Print

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

[levraininjaneer]

Hi

If I have a single track going down into the depths and I want to join a second track (also for a pushed cart) in the middle can it be done without risk of derailing the first track? I was thinking a "h" type junction might work where the "long leg" of the h is the existing track and the curly one is joining. I am hoping that some of the downward momentum is conserved when joining minecart hits the "T" and that it can then proceed... ?

[levraininjaneer]

Some experimentation and I have an answer. And a solution.

When a "pushed" minecart reaches a T-junction (from "leg" of the T), it just stops. No solution there.

But a cunning dwarf may build the "joining" track above the the "joined" track and then cut a hole. Under gravity, the minecart simply drops onto the bottom track and becomes railed there. Pretty nifty. Be careful, though: depending on speed, it quite likely takes more than one tile to make the z-level transition, so if your set-up looks like this (side view), it won't work:
_______  WWWWW (W = Wall _ = floor with track)
_______________
Because the minecart will collide with the wall first and then drop to the below track with zero forward velocity.
What you need is this:
______                     WWW
_______________________

This does require a bit of a stright on the bottom track, of course. But if you are going downwards, speed should be a problem of redundance, not lack.

[Fleeting Frames]

When a "pushed" minecart reaches a T-junction (from "leg" of the T), it just stops. No solution there.

Yes there is. Observe:

Destination
╚←cart 1
║
↑
cart 2

Carts coming from destination will go either straight or take the turn depending on their speed.

Short guide:
There are 3 kinds of tracks: flat track, corner with 1 input and 1 output, ramp with 1 track into exit and 1+ tracks into wall.
The first one is the same no matter how it is drawn - even on a ramp.
The second one converts it's speed to sideways direction, subtracting 1/20 of a push, centering the cart along the new direction, unless there isn't a wall blocking the exit and cart is going fast enough.
The third one substantially accelerates cart towards the exit direction. 4 ticks on one fall little short from 1 dwarven push.

Combining the second and third one results in the ramp acceleration applied after the corner, giving potential derails due sideways velocity and when chained dropping 6/20 of a push no matter the travel direction.

Whenever the cart passes from ramp with 1 exit direction to a normal track or ramp with different exit, it can teleport to the end of next tile, passing it in 1 tick instead of whatever the speed would dictate - convenient way to avoid speed loss!

[levraininjaneer]

Ah, I see what you did there.

Another, speed-independent way of achieving this is (side view)

\___  |W
____
       \___
Where |W = Wall  _ = Flat Track and \ = Track ramp (directly below hole)

I've also reckoned that a variation on the above setup is the most sure-fire way to make sure your speeding mine carts ever injure your dwarves.


\___  |W
      |
       \___T1__T2

Now, T1 and T2 are track stops with respectively high friction highest friction. I've tested (even without T1) no dwarf will get injured by the minecarts, regardless of what speed they had before hitting |W.
You can then set your route stop at T2 and know that your haulers are safe :-D

[Fleeting Frames]

That might be still speed dependent on whether carts coming from destination will go onto the first or second track - mainly, will they fly at high enough angle to jump the hole when travelling from east to west or not :p

Fixed by adding track/ramp on the tile right before the drop on the lower level to prevent flying when coming from destination, though it has left me curious: What sort of hauling system are you using that you want carts go from start 1 or 2 -> destination > start 1?

Also, you did test with max speed (2,7 tiles per tick) minecarts when sending minecarts up the ramps on second diagram, right? I recall dwarven space program used 6 highest friction track stops to stop the flying cart without injury to the dwarf riding it.

Though I suppose it is safe way to send empty carts from above :p

Speed-independent returning of carts from single destination to multiple start points is most simply done with pressure plates I think (each start point having a pressure plate linked to door covering a L-turn), but if they're empty carts and instant return isn't a concern might be able to have a system where N carts pile onto a non-accelerating track and have the N+1th cart go on a sideways bit to punch them towards their destination at 90° from the non-accelerating track before being left sitting at the end of of it for next wave to go to 1st destination.

[levraininjaneer]

No, you misunderstand my system somewhat.

My rails are strictly one way. Another track leads upward, and there joining is no problem as a dwarf is always guiding the cart. And this solution is speed independant as the cart is stopped "dead" as it hits the wall before falling and is then only accelerated by the ramp on which it falls, which is slowed and stopped by single Track Stops.

