creating a pit are also, believe it or not, part of game design. A pit, a channel, or a handcrafted underground stadium are all made with the basic game tools. They're parts of the game, that you create. In this specific case, we have two parts of the game - one is passable, looks irritating, and presents an immediate safety hazard - the other is clean, insurmountable, and perfectly safe unless someone pushes you in. In addition, the parts are essentially the same thing, and one can be seen as a different form of the other. In plain ingame terms, having the two being equally easy to construct is an imbalance, however small on the grand scale of things. It's not a squadron of cheap sentient cruise missiles that can dropkick your entire defence perimeter before you zoom in on the matter. But it ever so slightly stands out. You can overlook it - there's a reason the "Balance" is a sub-option in the list - but if there's a decent way of marginally increasing the more useful item's construction cost, it would make sense in the game, and ultimately allow for a better experience.