So I figured we should take the powered rail discussion from the various 1.5 discussions and make a thread exclusively about finding how powered rails work, how best to use them, and various tricks that can be used with powered rails. I'll start with my experiences so far.
To all who are not aware, below is the crafting recipe for powered rails: [] where red wool = redstone dust.
a single powered rail does not push a cart very fast.
multiple power rails can be chained together for increased effect (an all power rail track will be very fast)
multiple power rails can all receive power from the same power source as long as they are all connected together.
A single redstone torch can power the adjacent rail and up to 8 rails in either direction
power rails do not have curves, and will form two seperate tracks (presumably following the East and South rule) but carts will jump tracks between the two.
A cart hitting an obstacle (block, player) while on a powered rail will cause it to change direction. Presumably this would cause a cart to be able to move back and forth along the same track indefinitely without player input.
Carts can be "shunted" together to form a multi-cart "train" by adding multiple carts on the same track and causing collisions. After repelling each other they will eventually find equalibrium and move in tandem like the balls in a newton's cradle. So far I've successfully made at least one eight cart long train. Could be interesting to now have minecraft model railroads with both freight and passenger trains :tongue.gif:
I made a table/graph of powered rails and the distance it pushes an empty minecart. It starts out doubling, then slows down to increasing the distance by half a block every powered rail after 6 boosters. These distances are not including the powered rails.
The redstone travels the normal 15 blocks down the rails.
I also find them to be quite expensive, and so foresee them only being used as launchers for our current boosters. I designed a couple of these, minimizing it down to 3 wide x 2 deep x 2 high or 4 wide x 2 deep x 1 high, working great with only one powered rail.
I did some tests on a flat terrain with jnwhiteh, with the following configuration (top down view)
1 Powered Rail booster: ...
[] []
[]
2 Powered Rails booster: ...
[] []
[]
Where the stone slab is a button, and the torch is a redstone torch powering the "powered rail booster" that is 1 normal rails block away from the launcher rails connected to the button.
Number of not including the launching pad - Distance traveled from the launcher block to a near stop (I hopped out when it got so slow it takes about 10 secs just to travel half a block)
(A redstone torch is placed every 8 powered rails, since their powered distance is only 8, (yes I know we could have placed them 16 apart))
All distances above are with a player-loaded minecart. So the conclusion is that there is diminishing returns for every additional rails and seems to be roughly +40, +30, 20, 20, 15, 15, 10...
The resulting conclusion is that the speed boost is less if the cart speed is high, and the speed boost is more if the cart speed is low. Thus it is not a good idea to place 2 rails next to each other as the 2nd consecutive one will give less of a speed boost than if it was further down (when the cart is slower and thus gives a higher speed boost), and better to space them apart every X number of blocks. Determining what X is though, is difficult. However, spacing them too far apart will mean slower travel speed, so the repeat interval X determines how fast you want to reach your destination.
The minecraft wiki currently (as of 10 min ago) states that a single boosts +128 distance on flat ground, which I cannot replicate at all, so I will assume it is wrong, or it is tested in some manner that is not known to us.
It believe X cannot be larger than 64, because it appears each powered rail only boosts +64 distance max (based on the 0P distance of 65).
1 Powered Rail booster: ...
[] []
[]
gives a distance of about 105, but shifting the single powered rails down 16 blocks to
...
[] []
[]
(i.e 16 blocks between the launcher and the other :GoldBar:) gives about an additional 15 distance before the cart stopped moving, but the 120 total distance is still under the 128 theoretical max for 2 powered rails.
I have found that the detector rails do not transmit its power uphill. This is making it a little more difficult to create a decent system to go uphill.
You can start a cart if you put it on a powered rail next to a stop block, then power up the rail. It's a convenient way to start a journey without having to jump into a moving cart. I haven't been able to start a cart that did not sit next to a stop block.
That's because the stop block determines which direction the cart should start moving towards. Without the stop block, it would be ambiguous, so it does nothing unless the cart is already moving (above a min threshold that nobody knows yet).
I have found that the detector rails do not transmit its power uphill. This is making it a little more difficult to create a decent system to go uphill.
You don't have to use detector rails to power powered-rails. Just put redstone torches next to the powered rail, which are far cheaper to make than detector rails, and thus make the powered-rails "always on" (up to a max powered-rails distance of 8).
A powered rail without power will cause carts to slow down instead of accelerating them. I suppose putting a detector rail next to a powered rail will enforce one-way traffic.
Detector rail followed by booster rail, connected with a 1-torch delay = 1-way track.
