1. Announcing Mekorama on the Web!

    Now anyone can play levels from the forum online, with one click!

    Dismiss Notice
  2. Psst! If you're new here, welcome! Please visit these pages first for information about the forum and Mekorama:

    Welcome! ¡Bienvenido! Selamat datang! Добро пожаловать! Willkommen!
    and
    Everything you want to know about Mekorama

    Dismiss Notice

Bot Paths When R limbos

Discussion in 'Game Behaviours' started by Fendi, Apr 15, 2019.

  1. Fendi

    Fendi Famous Member

    Messages:
    54
    Levels:
    96
    Albums:
    4
    Likes Received:
    1,585
    Joined:
    Dec 24, 2018
    20190415_112019.png

    We found this when build the third collab level "Pill Mill". What do you think? Or, has anyone ever found something similar?
     
    Last edited by a moderator: Apr 15, 2019
    Ray Aznable, BangTe, trids and 2 others like this.
  2. Makaroni

    Makaroni Member

    Messages:
    24
    Levels:
    9
    Albums:
    1
    Likes Received:
    64
    Joined:
    Dec 30, 2018
    Thanks to @Fendi Rahardian & @DDD for presenting such Limbo bots which seem to have special powers for several moments and thanks for a fun time checking them out. :thumbsup:
    Nick B. Fury :cool: is already observing the Limbo bots in their gym and considers to form a new team: the :sneaky: a-R-vengers!

    But look :eek: what happens if they can't make it to the next ball motor in time. They are losing all powers and are only a :confused: shadow of themselves. Same if the next ball motor comes too early. :( But during the few seconds they are in Limbo mode, they are strong as any :mad: Hulk, I mean, as any Hercules (compare Hercules push by B here, B looks at that push a little Limbo as well).

    mekorama-46.png
     
  3. Fendi

    Fendi Famous Member

    Messages:
    54
    Levels:
    96
    Albums:
    4
    Likes Received:
    1,585
    Joined:
    Dec 24, 2018
    @Makaroni As you mean, we have considered this. In the design of the "Pill Mill" level and the "Red Bridge" demo level, R is flanked by a block on either side, this is to maintain the stability of Limbo R, and we only use it at the last step, while for Limbo R with a long lane, I think it needs design elements that need to be considered, maybe the Limbo R needs to be normalized first after passing the ball motor, Limbo R is normalized with a block or any element, then after normal R, it can be redirected to pass the ball motor mechanics. What do you think?
     
  4. Makaroni

    Makaroni Member

    Messages:
    24
    Levels:
    9
    Albums:
    1
    Likes Received:
    64
    Joined:
    Dec 30, 2018
    @Fendi Rahardian I would think at the moment, that there are two instability types playing possibily a role here.
    • I think, we cannot influence too much the intrinsic in-/stability of the Limbo walking R. I mean such a Limbo-ing R that is not in recognizable interaction with any other block. Such an influence would need to happen near the source of the effect, i.e. near the ball motor. The ball motor is a free movable object, maybe placing blocks around it, blocks that would restrict its movement, might have a fine influence on the rotation of the ball. But well, who can say if there is such influence or not, we might not see any effect, but it might be still there. As you demonstrated the other day, in presense of motors much unexpected stuff happens. And might it be still some long distance influence of the ball motor itself.
    • The other type of instability is enforced by blocks interacting (in a visible way for us) with Limbo R on its path through the level. It is true, for the a-R-vengers level card I needed to select carefully such interacting blocks, some worked, some didn't. For the moment I would put this instablility aside, which we would use on purpose in levels. In my video, I would say, this kind of instability enforces the Limbo walk on B (the low ceiling put B in Limbo position, the autopilot alone makes B not move the first draggable, it is easier to step across it, but in Limbo position the autopilot does that Herkules push for B).
    It might be as well, that there is no artifical separation of instability types as I just imagined. Might be just one type of interaction. Sometimes we see something, sometime we done recognize it.

    On my device the undistracted Limbo R walks normally only 8 blocks in Limbo mode before R returns to normal walking mode. It is 8 blocks plus / minus half a block, or maybe a plus / minus a full block, depending on the way we count it. Sometimes I also saw longer walks, but this was in preview mode of the level builder which behaves differently after certain changes to the level (has been reported in the forum by members several times). But sometimes, with additional blocks around, the Limbo walk ended already after 4 blocks. Currently I see 3 walking distances in Limbo mode: 1 block & R returns (this is R after first passage of the ball motor), 4 blocks and R turns right, 8 blocks and R returns.


    Have you noticed something in the video? Limbo R's tilting angle varies a little and it seems to be most stable shortly before the Limbo walk ends. Is it as in real life, the Limbo king is the one who can go deepest? Don't think so. The tilting angle is the same for the 8 block and for the 4 block walk, in both cases B is leaned back by 62° from the uprigth position. The 1 block walk after first passage is done with 50° leaned back.
    1_Block.jpg
    4_Blocks.jpg
    8_Blocks.jpg

    Currently I have no idea to Limbo walk with R beyond the 8th block.
    But I had a fun time with checking out H and P. P behaved like R. H was a little headless in his walking behaviour as always.