Reason for guiding back up is again, the thermodynamicist in me is ired by impulse ramps, and using a power source seems a bit FPS intensive, but more importantly: too much axles and stuff to build all over the place. Reason for a different track: well, considering high speeds on the downway track: safety.

Though I suppose with my newest design it could work well. I've made a set of Quickfort templates to designate upward or downward spiralling tracks: it's only 9 tiles surface area, so a vertical axle going up in the middle, with some gear assemblies and rollers could conceivably power the whole upward shaft with relatively little construction effort.

If I use Bumber's Waterwheel, design... hmmm: maybe this would be cool. I wish I could knew how much FPS impact to expect from that... But then I could always turn it off and revert to my old design if its too bad...

Interesting... (rubs hands together)

[PatrikLundell]

I don't think rollers and shafts have much of an FPS effect. The penalty comes from the power generation, unless a low impact one is used (I use flow induced aquifer slits).
My normal cart track design is an upwards spiral track (impulse ramp driven) with a drop shaft in the middle for the downward direction. I've converted this into a roller driven one (with the shaft in the middle and the downwards shaft shifted elsewhere) once just for the hell of it, and it worked quite well.

[levraininjaneer]

Ah, thanks!

Have you seen the cavern-water to map edge power generator? I figure it should be too bad on FPS. What do you think?

[PatrikLundell]

I haven't seen it, but if you use the suggestion of blocking the physical flow when logical flow has been induced I would expect the FPS impact to be virtually zero. Also, it doesn't seem a small flow has too bad an effect either.

[Fleeting Frames]

As far as power sources go, should you embark on waterless map windmills might be an option, and power transit can be toggled by pressure plates (should their radio not irk you :p).

Depending on your drop depth, it might be possible to have one cart travel upwards and roll back, then have different cart pushed from behind hit it and push it farther up.

Otherwise, a retracting bridge linked to a cart-toggled pressure plate might fling it up towards the start point, even if bridges are bad for your fps. A raising bridge would sometimes atomsmash carts as they fall straight back down, of course - iirc there is about 25% chance to remain on the bridge for 5x5 bridges. The height flung is not consistent either, but can be up to 10z.

If you have more digging room, then a fifth option might sometimes be to have the cart do a bit of flying. Idea would be something like this:
A                        B
  \                     /
   \                   /
    \/W ........W\/
So that cart pushed from A goes flying and goes straight to B skipping the ramps and vice-versa. Look ma, no track in between!
Bit tricky, though, given that gravity and ramps give quite close acceleration for tiles/second, but avoiding friction from track would help widen the margin.

[levraininjaneer]

Ah, some very creative ideas flying around.

Well, thing is, my carts are riding to transport stuff many z levels: about 30 to 60 at a time. I use them because transporting flux stone/coal/wood to my fortress would take many, very long hauling trips otherwise. At the moment, the 1 dwarf can do what about 5 can, normally and with a better FPS than the five.

Your ramp design would be an interesting solution for horizontal transport, which is not my objective at the moment...

I think I would need a crazy amount of windmills to do this, as they don't give so much power in the first place, and there each block of axle is 1 power needed, not to mention the rollers and gear assemblies in-between.

[PatrikLundell]

I'm not too keen on windmills, both because they produce paltry amounts of power, provided they produce any at all, and because they're a pain to build in a building destroyer safe manner, including securing the power transfer from infiltrators. Each windmill requiring a gear that soaks up a substantial part of the power (possibly) produced doesn't help either.

I have to say I don't understand how you transmit power via pressure plates, though, unless you mean using pressure plates to activate bridges.

[Fleeting Frames]

Link a pressure plate before a hole to gear assembly linked to rollers pushing up. Cart goes downwards, hits the pressure plate, thus disactivating the rollers for 100-110 ticks depending on cart speed and goes on to bottom. Coming back, the rollers will be on and bring the cart up.

In above post, each paragraph is different thing.

And yeah, windmills do provide some additional architectural challenge.

levraininjaneer: Well, just make sure your dwarves are not slowed down by carrying minecarts on their shoulders :p

[gordy]

What makes designating multi level tracks easy or convenient that you would consider it? (Understanding ease and convenience is not dwarvenly...)

[Fleeting Frames]

Macros help. Ctrl+R, designate repeating pattern and go to next one, Ctrl+R, Ctrl+P / Ctrl+U # Ctrl+P.

Desire to make a cool project also helps.

Though yeah, everybody has their own limit to time spent on thing. The main difference when hitting the cavern roof with stairway or ramp spiral is the time to designate, after all :p