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.

On realism: Stress, Strain, and Elastic Modulus

Discussion in 'Suggestions and Feedback' started by Victoriesx42, Dec 11, 2013.

Thread Status:
Not open for further replies.
This last post in this thread was made more than 31 days old.
  1. Victoriesx42 Trainee Engineer

    Messages:
    72
    In reality, ductile materials (metals) follow Hookes' law, where Stress = Elastic Modulus * Strain (amount of deformation).

    There are two types of tensile and compressive stresses: normal stress and bending stress.

    Force/Area is what is referred to as "Engineering Stress"

    where (Bending Moment * distance from the neutral axis) / 2nd moment of area is the bending stress on a structural member.

    If we arbitrarily apply elastic moduli to various materials, we could have a less math-intensive, more realistic game.

    More to come on shear stress, strain, and shear flow.
     
  2. Gentry Senior Engineer

    Messages:
    2,167
    Why did you post in the other 'stress' thread, then 6 minutes later make your own?

    It'll be better for everyone if we kept it all together.
     
  3. Victoriesx42 Trainee Engineer

    Messages:
    72
    I posted one weeks ago and nobody seemed to care.
     
  4. Gentry Senior Engineer

    Messages:
    2,167
    who cares if you're the one who posted it or not

    Well I guess this thread has sunk as well since we're off to such a good start.

    Though if you have good ideas, share them in that 9 page long thread on the subject.
     
  5. Victoriesx42 Trainee Engineer

    Messages:
    72
    Why didn't you tell the other guy to post his ideas in mine? Seems like a double standard if you ask me. Completely F'in Ridiculous if you ask me.
     
  6. Gentry Senior Engineer

    Messages:
    2,167
    I didn't know/care that yours existed. Nor does that matter. I'll post in the one that has generated the most discussion, and so should you.

    Its not a popularity contest.
     
  7. Praesumo Apprentice Engineer

    Messages:
    110
    Jesus Gentry. Every single post I've seen of yours lately is of you being a complete dick to someone.
     
  8. Sims_doc Junior Engineer

    Messages:
    876
    Oh, he isn't that bad once you get use to him i tend to enjoy reading his comments when he is replying to mine because he tends to be hardline towards the developers and will openly speak as such.

    but i do agree with you we need realism and of course the other thread doesn't offer that because it isn't specific on which way only that the game needs it.

    also, when i saw elastic modules? I've never heard of those before in my life.
     
  9. Gentry Senior Engineer

    Messages:
    2,167
    don't be hysterical
     
  10. BigDog Trainee Engineer

    Messages:
    90
    Actually I'm with Praesumo about the negativity but seriously both of you... shush. Trying to read over here keep your voices down.
     
  11. Ralirashi Apprentice Engineer

    Messages:
    306
    This is a game, not an engineering class.
    Kthxbye.
     
  12. Sims_doc Junior Engineer

    Messages:
    876
    why can't it be both?
     
  13. hrdrok Apprentice Engineer

    Messages:
    183
    Because nature's code is to much for a computer to handle. You want to introduce physical properties of metals. It would be cool if they did, but let's be honest, the strain on development and people's Pc FAR out weighs the benefits. Realism is cool, but I play games for the pure lack of true realism.
     
  14. Ralirashi Apprentice Engineer

    Messages:
    306
    Because this might be outside of a player's area of expertise and/or interest.
    Because I launch SE to build a ship and then have fun with it, not spend days calculating what stress will be applied to the 7th sloped block in the 5th row of the left wing of my ship. If I wanted to do that, I'd take a related class in college.
    There is such a thing as taking realism too far in a game. And this is a good example.
     
  15. damoran Junior Engineer

    Messages:
    608
    Well said. I second this.

    As I've stated before in an unrelated post, there is a sweet spot in between realistic and fun and so far SE has done well with that. If to heavy one way or the other you limit your audience, sales and ultimately support for the game.

    Remember, if this was completely realistic you'd have to ask yourself, why are we building with huge cubes in the first place? If you are calculating stresses and the like, you'd probably build with smaller parts of all different sizes, shapes and densities. Essentially, the entire game would need a revamp.

    My point being, the game is a creative sandbox environment with the primary goal of allowing players to tinker with different designs for space-craft and space stations and then "play" with those creations and have fun. As it stands there are only a few realistic aspects within the game, primarily, Newton's laws of motion and a rudimentary collision damage system.