    Limbo_Walk.png
     
    Last edited: Apr 19, 2019
    Ray Aznable, BangTe and Fendi like this.
  5. Fendi

    Fendi Famous Member

    Messages:
    54
    Levels:
    96
    Albums:
    4
    Likes Received:
    1,585
    Joined:
    Dec 24, 2018
    @Makaroni Yup, agree. I also think that using a ball motor for R (so R Limbo) with long tracks might be a bit risky, or at least the R moves becomes random or unexpected, such as the "Limbo Walk" demo level. But I think, the influence of the ball motor when R has Limbo, the possibility of stability is maintained. We may need to add a number of other elements around it so that the effect is maintained, or at least close to 100% stable. I think, like other players when build the levels, maybe he needs to set the design in such a way that the influence of the ball motor can run stably, like R can walk sideways, or so P can enter the hole. Yes, the ball motor mechanically moves freely, and the R round / R move is out of control, so I think the design effect is also very decisive. Just like other tricks (if Limbo R with a ball motor is said to be a trick), for example Autopilot for B, we might need to set certain elements above / below it so that when draggable is lifted for Autopilot, B runs stable and does not exit the area.
     
    Last edited by a moderator: Apr 20, 2019
    BangTe and Makaroni like this.
  6. Makaroni

    Makaroni Member

    Messages:
    24
    Levels:
    9
    Albums:
    1
    Likes Received:
    64
    Joined:
    Dec 30, 2018
    @Fendi Rahardian One idea might be to try out blocks as motor bearing with smaller physical size as the apparent one (see this thread). This could introduce some slight variations which influence in the end the stability of the Limbo R bots. I have not tried this out. But I tried out several blocks as pavements for R and found that stair blocks do a good job in increasing stability of Limbo R bots at their ball motor passage. In the demo level the Limbo bots walk (mostly) on blank ground (not on any pavement block) which might increase stability as well. The Limbo bots seem to not bother much on which block type they step, I was not able to influence their direction on their path by introducing pavement blocks in the path to which R bots usually react. But if there is a block in the direct way of their botic body then the Limbo mode comes to an end. But I find it difficult to start the Limbo mode properly again for an R bot that has lost the Limbo mode once before. Current question for me would be what to do to switch on and off the Limbo mode for R in a controlled and reproduceable way, again and again. The initial motionless standing of R at start of the level might be important two. How to nurse such R bots that returned from Limbo?


    Power_Limbo'ing.png
     
    Anomynous, BangTe and Źàiď like this.
  7. Makaroni

    Makaroni Member

    Messages:
    24
    Levels:
    9
    Albums:
    1
    Likes Received:
    64
    Joined:
    Dec 30, 2018
    Sometimes I just cannot search correctly. :unsure: The other day I searched this forum for search term 'Limbo' and received not much hits. :eek: Today I did it again and - voilá - hits over hits. :D I think, before I was not searching throught all content but only through some subset without noticing it. :p I'm getting the impression that in early forum times more discussion happened at the uploaded levels themselves and not in threads like this, any search needs to include the levels as well. In this special case, Limbo (the dance) is not Limbo (the light version of hell, which was abolished anyways 2007 by the Vatican); both Limbos are found by the search. Eventually I checked these levels tonight:
    • The Limbo Walk (24.07.2016, by @retrograde) which took several attemps for me to solve it since there are tricky moves which I did not knew before. But, yes, the solution involves Limbo walk by R caused by a ball motor.
    • Certainly the abolished Limbo (24.08.2016), not the Limbo walk
    • The Limbo (03.09.2016) with the unreal feeling. I solved this level but have not noticed any Limbo walk. Of course and for sure, spectacular jumps and moves. I checked the video game referenced therein and it looks like that abolished Limbo thing, not like the dance.
    • The Limbo demo (27.09.2016, by @Rohan Kumar) should be rather entitled 'Limbo Enforcer'. What a machine! Poor P. But definitely the Limbo walk.
    • Limbo and Limbo V.2 (both 06.04.2017, by @dDevilL99) are aiming both at the dance position for B, but I couldn't win these levels. Even at max. backward position poor B died through the zappers all the time.
    • Unfortunately it turned out that I'm not as Flexible (23.07.2017, by @meko) as required for this level. I'm seeing the path to the win but after about 50 trials I paused (to avoid 'stopped playing') this level. The bots were somehow instable, R often has fallen off the blocks by R-self, R often was hovering and vanished this way into nowhere, 2 times even both bots were in the right vertical arrangement and then just stopped any motion. Sometimes I really think that levels execution varies across devices with regard to accuracy of calculation and stability of moving elements in the diorama. Maybe I should get back an older mobile with the app installed, executing older levels on hardware from the same epoche. :sneaky: Hmm, probably true gamers are also not using any screen protector film. :oops: Bottom line: I just cannot tell if R is doing the backward Limbo in this level near the win block or some other kind of Limbo. But definitely the ball motor is there. ... Ok, I tried it, I got the app installed my old second hand S3 on Android 4.4 without protector film. Oh, I understand now, why older levels are often so small, there is no much place to tap on to on a mobile phone, says a surprised tablet player. :) But my hope was wrong, the level is not much more stable on the old device.
    • At the Limbo Machine (07.08.2018) I saw no ball motor and didn't want to invest another 50 trials tonight. The design of the level gave me the impression that it is this abolished kind of Limbo.
    • Most recent level is Limbo Crossing (15.12.2018, by @gmacpro). Luckily the caption offers a solution video showing P doing the Limbo after ball motor passage and walking this way across a fence element until Limbo is ended by running against a brick. It can be nicely seen that P walks a little off the 16x16 grid directly after Limbo mode (as in the video of message 4 above), but then is normalized while stepping via a stair down block onto a draggable. Could be that attaching the bot, that danced Limbo before, to a new coordinate system (that of the moving draggable) inside the level would normalize that bot again to be used properly in the rest of the level. Allways good to finish with an idea for the next day.
     
