TerraTech Community Build - Great Ball Contraption! [Finished]

ZeroGravitas

Breaker of Games
Jun 29, 2017
3,938
5,756
675
37
UK
www.youtube.com
Edit 2018-01-21: and here's the finished product!


▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬​

So, the devs have run some cool competitions, but how about a community collaboration?! :)

The concept: about 8-12 participating forum members will each build one small module to be integrated into a single 'Great Ball Contraption' - a continuously running type of Rube Goldberg machine. I made a 4 (and a half) module proof of concept build to demonstrate (save also attached, below):

Maybe take a quick flick through one of the several Lego GBC videos for inspiration too (e.g. below). In place of those little plastic balls, we have GSO regen bubbles (the best rollers I've found). They (or their components;)) will be moved through each novel mechanism to the next, in a big loop.

Don't worry about reading everything below before starting to post your ideas and preliminary builds! Just have a play and go for it, and we will work it out along the way!:)


End result: an R&D save file with all the modules set up together and in action, to be submitted to the appropriate weekly Community Core live stream.

Also, I will make and publish a video to showcase the whole thing and it's component parts, with full credit explicitly given to all builders and contributors. Other (contributing) TerraTech YouTubers are enthusiastically welcomed to take a copy of the save and showcase it in their own videos (to be released simultaneously or after my own, please).:)


Time frame:
About one month. More specifically:
  • 2 weeks of brainstorming and collaborative testing of ideas and mechanisms, ending on Friday the 27th [now completed on] Monday 30th October. Then...
  • 2 weeks of submitting finalised builds [Go for it now!] and integrating them into the save file (may be fiddly). [NOW ENDED] on Monday 13th November. The end build should be finalised for TerraTech version 0.7.8.0 (stable). That update should land somewhere during this time. (So blocks beyond this update to be included in submissions.)
  • 1 2 weeks, or so, [for me to construct, fix,] film and edit the result. Although this may well slip, depending on the final complexity of the build and my health is very unreliable too, unfortunately.
Other setbacks are fairly likely: my capabilities vary with my health, and game updates may interfere. (Although, hopefully the GSO regen bubble won't be redesigned to have square sides!)
 

Attachments

Last edited:

ZeroGravitas

Breaker of Games
Jun 29, 2017
3,938
5,756
675
37
UK
www.youtube.com
Module Specifications (AKA 'Rules'): These are a work in progress - thoughts, suggestions?

(1) Strictly 1 module entry per person. So make sure it's perfectly refined and awesome.

(2) Additional suggestions and tips can be made to other players, and may bag additional mentions.

(3) I'm hoping for 8-12 finalised entries, so block count and size per entry will be very restricted! Think elegant and minimalist! (No, seriously, still smaller than that!) Target 32 meters (i.e. 32 one blocks) length (about one big square in R&D). But we can stretch to 64m, max, if the module's block count is still very low and the mechanism really needs the space to show off it's stuff.

(4) Builds should be aesthetically pleasing, maybe decorative. But should not have more purely ornamental than functional blocks, since total block count will be such a hard limit.

(5) Avoid frame-rate dropping issues:
  • Tech-on-tech multi-tech action is, unfortunately a big strain on the game, although I'd still like to see a few simple mechanisms, please! But probably only 2-3 moving parts and as much anchored as is possible. (Physics is simplified for anchored techs.)
  • Also, some rotating anchored techs can cause particular problems, so be wary of bugs and test your entry carefully before submission!
(6) Modules must require no player input to function correctly. (Maybe enemy techs can be used to stimulate certain responses, for example.)

(7) Operation must be very reliable! Dropping 1 in 50 bubbles, passing through, might be considered acceptable, but the flow must not jam up altogether. And should easily run for 10-20 minutes without breaking.

(8) Re-load reliably, from a save file (while in action!).

(9) Be easily deployable (by me). So ideally, any multi-techs would auto-deploy with explosive bolts (as GKJ's did, here). [But functional reliability is more important and I can do some manual assembly if needed.;)] I might pass the save file around a little, if absolutely necessary. But simplicity will be essential, anyway, to avoid 'lag', and it's very likely that modules will need to be shuffled about and re-positioned to close the loop, debug, etc.

