Welcome to Keen Software House Forums! Log in or Sign up to interact with the KSH community.
  1. You are currently browsing our forum as a guest. Create your own forum account to access all forum functionality.

Load bearing and Structural/Civil engineering/building

Discussion in 'General' started by wuubb, Mar 7, 2016.

Thread Status:
This last post in this thread was made more than 31 days old.
  1. wuubb Apprentice Engineer

    Messages:
    114
    One thing that recently popped into my head was load bearing. This is arguably one of the if not the most important aspects of engineering things in the real world, especially buildings, bridges, tunnels, etc. Right now, on planets, I can make a bridge that is a 1 block thick sheet and drive a 1 billion ton truck over it and be fine. Same thing for building supports. Even if the structure is a "station", IMO there should still be shearing and loading forces involved to keep SE from being like Minecraft where people build floating cities and stuff and ground it back to reality a bit more.

    Just an implementation idea I had for bridges: the more you anchor the bridge by putting blocks into the ground, the more stable it is. Maybe we could implement concrete as a standard vanilla voxel material that adds strength as well? Drill into rock, place blocks in as anchors, and seal it up with concrete and then build your bridge/building (foundations or supports on either side of the bridge).

    Discuss
     
    • Like Like x 1
    • Agree Agree x 1
  2. Ronin1973 Master Engineer

    Messages:
    4,849
    Load bearing would be nice, but i think it's not practical for the game. That means a constant calculation on every block at every moment in relation to its tension and tensile strength in relation to every block around it.

    That'll kill your game faster than 10,000 jihad dogs that haven't eaten for a week.
     
    • Agree Agree x 4
  3. BlackUmbrellas Senior Engineer

    Messages:
    2,818
    This is basically just structural integrity, which is generally considered nonviable in Space Engineers because it'd cripple a lot of ship designs and encourage flying bricks.
     
    • Agree Agree x 2
  4. wuubb Apprentice Engineer

    Messages:
    114
    Maybe just apply it to building things on planets? Static objects like bridges and buildings?

    Wouldn't you only need to calculate for when you add/remove blocks (changing center of mass and such) and when there's collisions?
     
  5. DDP-158 Master Engineer

    Messages:
    3,748
    Both of the above are the reason we won't be getting SI like ME has. In ME the buildings are static and processing SI is easy. But when you apply that to a ship that is moving forward, rotating or turning, then calculations are constantly being made to see if all points on the ship can handle these forces. It would just be too much for most systems to even handle. This was all explained by the devs in one of the recent twitch feeds.
     
  6. BlackUmbrellas Senior Engineer

    Messages:
    2,818
    I imagine you could limit structural integrity to applying in a gravity field, but that feels... sort of selective. If you're making structures deal with load distribution and shear forces, why wouldn't that apply to your space stations, or your ships? Why does a 1-block pillar not support a bridge, but is fine to support a thruster? Why would crashing a ship into a tall structure on a planet have the potential to tear the whole thing down, but not crashing into a space station anchored to its asteroid by stilts?

    I mean, I understand where you're coming from, but ultimately voxel games are never going to be very good at encouraging realistic structures without severely curtailing ones ability to implement designs. Structural integrity implies maximum viable room sizes and/or awful ugly messes of support struts, because you can't create proper "load bearing beams" when those are constructed from the same blocks they'd be supporting.
     
  7. MechanizedIT Apprentice Engineer

    Messages:
    354
    Structural Integrity could and should be implemented in SE. Of course like ME, it would be an option, so no need to worry yourselves about your nonsensical ship designs.
     
    • Disagree Disagree x 4
  8. DDP-158 Master Engineer

    Messages:
    3,748
    If you ignore the whole need to have a high end pc then sure, it should.
     
    • Agree Agree x 1
    • Funny Funny x 1
  9. BlackUmbrellas Senior Engineer

    Messages:
    2,818
    Again, one of the problems is that structural integrity prevents scaling, or at least makes it significantly harder.

    A ship doesn't even have to be "nonsensical" to suffer.

    Say you make a light-armour hanger for a handful of fightercraft. With structural integrity, you need to brace the roof of the hanger; for the sake of argument, let's imagine that heavy armour can serve this purpose because its tougher and has more load points or whatever. You want to make the hanger twice as large; suddenly, your roof is crisscrossed by tons of heavy armour blocks. Say you make it bigger. The heavy armour can't support itself anymore, and your roof can't support its own weight.

    Realistically, you'd be able to install cross-beams. But with structural integrity and a block-based system, you can't, because the beam is just another collection of blocks, and their weight distribution cannot scale as rapidly as one would expect of real designs.
     
    • Agree Agree x 2
  10. tankmayvin Senior Engineer

    Messages:
    2,863
    At a certain point it stops becoming a game and turns into real engineering as well. You need to balance deep design with things actually being fun and game-like.

    If you want to "engineer" things, there are tons of real life projects, or even CAD based virtual projects. At a certain point a game just cannot fulfill that sort of experience.
     
    • Like Like x 1
  11. tankmayvin Senior Engineer

    Messages:
    2,863
    Combat simspeed would easily drop to <0.1 with structural integrity. Which basically means no one would play with it turned on.

    Also, things like merge blocks and connectors can apply INSANE forces on grids, with structural integrity they could literally rip ships apart.
     