    Last edited: Apr 23, 2019
    BangTe and nGord like this.
  8. Makaroni

    Makaroni Member

    Messages:
    24
    Levels:
    9
    Albums:
    1
    Likes Received:
    64
    Joined:
    Dec 30, 2018
    I tried several block combinations to stabilize Limbo R but didn't succeed in creating an endless Limbo loop with entering and leaving the Limbo mode. :(

    Limbo mode ends in some kind of coma when Limbo R steps on blocks that move or could be moved. With some hard blow (either by hitting R with a draggable, or because Limbo R is running against some block, or because R falls down) it is possible to get Limbo R out of the coma. Then R could participate in the level again but R remains instable and it is pure(?) luck(?) to have another Limbo cycle.


    The_last_Limbo.png
     
    MekaSage, Ray Aznable and BangTe like this.
  9. Ray Aznable

    Ray Aznable Indonesian Language Helper

    Messages:
    171
    Levels:
    30
    Albums:
    4
    Likes Received:
    4,842
    Joined:
    Dec 2, 2019
    genius

    mad scientist.

    testing it over right away
    with any possibilities may occur.
     
    Last edited by a moderator: May 20, 2020
  10. Anomynous

    Anomynous Urdu/Hindi Language Helper

    Messages:
    84
    Levels:
    136
    Albums:
    29
    Likes Received:
    2,378
    Joined:
    Jan 26, 2020
    Such an interesting thread and demos.

    BTW @Makaroni u spent alot of time in observing a lot of behaviour so why you left making and playing?
     
  11. Anomynous

    Anomynous Urdu/Hindi Language Helper

    Messages:
    84
    Levels:
    136
    Albums:
    29
    Likes Received:
    2,378
    Joined:
    Jan 26, 2020

    Loved the ramp climbing one.
    It's little bit unstable in level bcz if we take r somewhere else to the ball sometimes this limbo effect don't word.

    But I'll find a way to use it perfectly ;)
     
    Ray Aznable likes this.
  12. Makaroni

    Makaroni Member

    Messages:
    24
    Levels:
    9
    Albums:
    1
    Likes Received:
    64
    Joined:
    Dec 30, 2018
    @Anomynous Thanks for asking! Exactly, in the end it was too much time, which I couldn't afford any longer.

    But the other day I opened again the app on my tablet and noticed much change. I could'n remember blue bots and stars, but both make perfect sense to me, increasing the depth of the play and moving the focus a little away from all the funny behaviours. I also noticed that many of these strange things don't work any longer, which is ok for me. I need to go through my old stuff and will delete things which would no longer work or will leave a deprecation notice. Without the old videos one would not believe what we did years back. o_O

    Deprecation notice: I tested the QRs in this thread and found that the cards do not work any longer as originally intended with the most recent app version (Android tablet). The older limbo levels listed above I have not tested. This thread is for the archive museum. :eek:

    ... ah, not more than 5 stars ... grid lines ... undo/redo ... :rolleyes:


    You gotta learn to dance before you learn to crawl - Meatloaf
     
    Anomynous, MekaSage and Labyrenight like this.
  13. Labyrenight

    Labyrenight Famous Member

    Messages:
    371
    Levels:
    184
    Albums:
    3
    Likes Received:
    2,210
    Joined:
    Jan 22, 2021
    I have a structure so the robot can climb up the wedge, but can't go down like in limbo. If you are willing, you can see it here Wedge Climbing. Maybe this can be further developed.
    There's also a limbo-inspired way, but I reversed it.
     
    MekaSage likes this.

Share This Page