(10) Run at an average rate of at least 1 regen bubble every 4 seconds. Faster would be OK, as long as it's not a problem to be limited by the module(s) before.

Side note - 4 seconds is the time to drop a component from a conveyor with pacemaker set to Slow. I was initially thinking of ~1 ball per second (fabricator set to Turbo), but that would probably be too frantic and involve too many simultaneous balls in motion.


(11) Entry must specify it's acceptable input and output types: bubble blocks or scrapped components? Continuous stream of balls only, or will a large influx in one lump be OK? (E.g. with a hopper as a buffer, like my GeoCorp up-ramp.)

For height defaults: ground level, 10m, 20m and 40m 8m, 16m and 32meters. Output height (and type) need not be the same as input. In fact, modules should ideally be all about either raising or lowering the balls in an interesting fashion.:) So they can be paired up with each other and minimise overall block count. I may ask some people to adapt their outputs/inputs to fit with other entries, in the end-build. Or ask to adapt it myself, if needed.
GBC Default Inputs 1.png Snapshot of tech demonstrating the default input heights. Set filters and add regen bubbles.

(12) Entrants should be prepared to adapt and refine their builds to fit within the whole and ensure meeting the other rules.

(13) Be unique and distinct, so we get a diverse set of modules using as many different types of blocks and conveyance mechanisms as possible

(14) If you're not sure (or politely disagree) about any of this, then ask below!:)
 
Last edited:

ZeroGravitas

Breaker of Games
Jun 29, 2017
3,938
5,756
675
37
UK
www.youtube.com
Notes:
My proof of concept build (submitted to this week's Community Core) is a placeholder; I'm trying to work up a more detailed video of this build, for my channel. Full video in top post.

I have a fairly powerful, new gaming desktop, and saw no real signs of frame rate degradation, there. But I was startled at how quickly simple ideas got fairly large. Like the GeoCorp diagonal upwards conveyor (using resource chunks as pushing slats) - I think that is borderline too big.

If there's a lot of interest, we may need to be even more strict on size, and possibly run the machine with just a wave of 'balls' propagating along, rather than populating the entire track at once. Filming their progress as they go.



Ideas, Suggestions (to start the ball rolling;):p):

Don't be afraid to keep it simple!: it will be very useful to have straight forward, reliable ramps, bridges and lifters to feed between other, more complex and sensitive modules.

Resource relay:
scrap the balls, send the resource components on a wild ride! A convoluted (or linear) series of droppers, collectors and receivers. Maybe scatter the resources in a mess, before gathering them up.

Note: we can't have too many loose resource chunks on the ground, as lag-tastic, but they're much less of a problem on conveyors. Beware that collectors have a preferential order for resource type, too, so watch out for Celestite being left behind and accumulating.​

Transmutation: after scrapping, you could use the resources, mixed in with others, to fabricate different blocks to pass through your module. Just scrap and convert then back again by the end.

Counter/delay/chaos: make the behaviour as complicated as possible; have a 'state machine'. You can very easily make a binary (or trinary) counter with the alternator conveyors (as I showed in this thread).

Tipper: I meant to incorporate a simple 2-part multi-tech into my HE down-ramp, that would accumulate, say, 5 balls before it over-balanced and let them all roll out together. A couple of these mechanisms might form a (very simple) counter/clock (example vid), and be interesting to watch for a minute or so (in time lapse).

Train/truck/wagon: I really want to see a little build loading up a hopper on one side, then ferrying a whole batch across, dumping them and returning.:cool:

Tilted table: simple zig-zag channel, or pinball setup with automated shovels firing (only the tips collide, I think). Tilt could change, or pins could be added to impede the balls escaping through a hole.

Rotation: would definitely love to see a module (not all!) using the 'hover bug'. But automatic (horizontal) tech turning can also be achieved with AI on a rotating anchor, by pointing the primary cab at the sky.

A lifter fashioned like a Ferris wheel would be awesome, but, as a multi-tech, it would have to be small and constrained to 2 or 3 parts. Maybe more like a rotary digger/shovel.

Or simply use propellers as agitators, to stir, redistribute, randomise or flick the balls along (will their blades collide?).
Pusher: an articulated rotary piston arm, perhaps. With slow action that allows balls to accumulate before shoving them out en-mass (down a rap, or whatever).