I also discoverd that booster rails only have power to go up 1 (!) block.
so you can only have a 1:2 upward track, half made of boosters.
Rollback Post to RevisionRollBack
Feel free if you to ^ my posts if you find them helpful.
also this is quite strange, booster tracks seem to have some kind of memory, if a cart ran over them, south-west rule seem to be overitten in case it runs orthogonal into that track after a few turns.
Rollback Post to RevisionRollBack
Feel free if you to ^ my posts if you find them helpful.
You slow down when you turn corners so you should put the Power Rails on corners. My post doesn't sound as mathematical as all of yours! 1+1=3 YAY!!! I'm clever like you guys!
hom much do you slow down? test results pls!
Rollback Post to RevisionRollBack
Feel free if you to ^ my posts if you find them helpful.
I have found that the detector rails do not transmit its power uphill. This is making it a little more difficult to create a decent system to go uphill.
You don't have to use detector rails to power powered-rails. Just put redstone torches next to the powered rail, which are far cheaper to make than detector rails, and thus make the powered-rails "always on" (up to a max powered-rails distance of 8).
I was aware of this, but if people were wanting to create a one-way track that goes uphill, they would have to put the detector rail on flat ground.
Also, for going uphill: 5 booster tracks will allow you to go 4 blocks without slowing too much, and 5 blocks with some slowing. But, you cannot make it to the 6th block. 4 booster tracks: 3 blocks w/out slowing, 4 with slowing, and can't make it to the 5th. 3 booster tracks: 2 blocks w/out slowing, 3 blocks with slowing, and can't make it to the 4th block. I see a pattern here. Note, that this is with the boosters and following tracks on an uphill incline.
Edit* This would mean that, for the smoothest ride, have one fewer normal rail in between each set of powered rails than you have powered rails. Now the question is what would be the most efficient ration of normal rails to powered rails for uphill travel? I think a 3:4 ration is pretty decent.
Maybe someone knows this. If a powered rail segment on a hill is sufficient to carry a person, does that mean it will be sufficient to carry a container car as well? Or are the two `weighted' differently?
Has anyone tried combining powered minecarts with boosters? Perhaps they synergize well?
From the #4 post where I posted some test results, it would appear to me that traditional boosters do not respect the 8m/s travel speed limit (they increase the internal speed far beyond that), but powered rails do respect the 8m/s speed limit. Because of this, you need N powered rails on a slope to drive a cart over a slope that is N high/long (by ThatOneLundy above) to maintain the same resulting speed at the top, or to start it up from a standstill (you can deduct some from N if you have initial velocity on the cart).
So they don't synergize well.
We're also predicting that Notch might remove the bug that causes the traditional cart boosters to work in the future and patch 1.5 might be the time to start transitioning over.
Hey gang, I just invented a way to detect an empty minecart with the new 1.5 update.
I am making a video as we speak which I will attach to this post once it uploaded.
The basic principle is this:
If a cart comes along from the left in this crude drawing (sticks are rails) an empty cart will have less speed than a full one, causing the empty cart to drop on the first (left) track, and a full one to jump further and land on the second block. Using this, you can make a cart detection system without relying on the traditional boosters.
Just wanted to get this out there, hope you can use it :biggrin.gif:
i followed your vid and this did not work for me it jumped 2 spaces
Thought I would share my personal research on the comparative strength of powered rails vs. boosters. I conducted three types of tests:
[*:9yryqxf8]Distance per rail
[*:9yryqxf8]Optimal spacing for speed
[*:9yryqxf8]Climbing
Control Group
The control group was a simple resetting booster, using a single oscillating cart to propel double-stacked booster carts into the player cart when a stone pressure plate is depressed. The double-stacked booster carts parallel the player cart for 2 rail lengths plus the tile with the pressure plate.
Distance Per Rail
To test distance per rail, I created a loop around my fort. I connected both the resetting booster and the powered rail input tracks to the loop. I then measured how many flat tiles various power sources could cover.
Clearly these numbers agree with previous measurements. The first rail gives you the biggest boost, with diminishing returns thereafter. The disparity between rail power and the booster was also striking. This booster was pretty much as short as I could make it.
Optimal Spacing for Speed
I wanted to see what interval between tracks yielded good performance, because distance per rail doesn't tell the whole picture. To test this, I measured the time it took for a given configuration to complete the first, second, and third loop around my track. The control booster can complete the loop in roughly 18-20 seconds, slightly contingent on server lag.
