Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - z26

Pages: [1] 2 3 ... 7
1
Screenshots & Videos / Re: Owl's builds
« on: April 12, 2017, 04:52:49 am »
The transmission is pretty cool.

2
Challenges / Re: Mini challenge: Extreme hill climb
« on: April 07, 2017, 04:14:32 am »
A while ago i've made a second bot for the loop, using mass driving tech that can stay upside down for an undefined amount of time.
I guess my easy entry of this is already ready

the slinky is a "walker" that can easily do this, doesn't deserve the extreme category imo

Oh and yeah, your track totally is a interplanetary roadway.

sorry for sucking the fun out of everything :p

(actually making a roadway between two or more planets might be cool, might give a sense of how smaller the map is than it looks)

3
Challenges / Re: Mini challenge: Extreme hill climb
« on: April 06, 2017, 12:02:33 am »
no hooks allowed?

what are the rules on continuum mass manipulation?

4
Introductions / Re: Petit Scarabee
« on: March 30, 2017, 03:09:20 am »
Our last interactions were in a skype chat where we came up with 2 pistons engines.

Not everything shared between us ultimately ends up in this forum.

5
Screenshots & Videos / Re: Wheeled mass driver
« on: February 24, 2017, 04:56:53 pm »
Yes, no.  the 3 continuui stages  are axially spaced apart by a small gap, which I've done to be sure they don't clash, but I don't know if that is truly necessary.  This can be seen visually by spawning the blueprint, the blocks are first aligned before they spread out.

Two types of accuracy errors are present: firing not in a straight line and firing with much less velocity than wanted.  The second type is more frustrating in my opinion.  It can happen even when there is only a single continuum left in the cannon
(i guess this happens because the block is interacting with the piston head, should I add a gap there?)

Rockslide is surprisingly easy to distract by shooting at it, it's just not made to filter out random parts flying around.  This helps my cannon a lot, but should be partly preventable by ignoring parts moving TOO fast.  A more complete solution would be more complex however.

This physical cannon has a good ammo parts to total parts ratio.  Maybe a more accurate 4 cannons with 6 continuui each would be possible, but the aiming and shooting system would be much bulkier. having a 1 or 2 x 6 system (one cannon on each end with a secondary weapon might me the most sensible, pragmatic option if the goal is to make an acl bot, but making something with only a physical cannon would be more impressive.

Normal segways need to lean to go forward/backwards.  The jets however apply a force directly to the center of mass, while the wheels only manage the vertical aiming angle (balance) + turning.

I've experimented with a complex camera system to avoid wasting shots, but that system is cpu intensive and the bot is very sensitive to performance issues (ruins the cannon accuracy)  Make sure you don't have a lot of code loaded in the vp screen or the bot will suffer

6
Screenshots & Videos / Re: Wheeled mass driver
« on: February 24, 2017, 08:19:49 am »
Thanks.

blueprint https://www.dropbox.com/s/b0p14s1i4mdegln/bp_dicycle31?dl=0

If anything confuses you feel free to ask me about it.

7
Screenshots & Videos / Wheeled mass driver
« on: February 23, 2017, 09:45:08 pm »
Hi,

I've made this bot a while ago and its unfinished, but with me being busy + new rawbots coming eventually (will turn it obsolete probably) I think its better that I share it right away.  If anyone wants the blueprint feel free to request it (it also contains a complicated vision system that probably impact the game's performance, but that is easy to remove)


8
Development / Re: Reboot
« on: February 03, 2017, 06:05:47 pm »
The bendy joints ironically seems to help in that video, working as a suspension.  The wheels seems quite grippy, but bouncy

Oh, and "custom wheel physics" got me curious.  AFAIK most games don't actually need to simulate rotating cylinders and so don't bother doing it, so I'm curious about Rawbots' approach.

Multithreaded physics is good to hear, dunno how it works internally however. Have parts distributed between threads, or instead something more complex?

9
Screenshots & Videos / Re: Random Screenshots
« on: February 02, 2017, 10:36:19 pm »
It feels strange to see a rawbot inside a realistic world.

Oh and the central continuum's edges looks straight, so probably not the camera.

10
Challenges / Re: Planetary ACL
« on: December 07, 2016, 10:24:03 pm »
What kind of planet size do you have in mind?  With planets that are too big this will eithier turn into a missile contest or a game of hide and seek, depending on how much ground cover there is.  With planets that are too small, being pushed out of the planet will be pretty easy (overlapping two planets for stronger grav might help with that)  So I guess there is a certain range of sizes that offer the best experience.

11
Haha, playing fetch with an hypercube, thats clever.

12
I cannot see the video in the last post, "This video is private"

Btw i have no solution to your problem, but you inspired me to give another go at a failed project of mine.  Thanks for that :p

13
General Discussion / Re: Progress on Navigation ?
« on: December 02, 2016, 06:09:38 pm »
I've made this for absolute coordinates.  As long as the reference continuums are level and that were talking about a blueshift style map it works quite well.

https://www.dropbox.com/s/xx6469kihdx5ebz/bp_absolutexyz?dl=0

14
General Discussion / Re: Offering an idea
« on: December 02, 2016, 04:25:16 pm »
With glovepie i managed to send the y position of my mouse as an analog input to a piston.  I had to use a workaround: i encoded the value in binary using one key for each bit, then decoded it in rawbots.  It works, but using osc instead would probably be preferable.

Also, I had something similar in mind (outputting a scan of the environment)  I don't know if osc supported outputs in rawbots.  For inputs you simply had to write something special in the "positive/negative" keyboard hex,  but how would you output something?

If someone's curious heres the code.  replace "1050" by the vertical resolution of your monitor and on rawbots side, multiply the output of the
"5" key by 32, 4 by 16, 3 by 8, 2 by 4, 1 by 2 and 0 by one.  Then add them together and you have an analog value going from 0 to 64.

Spoiler: show

var.positionx = mouse.CursorPosY / 1050 * 64
var.positionx = var.positionx - (var.positionx mod 1)
var.positionx = 64 - var.positionx


if var.positionx < 32 then key.5 = false
if var.positionx >= 32 then
var.positionx -= 32
key.5 = true
end

if var.positionx < 16 then key.4 = false
if var.positionx >= 16 then
var.positionx -= 16
key.4 = true
end

if var.positionx < 8 then key.3 = false
if var.positionx >= 8 then
var.positionx -= 8
key.3 = true
end

if var.positionx < 4 then key.2 = false
if var.positionx >= 4 then
var.positionx -= 4
key.2 = true
end

if var.positionx < 2 then key.1 = false
if var.positionx >= 2 then
var.positionx -= 2
key.1 = true
end

if var.positionx < 1 then key.0 = false
if var.positionx >= 1 then
var.positionx -= 1
key.0 = true
end

15
General Discussion / Re: Offering an idea
« on: December 02, 2016, 02:03:39 pm »
Also, we do have an archive to the old forum, but its quite incomplete.  I discuss this subject there http://forum.rawbots.club/index.php?topic=3.msg219#msg219

https://glovepie.en.softonic.com/
I'm currently investigating Glovepie, an old program meant to control games with stuff like wiimotes an virtual reality gloves.   it's a bit like hot2key, but seems to be much more programmation oriented.  I've managed to shoot plasma shots by middle clicking so far.

Pages: [1] 2 3 ... 7