Print

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

[fricy]

I have a double helix minecart spiral carved, the blue track is for going down, the red one has highest rollers on every odd level for bringing up carts.
I double checked to make sure that all tracks are connected, all ramps have two directional tracks carved on them, the game does not give me the dir/connect error, so I don't make the common mistakes.

The problems is, the downward track is slow. It looks like somehow the cart gets stuck on every 4th downward ramp for 18-20 ticks, collects some momentum, continues, then get stuck again on the 4th ramp. I counted the ticks, and this pattern emerges:

Spoiler: Ticks spent on a tile (click to show/hide)

Code: [Select]

feature tick
track 1
DOWN ramp 4
track/corner 1
DOWN ramp 4
track 1
DOWN ramp 3
track/corner 1
DOWN ramp 18
track 1
DOWN ramp 4
track/corner 1
DOWN ramp 4
track 1
DOWN ramp 3
track/corner 1
DOWN ramp 19


So it looks like somehow every second corner track slows down the cart to a crawl. Which shouldn't be possible, as we can see the cart spends 3-4 ticks on the other 3 ramps, resulting in at least 40-50.000 extra speed, and the corner speed penalty is only ~1000.
The upward ramp with the rollers does not show this slowdown, the speed is consistent, but when I changed the direction by switching off the rollers and  pushing the cart down that way the same thing happened.

Does anybody have any idea what's happening here?

EDIT: It's even more weird than I thought before: I can derail the cart at the corner (just before the slowdown is happening) if I remove the corner wall. So it has more than 50.000 speed on entering the level, that goes close to zero at the corner or the entry point of the next down ramp. I built a quick test fort to make sure that it's not my save that's bugged. Same thing is happening.
It looks like the ramp is behaving as an up and down ramp at the same time. The cart enters, loses momentum as if on an up ramp, then rolls down, continues and gains momentum. Weird.
Edit2: IT's definetly the ramp, not the corner: If I build a track over the ramp and lead the minecart onto another ramp it can go up without a problem.

[Larix]

As i understand it the slowdown is due to the cart actually derailing in the spiral. A derailed cart will fail to go down a ramp and instead jump over it, either landing on solid ground on the far side, or, more likely in a ramp shaft, colliding with a wall, dropping its contents and 'falling' onto the ramp below with zero horizontal speed. The cart's coming to a stop and speeding up from zero over and over.

I haven't tried out series of corner track ramps much, so don't know if there's a usable configuration for a 'helical' 10+ z-level course, but for going down a lot of levels with push or ride, a straight ramp works perfectly (can even take a corner ramp or three, if they're far enough apart) and a high ramp made of nothing but corners is still perfectly usable with guided carts, although that feels kind of lame on the downward route.

[fricy]

Yeah, on second thought it looks like simple derailing. I thought a spiral would work with high speeds, but instead is limited around 2tick/tile. Damn, than it may be better to drop the carts.

[krenshala]

You could do the spiral up and a straight ramp down.  You would just have to either run it off to one side, U turn then head down, or head down then U turn and come back.  It is probably more likely to derail with the latter, however.

[fricy]

Ok, next question: what's the secret of a straight ramp that does not slam high speed carts into a wall?
I have a few, but they are only used at high speeds with empty carts, loaded cart are guided. I tried modifying the spiral design, so the ramps are not directly adjacent to a corner to see if it's enough to count as a straight ramp, but that does not help. Do you just channel away multiple floor tiles where you want to change a level, so the cart drops onto the track below? Something else?

[Larix]

I have a track route going down ~100 z-levels on two long ramps (have to bleed off speed and change course once), and it's set to "ride" downwards. Very fun to watch, and the _riders_ have so far not come to harm. Anything that gets in the way has good chances of being flattened.

From what i've observed, past a certain speed (probably standard derailing speed), a cart will continue on its way even if the track suggests otherwise, unless forced by walls blocking the 'straight route'. This includes 'jumping' over ramps instead of going down them. A derail-speed cart moving on flat track will ignore a down ramp opening under it (it can even jump up, causing even more problems). When rolling down a ramp, the cart actually counts itself as going downwards, thus a directly following downward ramp continuing straight onward is compatible with the cart's desire to continue in its direction and gets followed without issue. This makes straight rows of ramps work.

I don't know what exactly a cart going down a spiral ramp actually tries to do, but the evidence of continuous collisions says it fails to follow either the corners or the down direction at some point. This is even more confusing because it only happens if the corner ramps are close to each other - directly adjacent or less than five or so steps apart.

[Snaake]

As others have said, it's just because the cart is ignoring the downward ramp on the straight sections, tries to keep going straight, but there's a wall there that it slams into, so it then drops down onto the ramp and starts speeding up again.

The aforementioned straight ramp down would work, I think. Or you can shrink the spiral to only be 2x2, you don't really need the extra space for a purely downward spiral (you would for an upward one to fit rollers and power transfer, or impulse ramps). Keep in mind that you need to have a wall tile behind each turn (or do doors work? can't remember). Or you could just channel 1 tile on each level (with the ramp up from that level provided by a channel from the level above), I think that should work too. The (d)ug tiles are your problem. In effect, you don't need the "straight ramp down" to actually be straight, it just needs to only have ramps, you can put bends in it all you want.

Or you could enforce a speed limit at just below the derail-into-roof speed using track stops.

[fricy]

Thx for the advice, looks like I'm either regulating the speed, or dropping the cart. The minimum I need is a 3x3 helix to put the gears in the middle. I built one with all ramp tiles, but it still derails.

[Gentlefish]

Track stops work wonders for slowing carts down. Try installing a low powered track stop every odd level as well?