The configurations vary by how many rails are used each time a boost is applied and by how far apart each section of powered rail is. For example, the first test used 1 powered rail at a time, 50 blocks apart. Ultimately, all configurations were able to reach their best performance by the second lap, so I've only included the first and second lap data here. The efficiency value is interval/rails, giving a snapshot of how many rails per distance traveled you will pay to achieve this performance.
It's pretty clear that 18-20 seconds was the top speed around my loop. Most configurations were able to achieve it by the second iteration, which is to say that using most of these set ups will get you to top speed after about 100 or 150 tiles. The points of note are:
[*:9yryqxf8]1 powered rail every 50 is the most efficient method tested, but quite slow initially and never achieves top speed
[*:9yryqxf8]2/50, 1/25, and 3/75 are all able to achieve the next best efficiency, but 3/75 couldn't quite maintain top speed
[*:9yryqxf8]Once a certain separation distance is reached, it is very difficult to maintain maximum speed over the interval. Even 8 powered rail once per loop couldn't keep up top speed the whole duration.
The obvious recommendation as a result of this test is to use 4 powered rail to launch and then either 1 every 25 or 2 every 50 to keep it going. Of course, these numbers are just recommendations - there may be a more optimal spacing, but of my test data, they seemed to perform the best.
Climbing
To test climbing, I created a 1:1 ascent for 35 tiles. The control booster had absolutely no problem rocketing to the top and probably could have kept going for awhile. No slow down detected during the ascent.
I tested three configurations of powered rails. The first method used powered rails on the ground to gain speed and then attempt to climb.
rails height
1 2
2 4
4 7
7 9
10 10
20 15
No surprise here - generally diminishing returns as we go from 2 height/rail at 1 to to .75 height/rail at 20.
Next I looked at how many tiles in height you could continue rising after using ascending powered rails. I used a single powered rail on the ground to inch me into the base of the slope, at which point I used ascending powered rails to boost me upward. I counted how many additional tiles I covered. I didn't record a lot of the data points here, because it was apparent that I wasn't gaining a lot over increased number of rails.
rails tiles
3 3
16 9
So in the first case, I used 3 rails to cover 6 vertical tiles, and in the second, I used 16 rails to cover 25 vertical tiles. We can see that vertical rails help actively overcome the deceleration of the slope, as our diminishing returns from adding new rail is not as severe as when adding rail from the ground.
Lastly, I tried using alternating sets of powered and unpowered rail. I tried 1:1, 2:2, 3:3, 4:4, and 5:5 configurations, as well as 1:2 and 2:3 after building up speed. 5:5 ratio provides insignificant lift to make it past the first block of powered rail. The other four equal ratio splits all made it to the top in 9 seconds. Their performance was deceptive. 1:1 feels smoother and faster, but did not actually arrive any faster.
After building up speed, all configurations could manage a couple repetitions of a more efficient ratio (1:2 or 2:3), but they were not able to maintain speed using these set ups.
Conclusions
[*:9yryqxf8]4 consecutive rail will accelerate a cart to maximum speed.
[*:9yryqxf8]Maximum flat travel speed can be maintained with 1 powered rail per 25 tiles covered. Both 1:25 and 2:50 performed well. 3:75 showed slight decrease and longer distributions were slower.
[*:9yryqxf8]Using inclined powered rail to ascend is more efficient that building speed while flat to ascend.
[*:9yryqxf8]Ascent requires roughly 1 rail per 2 altitude. 4:4 seems to be the longest stretch that will work, with alternating powered and unpowered at 1:1 providing the smoothest ascent. This is unfortunate, because it's a pain to power them all.
To all who are not aware, below is the crafting recipe for powered rails:
a single powered rail does not push a cart very fast.
multiple power rails can be chained together for increased effect (an all power rail track will be very fast)
multiple power rails can all receive power from the same power source as long as they are all connected together.
A single redstone torch can power the adjacent rail and up to 8 rails in either direction
power rails do not have curves, and will form two seperate tracks (presumably following the East and South rule) but carts will jump tracks between the two.
A cart hitting an obstacle (block, player) while on a powered rail will cause it to change direction. Presumably this would cause a cart to be able to move back and forth along the same track indefinitely without player input.
Carts can be "shunted" together to form a multi-cart "train" by adding multiple carts on the same track and causing collisions. After repelling each other they will eventually find equalibrium and move in tandem like the balls in a newton's cradle. So far I've successfully made at least one eight cart long train. Could be interesting to now have minecraft model railroads with both freight and passenger trains :tongue.gif:
The redstone travels the normal 15 blocks down the rails.