Jump: any amount of time spent mid-air will gain bonus kudos and admiration. A ramp jump? Or controlled explosion (using an enemy tech to fire missiles continuously...).


Preliminary Builds Index:

@Legionite - "Xpressway Train Station" - (1) Resource carrying train (2) piston raisers (2.5) gently sloped exit rail ramp.

@QuackDuck - Scrap magnet (duck) crane.
 
Last edited:

ZeroGravitas

Breaker of Games
Jun 29, 2017
3,938
5,756
675
37
UK
www.youtube.com
Edit: excuse me, I've merged @JohnMatley's first posts, here, to tidy the thread for other readers:
All I see is Error
Where? What error, specifically? One of the videos...?
Video error, although I'm surprised at some sections where I thought for sure the balls would fall off. Are there any hidden tricks?
So the first video has a problem for you...? Or both OK now?

You can examine the build for yourself if you download my save from the submission (I'll add the tech snapshots here later, too). TerraTech physics works a little different to the real world; certain things that look hard just happen naturally.
Can't check it at the moment
 
Last edited:

Masakari

Well-Known Member
Nov 2, 2016
951
999
505
37
Those new bumpers work great for that track and didn't know the Venture grooves were physical. Thought those were just decoration/bump-mapping, or is it just due to gravity and the straight shots that keep them in place similar to the HE run section?

Regardless, GJ on this one. I don't know if I will add anything though as I find it hard to make just random things like this. Maybe my mind just don't work like that.
 
  • Like
Reactions: ZeroGravitas

Masakari

Well-Known Member
Nov 2, 2016
951
999
505
37
Just watched (without audio) the lego GBC video you posted......my god that must have cost a fortune even secondhand.

Loved the claws and track-based wave-rider sections the most. Seeing that makes me want even more MT specific parts like pistons, gears, motors/actuators, remote control blocks (allowing functions on multi-grids), and true circular shape(d) module/part blocks in varying sizes.

Got quite a few ideas I want to try out now (not sure how they will translate into TT though).
 
  • Like
Reactions: ZeroGravitas

ZeroGravitas

Breaker of Games
Jun 29, 2017
3,938
5,756
675
37
UK
www.youtube.com
I've re-organised the layout, above, between my first few posts. Plus added a whole bunch of suggestion ideas! As I've said, don't worry about reading everything before giving this a go. There's flexibility and time to hammer out details later.:)
 

ZeroGravitas

Breaker of Games
Jun 29, 2017
3,938
5,756
675
37
UK
www.youtube.com
Rolling race car bubbles on a track.
GSO regen bubbles, not the Venture regen bubble (I don't think they'll roll very well with protrusions...?)... Oh, the Venture shield bubble, you mean, is that perfectly spherical (too)?

[Edit: after some testing - no, neither venture bubbles are good enough, nor the GSO shield. But, the GSO wireless charger is just as good as the regen bubble! Despite it's wobbly appearance.]

the Venture grooves were physical. Thought those were just decoration/bump-mapping, or is it just due to gravity and the straight shots that keep them in place similar to the HE run section?
Not sure the colliders are that complex. The bubbles just seem to follow the gradient of the slope exactly; I had zero trouble with that impossible looking HE down-ramp - it just worked first time and has never dropped a ball.
Those new bumpers work great for that track
These, on the other hand, while looking perfect, where very fiddly to get to work reliably. I had to replace the outside corner pieces with HE corner brackets to stop occasional, random jams that blocked the helter-skelter.
I don't know if I will add anything though as I find it hard to make just random things like this. Maybe my mind just don't work like that.
You mean, it doesn't like to work to specifications? It rather just goes with what grabs your attention at the moment...? Well, just let this simmer and see if any little motivations crop up in the next few weeks.:)
 
Last edited:
  • Like
Reactions: Masakari

ZeroGravitas

Breaker of Games
Jun 29, 2017
3,938
5,756
675
37
UK
www.youtube.com
The article is too long! Can't read:oops: mention me?
Sorry. Basically: would you like to make one section of a build like in my video, but bigger? You and everyone else.:)

Your "Power transfer" build probably helped inspire me to start this project:
Each person's part will have to be smaller, though.



An Xpressway Train Station.