    • Agree Agree x 2
  12. Ronin1973 Master Engineer

    Messages:
    4,849

    Same thing. I understand what you're asking for. I think it would be great to have more careful consideration when building things. Whip your fragile ship around too fast and tear off the front end because it's only secured by one light armor block.

    However, it's just not feasible in a game that is already CPU intensive without it. The cost wouldn't equate to the benefit. I'd rather suspend belief about a structure that would never stand in the real world or a spaceship that would torque itself in half than to have those things and have a game that was unplayable for most people.

    Both the major graphics card makers have schemes for running physics calculations on the parallel processors in their graphics cards. Perhaps it would be possible with a high end graphics card or two chained together. But that's a $1000-$2000 investment for just the cards. Most people wouldn't go for it so it would be only a small number of people who could enjoy that feature with a high cost in labor to develop it. The ROI wouldn't be enough.
     
  13. Nacon Junior Engineer

    Messages:
    591
    Keen did finally drop 32bit and DX9 in favor of DX11(12?) and 64bit though. So yeah, why not? :)
     
    • Funny Funny x 1
  14. wuubb Apprentice Engineer

    Messages:
    114
    Oh well. I thought that SE already did some of this with pistons and rotors, cause one time I made a drill on a planet and had like 10 pistons attached and they broke and the drill fell, which is why I thought maybe the idea could be applied to other static objects.

    I guess for now I will just keep building my stuff to look real with supports and foundations and such.
     
    • Agree Agree x 1
  15. DDP-158 Master Engineer

    Messages:
    3,748
    You want the people with decent range pc's to end up in the potato crowd?
     
    • Funny Funny x 2
  16. Nacon Junior Engineer

    Messages:
    591
    Potato are pretty decent anyhow. :)
     
    • Disagree Disagree x 1
  17. Ronin1973 Master Engineer

    Messages:
    4,849

    Yes. But that's checking one block against the mass and physics of one grid. It's one point, not every block.
     
  18. StuffYouFear Apprentice Engineer

    Messages:
    416
    This might have been fesable before planets were introduced. I dont see this happening within the next year as honestly the potato crowds average computer would need to catch up to whats now consitered high end, and what they think is high end is really just middle of the road. My PC is fairly middle of the pack, and is still above the price range of what alot of people are willing to spend.($800-1200 w/o monitor)
     
    • Agree Agree x 1
    • Disagree Disagree x 1
  19. VanGoghComplex Apprentice Engineer

    Messages:
    246
    I used to advocate for structural integrity, but then I realized something. In ME, when a block bears too much force, it crumbles. That makes sense for ME, but it really doesn't make sense for SE.

    You'll recall the intro video with Red head-butting Blue in its long spindly swan neck on the way to the planet? Those ships are made of steel, not rocks. The result there, as it should be with SI, would be bending, not sudden crumbling away to nothing.

    I'm pretty sure it's beyond Keen's ability to make grids themselves deformable in this engine. If I believed they could do it, I'd be all for it, but since the most likely thing is a copy-paste of ME's SI mechanic with suddenly-dissolving blocks and all, I hope this never comes to pass.

    Though, watching a bunch of people cry about their designs that they built in a weekly-changing early access game being obsoleted would be comedic gold.
     
    • Informative Informative x 1
  20. Frigidman Apprentice Engineer

    Messages:
    306
    Even though I enjoy it in ME, I would not enjoy it in SE.

    Reason being is SE "to me" is science fiction, and everything I create is not "realistic". I would not like for realistic structural integrity to be apart of the field, preventing me from making wondrous space based nonsensical creations and ships. For me, realism is best left for education tools, and real life. When I play a game, I want entertainment and freedom to do what I could not do in real life.

    PS: I've nothing against those who do want ultra-realism in games.
     
    • Agree Agree x 1
  21. VanGoghComplex Apprentice Engineer

    Messages:
    246
    I can't go to space in real life, so while I feel similarly about the purpose of games, I must respectfully disagree with you sir. =)

    I don't dislike sci-fi games at all. I just found KSH's original approach (feasible tech based on current tech) to be refreshing, in a genre awash with Star Trek stuff.
     
    • Like Like x 2
  22. Einharjar Junior Engineer

    Messages:
    522
    As Freeman Dyson quotes readily on SE's own loading screens: "Aviation is the branch of Engineering that is the least forgiving of mistakes". One of those easily made mistakes is Structural Integrity.
    However, I do agree that constantly calculating the tensile strengths and ductility of your structural components and systems would bog the game to a crawl. I mean, my old i7 dinosaur is already slogging on Planets at the 25 to 15 FPS range. Couldn't imagine plugging SI into it.

    I do want it though, badly. I'll be upgrading my rig soon because I know that's the only way to enjoy the game more. I hate seeing ships blown nearly in half though and still fly along just fine because one shrivel of light armor is holding the two halves together - plus some how delivering full power to the engines and digital control signals and all sorts of other unbelievable feats >.>
    I'd give up a little sim speed to see that go away and have ships and structures dynamically fall apart under stresses. I so would.
     
  23. tankmayvin Senior Engineer

    Messages:
    2,863
    I don't really want to do FEM calculations for SE either.....
     
Thread Status:
This last post in this thread was made more than 31 days old.