I also find them to be quite expensive, and so foresee them only being used as launchers for our current boosters. I designed a couple of these, minimizing it down to 3 wide x 2 deep x 2 high or 4 wide x 2 deep x 1 high, working great with only one powered rail.
What are these "detector" rails Notch mentioned?
-
View User Profile
-
View Posts
-
Send Message
Curse Premium1 Powered Rail booster:
[]
[]
2 Powered Rails booster:
[]
[]
Where the stone slab is a button, and the torch is a redstone torch powering the "powered rail booster" that is 1 normal rails block away from the launcher rails connected to the button.
Number of
0 - 66
1 - 105 +39 (configuration shown above)
2 - 135 +30 (configuration shown above)
3 - 161 +31
4 - 179 +18
5 - 200 +21
6 - 213 +13
7 - 228 +15
8 - 238 +10
9 - ?
10 - 262
11 - ?
12 - ?
13 - ?
14 - ?
15 - 307
(A redstone torch is placed every 8 powered rails, since their powered distance is only 8, (yes I know we could have placed them 16 apart))
All distances above are with a player-loaded minecart. So the conclusion is that there is diminishing returns for every additional
The resulting conclusion is that the speed boost is less if the cart speed is high, and the speed boost is more if the cart speed is low. Thus it is not a good idea to place 2
The minecraft wiki currently (as of 10 min ago) states that a single
It believe X cannot be larger than 64, because it appears each powered rail only boosts +64 distance max (based on the 0P distance of 65).
1 Powered Rail booster:
[]
[]
gives a distance of about 105, but shifting the single powered rails down 16 blocks to
[]
[]
(i.e 16 blocks between the launcher
-
View User Profile
-
View Posts
-
Send Message
Curse PremiumThat's because the stop block determines which direction the cart should start moving towards. Without the stop block, it would be ambiguous, so it does nothing unless the cart is already moving (above a min threshold that nobody knows yet).
-
View User Profile
-
View Posts
-
Send Message
Curse PremiumYou don't have to use detector rails to power powered-rails. Just put redstone torches next to the powered rail, which are far cheaper to make than detector rails, and thus make the powered-rails "always on" (up to a max powered-rails distance of 8).
Detector rail followed by booster rail, connected with a 1-torch delay = 1-way track.
I also discoverd that booster rails only have power to go up 1 (!) block.
so you can only have a 1:2 upward track, half made of boosters.
hom much do you slow down? test results pls!
I was aware of this, but if people were wanting to create a one-way track that goes uphill, they would have to put the detector rail on flat ground.
Also, for going uphill: 5 booster tracks will allow you to go 4 blocks without slowing too much, and 5 blocks with some slowing. But, you cannot make it to the 6th block. 4 booster tracks: 3 blocks w/out slowing, 4 with slowing, and can't make it to the 5th. 3 booster tracks: 2 blocks w/out slowing, 3 blocks with slowing, and can't make it to the 4th block. I see a pattern here. Note, that this is with the boosters and following tracks on an uphill incline.
Edit* This would mean that, for the smoothest ride, have one fewer normal rail in between each set of powered rails than you have powered rails. Now the question is what would be the most efficient ration of normal rails to powered rails for uphill travel? I think a 3:4 ration is pretty decent.
By 'stop block' do you just mean any block that stands at the end of a piece of track? Or is there some specific block type/configuration?
http://www.minecraftforum.net/topic/1763406-absolom-a-maze-in-works/ - My current project, help wanted!
-
View User Profile
-
View Posts
-
Send Message
Curse PremiumYes, just a block that stands at the end of a piece of track is my definition of a 'stop block'.
[SSSS] [Skeleton]Support nice ideas. [Spider] [Zombie]
I've been wondering about this problem all day. Your solution is very interesting, I shall experiment!
Thank you
-
View User Profile
-
View Posts
-
Send Message
Curse PremiumFrom the #4 post where I posted some test results, it would appear to me that traditional boosters do not respect the 8m/s travel speed limit (they increase the internal speed far beyond that), but powered rails do respect the 8m/s speed limit. Because of this, you need N powered rails on a slope to drive a cart over a slope that is N high/long (by ThatOneLundy above) to maintain the same resulting speed at the top, or to start it up from a standstill (you can deduct some from N if you have initial velocity on the cart).
So they don't synergize well.