Just press X, and you'll be on your way!
Wow, that's looking awesome already!:D I was just editing in a suggestion about trains when you posted this...:) I'll give it a proper look in a bit, but be prepared for me to say it may be too big/laggy (however amazing!:cool:); the test will be if 8 of these in series will run OK without major lag.o_O
 
Last edited:

ZeroGravitas

Breaker of Games
Jun 29, 2017
3,938
5,756
675
37
UK
www.youtube.com
I'll give it a proper look in a bit
OK, @Legionite, so:

(a) That's amazing work for less than 24 hours, worthy of it's own showcase (might add a quick video, here, in a day or two). And it teaches me a bunch of stuff.:D
Conveyor based lift and car/train mechanisms.
Easily setting up (multi-tech) slanted/pivoted chutes.
(b) I would definitely count that as 2 (fairly large) modules, given the difficult limitations. So will likely ask you to chose one or other for a final submission.

(c) Obviously the reliability is poor, as is (but then this is only the blue-sky, brainstorming phase, plenty of time).

I think you could work on:
  • The smoothness and stability of the piston lifters. Support them with multiple actuator conveyors...? Different rod and axle parts...? Do wheels (or other things) work/help as runners? Possibly cut it down to 2 moving parts, with a solid mid-section (to reduce CPU burden and aid reliability of transfer. Make input hopper bigger.
  • Feeding the bubbles to the bumper bar chute without drops. Obviously.
  • Rail system. Maybe the conveyors can be incorporated underneath, maybe as the rails themself? Pushing resource chunks(s) inside the tech, such that they hold it down too (without the hover bug pusher)? Resource collectors currently throw chunks everywhere (might be difficult to perfect). And collection off the train needs to ensure celestial doesn't get left behind for the fabricator.
I also wonder if you think it would be feasible to have a train carry the solid bubbles themselves? Or is their weight guaranteed to through things off, or be too hard to load/unload?

By the way, I'm prepared to do a little bit of manual setup work, rather than have the entire tech pre-packed with just add "X", if it aids the reliability/elegance of a design.

I'm sure you have most of this in mind, better than I do, already. And again, great work so far!:)
 
Last edited:
  • Like
Reactions: Legionite

ZeroGravitas

Breaker of Games
Jun 29, 2017
3,938
5,756
675
37
UK
www.youtube.com
Can I contribute to this even if I don't have R&D?
Abso-flipping-lootly.:)

Using the current unstable version you should be able to do pretty much everything in Creative mode. The snapshots'll be just as good, right. The main deficiency will be lack of Pacemaker, currently. But a fabricator at normal speed will also take 4 seconds to make a regen bubble (I believe), so still perfect for the ball timing I think we should aim at.

Do we need some sort of universal connection?
It's a very good question. I think I've been way too vague about input/output format, so far. So something needs working up.

Easiest approach would have been to say that every module needs to output regen bubbles to (and collect bubbles from) the ground. But that would look boringly repetitive, and involve adding unnecessary adapter sections to front/back of modules. (And I really think we need to push to shave the block-count down to a minimum, wherever possible!)

Maybe we call that a "type-A" connection? Then type-B would be at, say at a single point 15 10 meters off the ground...? (e.g. delivered by a straight ramp/bumper rail.) Type-C would be resource chunks on the ground, etc...?

As with my demonstration build, the entire point of some modules was purely to raise the bubbles up in a stylish way, to feed stylish ramps. So maybe ascender or descender would be the main module types.:) I think we need to match up people's modules like this. Team work! Co-operation. ;):D
 
Last edited:

ZeroGravitas

Breaker of Games
Jun 29, 2017
3,938
5,756
675
37
UK
www.youtube.com
I would definitely count that as 2 (fairly large) modules,
Or 2 and a half, even (@Legionite); that output twin bumper rail is the start of a nice descender module. Someone (anyone!) could make a whole terrace of long down ramps, packed up into a single tech. Tippers too. Have the descent take a really long while. Then maybe have something like, when the bubble/ball reaches the bottom, it triggers a (mechanical) mechanism that releases all the rest of the accumulated balls in one big rush. Maybe down a steep ramp... With a jump at the end...:eek::D

I am a little wary about having a lot of non-anchored tech parts, though. My demo ran at great FPS, but pretty much everything is a solidly anchored tech, there...
 
Last edited: