Disposed-open world mechanical survival 3D

Hey guys, thanks for peaking in. This will be a long thread of complete PC video game development process from start to finish(when game engine and some assets from store are excluded) and i will represent it with pictures and videos mainly and some descriptions. I am pushing this solo and development is started recently.

related website : https://devsgoingviral.wixsite.com/devsgoingviral?fbclid=IwAR1rO44ZBR9rCoRqOdg9AmX9l0yEDATqPnny9_z3ew2MN47l53wUBwCvTfo

Sneak peak into the project basics:
Disposed video game game is(any person view) blockbuster open world mechanical on unique way story driven survival where survival is re-defined, amplified, is infinite times more creative and better to play where repetition is hidden and basic survival and planet exploration elements expanded by a large margin.

What you do is exploration and survival with a goal.
Innovative(never used before) multi layer goals and reasoning for existing on the map and playing this world is featured inside this video game all wrapped tightly and dynamically with
an blockbuster storyline.

What Player do is : exploration, surviving, base management and establishment, tech research, story reveal, interactions with planet and lifeforms,
resource gathering, vehicle detailed management and customisation, maintaining yourself as mechanical assembly and maintaining yourself as biological form, terraforming planet until rescue comes(if you desire so), terrain and resource mapping and management, fighting, inventing, running for your life from the planet dynamics and from the lifeforms in the same time, saving lives.

I did some quick tests on searching the visuals by using some assets that i got from asset store and combining with my own assets such as 3D models and shaders and this is some aprox visuals i am aiming for:

And this is some modelling progress of one of the vehicles Player will be able to manage and drive:


This mecha spider will be scripted soon and it has procedural leg placements, doing him first so i can define the details about that subject after which i can slightly more define how my terrain will be.


1 Like

Here is some progress on modelling this mechanical walker . And from tomorrow texturing will start.

1 Like

Spider will be customisable from inside out : both the functionality and visuals.

1 Like

Making progress with modelling, and unwraps are starting, soon the texturing will follow , then shaders and related scripting and after that finally some mechanics scripting will happen where things will become way more interesting.

1 Like

Sounds very cool, the graphics look great.
Are you at a point you already have some gameplay?

I tried doing a system that can take a “tank” with 3+ legs and procedurally walk it, I was unsuccessful (but to be fair I didn’t give it all that much effort), waiting to see what you come up with.

1 Like

Thank you man! Not yet this PC attempt of the old mobile project is fresh started. Thats the half of the reason that gameplay showcase about why i am working this much on the spider so once i finish him i can showcase some mechanics and so i can do better planning of my terrain.

Ugh its gonna be nasty complex for my little brain but i will do best i can related to that procedural leg placement.
Features of the procedural leg placement would be :
Walking on the ground, walking in between two steep walls/side walking, incoming flying hazard leg catch prevention(counting firstly for grounded legs and deciding based on that), and few more neat leg decisions such as sharing one tight ground spot with two legs, understanding the slope+player wanted movement direction which decides what legs are pushing legs/i just want it to have visual sense, obviously mix of ground + wall push is possible. I did put on paper and its indeed complex with all the logic legs would have, but i cant wait to start scripting it and start crying.

Do you have some pitfalls to share hehe :smile: ?

1 Like

I wish, haha
I could barely wrap my head around doing this, I’m afraid if I try and give you tips I’ll be misleading you, but if you face some problems and ask for help around here on the forums i’ll be sure to give my best attempt at helping you out.
I had such a hard time with it I thought about throwing ML on it and hope it works, but ended up just leaving it aside because I have other stuff I actually know how to do, lol

Alright fair point about misleading hehe.
I also was thinking about ML since it looks as ideal solution since it could improvise if properly trained however
sadly i have none practical experience with ML and rushing to put some gameplay out there so that’s ditched :frowning: .

It seems you was going for all case scenario and everything to be correct possibly adjustable and modular hence way more intelligent and aware of its parts, and all that brings great complexity. However all i want for my spidey to do is those few named things and he needs to do them in amazing way which makes the system way more simplier, since there are only few things to do and interpolate in between few main logics, and if i manage to create something good i’l be glad to share my silly code.

This is quick prototyping i did, on this video i was using asset for procedural leg movement which is really robust/wide usage and yet it fails doing many things i need for my game and thats why i decided doing my own system.
Asset is called moveen and is way more advanced then my system will be, its just that for my needs my own system will look false 10x more advanced , which would not be the actual case,it will just look like most advanced thing ever for the player which is the only thing i care about.

1 Like

What does it fail at doing?
A lot of time a simple tailor made system is better than a do-it-all monster.
And smoke & mirrors are definitely the way to go 99% of the time, if the player can’t tell - do it!

1 Like

I could not agree more on that with you hehe!
To not sound wrong that asset i got is great by all means and is advanced and it does everything it claims to do, its really powerful it has bunch of settings and for anyone looking at bipedal quadruped or anything in between walkers on non interrupted ground surfaces i warmly recommend it and taking control over the legs is possible with code.

I simply want my own simplified system that i know from inside out within the code. And by doing so i can do things
such as : walking on interrupted surface, have easy control over each leg and having deep understanding how that will reflect further the whole walking system means i can do those fancy leg grabs(stop incoming hazards such as falling rocks, incoming objects, grabbing onto the surface instead of only stepping into it), and walking by “pushing” itself against two steep walls and hovering pretty much in air suspended like that.

I’m thinking on having same as modern CNC industrial machines where you have interchangeable parts and you swap the drilling head tools on the fly , so i need to do test rig in modelling software and to see if spiders current mechanical leg properties allows particular reach toward the body from where many tools can be swapped.
Basically i designed foot of the leg to be interchangeable with other parts(drill, weapon or any attachment i can come up with to benefit the gameplay), so in this case one single leg, or multiple is controlled by the player while other legs serve for walking. Basically i have in mind that such robotic leg/arm is very useful for many other operations and not just for walking, i think it would be nice touch and makes better gameplay while player with this particular asset.

Currently i am unwrapping the spider still, fixing geometry on the way, doing texture planning and similar.
I hope i finish all the unwraps today, there is just so many parts…

1 Like

I did test on the leg limits and they look great in terms of the reach.
However point on where leg is connected to the body as a hard point i think will not allow body tilt, or in other words it would need to be perfectly level no matter what which is a problem, so i will re-design just those attaching points most likely to be ball join.


That squash under the robot is needed because i forgot to mention spider should be able to walk even if there is thin beam of surface right under him and nothing else around.

1 Like

This is my simple logic i put together that i have desire to follow and fail hahaa.

1 Like

Mechanical mesh limits i am speaking about in previous message is this mesh that i am creating and this is bounding box/volume that says any spot inside the volume is a mechanically valid, without any scripting.
But to be fair after raycasts pick a point(asset mechanics correct point) i will check only to see if that point is indeed inside of this mesh volume(visually correct point).
One thing i am aware is each time i want to check i will need to read the mesh which is no good, so later when i start optimising whole code definitely i am planning on cashing the mesh verts into variables so it does not need to recalculate it every single time. I need to learn later how to catch those verts properly, how to store and what math to use since after catching all that verts i cant use any unity mesh class but i guess ill need to figure out what kind of math gives me the point inside of that volume.

This mesh represents leg limits/max foot reach in all possible directions which is what i need for my scanning systems to work properly.
If helpful for anyone thing that i noticed while mapping out the shape is it gonna be this banana shape in overall for sure for any similar leg design but exact shape that represents safe margin for your leg meshes to not intersect will greatly be defined by your art leg design choices . Intuitively i would expect that this first shape i got while stretching and compressing the leg while in straight line becomes same shape but smaller as i rotate the leg away from straight line and do the same process of mapping the max limits, but that is not the case, shape changes quite.
This is just a portion of the shape and since my leg ergonomic design allows for greater angles i will map out the whole range where shape will become more wide and changed each iteration as i approach the end of the how wide leg can reach.

1 Like

Started finally texturing first parts of the spider. And after this post i will take my time to finish all the texturing, shaders and some visual functionalities/tiny asset mechanics scripted(belt spinning, flames, sounds, basically will hook up basic Player input to my visuals…) so i can have fun while later developing actual walking system(right after that). Soon i will post some videos of all things working together.

1 Like

Finally can start scripting the visuals and sounds mainly. For the turbocharger belt i will simply scroll the texture and additionally there will be bones rigged to it and animated so i can simulate/fake the belt tension while picking up the speed and with sounds i hope to get that belt working nicely.

For the fancy volumetrics in my scene i am using free asset called “Aura” .

1 Like

I wanted this belt to feel correct so i took my time with half of my will and other half was some lerp line screwing me up hahaha . I hope that belt behaviour is neat.
Modest video for now(warning loud music):

1 Like

Tiny sounds progress of belt and 4 dynamos that are behind the engine. Now im going to script and make some particles and related sounds that will be coming from exhaust as also engine sound itself is missing for now. There is also air intake that will move/open once the engine needs more air, this intake will attract all nearby particles/and suck them in inside the engine which will influence exhaust smoke color as also quality of the air will have great impact on the how engine works.

1 Like

Tiny sounds progress. But today im starting to script procedural leg placement

1 Like

This is what i have so far with the spider in overall.

I am writing my own FK/IK system because of the way of how my spidey legs are made/design choices.
To achieve that robotic step and to not wiggle in all directions while stretching its leg i am constraining each leg joint to spin on single axis toward the target, targets are elbow and foot target, that would be the FK system but i want that “triangle” logic from IK systems on top of my fk system that will deal with foot and elbow positioning + leg mount which creates that imaginary triangle. But once i have that in place i will make it as optional because i want also to keep FK goodies where i have manual rotation control over the each individual leg part.

1 Like