We're also predicting that Notch might remove the bug that causes the traditional cart boosters to work in the future and patch 1.5 might be the time to start transitioning over.
i followed your vid and this did not work for me it jumped 2 spaces
[*:9yryqxf8]Distance per rail
[*:9yryqxf8]Optimal spacing for speed
[*:9yryqxf8]Climbing
Control Group
The control group was a simple resetting booster, using a single oscillating cart to propel double-stacked booster carts into the player cart when a stone pressure plate is depressed. The double-stacked booster carts parallel the player cart for 2 rail lengths plus the tile with the pressure plate.
Distance Per Rail
To test distance per rail, I created a loop around my fort. I connected both the resetting booster and the powered rail input tracks to the loop. I then measured how many flat tiles various power sources could cover.
Clearly these numbers agree with previous measurements. The first rail gives you the biggest boost, with diminishing returns thereafter. The disparity between rail power and the booster was also striking. This booster was pretty much as short as I could make it.
Optimal Spacing for Speed
I wanted to see what interval between tracks yielded good performance, because distance per rail doesn't tell the whole picture. To test this, I measured the time it took for a given configuration to complete the first, second, and third loop around my track. The control booster can complete the loop in roughly 18-20 seconds, slightly contingent on server lag.
The configurations vary by how many rails are used each time a boost is applied and by how far apart each section of powered rail is. For example, the first test used 1 powered rail at a time, 50 blocks apart. Ultimately, all configurations were able to reach their best performance by the second lap, so I've only included the first and second lap data here. The efficiency value is interval/rails, giving a snapshot of how many rails per distance traveled you will pay to achieve this performance.
It's pretty clear that 18-20 seconds was the top speed around my loop. Most configurations were able to achieve it by the second iteration, which is to say that using most of these set ups will get you to top speed after about 100 or 150 tiles. The points of note are:
[*:9yryqxf8]1 powered rail every 50 is the most efficient method tested, but quite slow initially and never achieves top speed
[*:9yryqxf8]2/50, 1/25, and 3/75 are all able to achieve the next best efficiency, but 3/75 couldn't quite maintain top speed
[*:9yryqxf8]Once a certain separation distance is reached, it is very difficult to maintain maximum speed over the interval. Even 8 powered rail once per loop couldn't keep up top speed the whole duration.
The obvious recommendation as a result of this test is to use 4 powered rail to launch and then either 1 every 25 or 2 every 50 to keep it going. Of course, these numbers are just recommendations - there may be a more optimal spacing, but of my test data, they seemed to perform the best.
Climbing
To test climbing, I created a 1:1 ascent for 35 tiles. The control booster had absolutely no problem rocketing to the top and probably could have kept going for awhile. No slow down detected during the ascent.
I tested three configurations of powered rails. The first method used powered rails on the ground to gain speed and then attempt to climb.
No surprise here - generally diminishing returns as we go from 2 height/rail at 1 to to .75 height/rail at 20.
Next I looked at how many tiles in height you could continue rising after using ascending powered rails. I used a single powered rail on the ground to inch me into the base of the slope, at which point I used ascending powered rails to boost me upward. I counted how many additional tiles I covered. I didn't record a lot of the data points here, because it was apparent that I wasn't gaining a lot over increased number of rails.
So in the first case, I used 3 rails to cover 6 vertical tiles, and in the second, I used 16 rails to cover 25 vertical tiles. We can see that vertical rails help actively overcome the deceleration of the slope, as our diminishing returns from adding new rail is not as severe as when adding rail from the ground.
Lastly, I tried using alternating sets of powered and unpowered rail. I tried 1:1, 2:2, 3:3, 4:4, and 5:5 configurations, as well as 1:2 and 2:3 after building up speed. 5:5 ratio provides insignificant lift to make it past the first block of powered rail. The other four equal ratio splits all made it to the top in 9 seconds. Their performance was deceptive. 1:1 feels smoother and faster, but did not actually arrive any faster.
After building up speed, all configurations could manage a couple repetitions of a more efficient ratio (1:2 or 2:3), but they were not able to maintain speed using these set ups.
Conclusions
[*:9yryqxf8]4 consecutive rail will accelerate a cart to maximum speed.
[*:9yryqxf8]Maximum flat travel speed can be maintained with 1 powered rail per 25 tiles covered. Both 1:25 and 2:50 performed well. 3:75 showed slight decrease and longer distributions were slower.
[*:9yryqxf8]Using inclined powered rail to ascend is more efficient that building speed while flat to ascend.
[*:9yryqxf8]Ascent requires roughly 1 rail per 2 altitude. 4:4 seems to be the longest stretch that will work, with alternating powered and unpowered at 1:1 providing the smoothest ascent. This is unfortunate, because it's a pain to power them all.