    Other than that the game is about having fun with your creations.
     
  16. Victoriesx42 Trainee Engineer

    Messages:
    72
    My point is you can have the realism AND the fun. Replace the games damage system with the real formulas. The calculations in question are fairly simple, and I believe they would take less memory than the current system. Furthermore, you don't actually have to understand the concepts, as they would be behind the scenes.
     
  17. GrueX Trainee Engineer

    Messages:
    12
    Meanwhile there is a conversation to be developed over at the suggested thread...:)

    Structual Stress on Ship-Parts
    http://forums.keenswh.com/post/structural-stresses-6644237?trail=15

    Im interested in more realism too. I dont think it has to be an either/or proposition though. Right now Im thinking 'enhanced' struc damage from direct action on an object.
     
  18. Victoriesx42 Trainee Engineer

    Messages:
    72
    Condescending much? I'll discuss my ideas where I please.
     
  19. Sims_doc Junior Engineer

    Messages:
    876
    its not like there trying to convert you, personally i don't mind either of the topics because they have boarder the views.
     
  20. Siward Trainee Engineer

    Messages:
    13
    In no way would implimenting elastisity reduce the required number of calculations for the simulation. To calculate the applied stress to each block, a set of modal solutions would need to be formed every time the object changed shape. At the moment each body is only required to perform rigid body calculations (6dof), the modal solution you're asking for would require 6 more degrees of freedom per block unless risidual vectors were calculated at which point you're still only increasing the required calculations. This is the approach used in Kerbal Space Program. This approach would require a substantial increase in resources.
     
  21. Victoriesx42 Trainee Engineer

    Messages:
    72
    I disagree, if one were to apply the five unique stresses a differential element can experience, and treat each block as a differential element, there would only be 5 parameters on which to calculate the state of stress. One could even use superposition to further simplify the process.
     
  22. Siward Trainee Engineer

    Messages:
    13
    Ok I'm not sure if you followed... the problem has nothing to do with how you calculate the stress, the mesh is already created as the user builds the ship (blocks). How do you want to determine the strain be calculated. You need displacement. Anyways, you've already countered your own argument. it does not reduce the required resources to calculate stress, because you'll already need to just calculate strain in the system. The matrix inversion alone would be a pain in the ass for a static model, much less a dynamic one.

    personally i enjoy the damage model.
     
  23. Siward Trainee Engineer

    Messages:
    13
    that being said, i agree. stress calcs would be cool, but it would be resource intensive.
     
  24. Skeloton Master Engineer

    Messages:
    4,069
    And resources are in precious short supply...excuse the pun.
     
  25. Sims_doc Junior Engineer

    Messages:
    876
    Yeah, with people having high end gaming systems resources are really short.. but for the rest of us casuals it does present of problem that will be fixed when the game is optimized so I'm all for heavy realistic calculations because if its done now in alpha we won't have to worrier about it later..

    because after all this alpha!
     
  26. TechyBen Junior Engineer

    Messages:
    542
    You cannot "optimize" math. It's math heavy to do such a thing over so many blocks in all instances. There is only so much math your computer can do.
     
  27. Sims_doc Junior Engineer

    Messages:
    876
    Really? so you're trying to tell me that i can't find a better way to calculate this and save on resources. I don't want to be rude and it won't necessarily come out kind but you're new to the hole math and computer stuff right? because that happens all the time people figure out a better algorithm to calculate something and BAM! WOW COOL you've optimize it or made something that requires more resources depends on what you're doing.

    That's why mathematicians are heavily used in the computer science still.
     
  28. GrueX Trainee Engineer

    Messages:
    12
    Victoriesx42 - Sorry. Im new.
     
  29. Leonhardt Senior Engineer

    Messages:
    1,930
    Too much realism would kill this game.
    And this is exactly what this suggestion is... suggesting.
    I came here to build space ships with blocks in space.
    Not to play a hyper-realistic zero gravity physics simulator.

    Feel free to mod it all you like, but keep the computation-heavy calculations away from my computer.
    I'd like to be able to play this game when we get to beta, thank you.
     
  30. RedPhoenix Moderator

    Messages:
    817
    Thanks. Please keep on discussing there.
     
Thread Status:
Not open for further replies.
This last post in this thread was made more than 31 days old.