Jump to content

Heliomance

Alpha Tester
  • Posts

    19
  • Joined

  • Last visited

Posts posted by Heliomance

  1. No, but an interesting thing to discuss when considering a "world" where our reference frame is gone, many of our secondary units are derived or use an Earth based reference, and it's many months till we actually play anything. (And while people insist on going off topic from the "agree on a date format" proposal ;). Oddly couldn't get people interested in this side of things when I posted on SI units more generally. Go figure :P )

     

    (I actually don't think Newtons are an issue. The kilogram is essentially an arbitary amount of mass, so a Newton can still be the force required to accelerate the agreed "thing" to 1ms-2. Not saying that it would not be messy or complicated mind you...)

    All the SI units are essentially arbitrary. The thing is, though, they're arbitrary in such a way that they all interact nicely. One Newton is the force needed to accelerate one kilogram by 1ms-2. One Joule is the energy required to exert a force of one Newton on an object over the space of one metre, and is also the energy dissipated as heat when a current of one ampere passes through a resistance of one ohm for one second. One Pascal is the pressure generated by a force of one Newton exerted over an area of one square metre.

     

    If you mess with any of the base values, you mess with every single one of the derived values that interacts with that thing in any way. And as they're all interconnected, you essentially end up having to redefine everything.

  2. Almost forgot - the prototype kilograms mass would not change, but it would not weigh "1.0 kg" by our instruments on Alioth.

     

    (We really should move this part of the discussion to a different thread, but the immediately obvious solution (in-game) would be to ensure consistency between the master measures (by making sure they all measure the mass of the prototype Earth kilogram exactly the same) and then make, probably also from an iridium alloy, a new Alioth prototype that equals "1.0 kg" when measured.

     

    If the measured mass of the kilo is greater than 1.0kg on Alioth, we get our handy dandy AI and Arkship tech to shave the prototype down to size.)

    That's because we don't actually have any instruments capable of measuring mass directly. We have instruments capable of measuring force, and of measuring acceleration. So the prototype would still mass 1kg, it just wouldn't exert a force of 9.81N on Alioth.

     

    This means that shaving the prototype down and redefining the kilogram would cause utter havoc all across the board in derived values. If you redefined the kilogram, then the Newton would no longer be the force required to accelerate 1kg by 1ms-2.

     

    Redefining SI units is... not a good idea.

  3. I'm far from an expert on this kind of stuff.  And i only read the first few posts because it looked like the rest of the thread was just one person derailing.

     

    But, i would guess that the game is going to have to look at a construct and calculate how it will perform.  Taking into account which elements are used and how they are positioned, and what voxel materials are used and the weight and maybe even the shape they are in.

     

    How else would the game know how the construct should perform?

     

    I don't see any reason why that information couldn't be translated into some type of Stats or performance graph.

     

    What the physics engine knows:

     

    - The ship's velocity at the current moment

    - The ship's position at the current moment

    - The force being exerted by each of the ship's thrusters at the current moment

    - The mass of the ship

    - Gravity

    - Potentially air resistance if it exists

    - How long it's been since the last update tick

     

    What the physics engine calculates every tick:

     

    - The net force provided by all thrusters at the current moment

    - (If implemented) The resistive force of air resistance at the current moment, based on the ship's velocity

    - The net acceleration of the ship this tick, calculated from the force of the thrusters, the mass of the ship, the value of air resistance, and gravity

    - The position of the ship next tick, from its current position, velocity, acceleration, and time since the last tick

    - The velocity of the ship next tick, from its current velocity, acceleration, and time since the last tick

     

    What the physics engine does not know:

     

    - How to fly the ship

    - The optimal configuration to fire the thrusters in for maximum acceleration

    - The turning circle of the ship

    - The ship's top speed (unless there's a hardcoded limit)

    - The ship's maximum acceleration

    - Anything that happens more than one tick in the future

  4. I think the alpha team members will have a huge advantage over other players after general release because of the game skills we'll be developing, the familiarity with the construction and coding techniques and the relationships we'll already have in place.

    I wouldn't have thought they'd have that much of an advantage over beta players tbh. I imagine the beta period will be long enough to build all those skills up from scratch before live happens. Alpha and beta players will have that advantage over non-betas, though.

  5. What tools are going to be available to orgs for enforcing their rules? The idea of this game is emergent gameplay, of building a world and civilisation. Civilisation requires laws, and it requires the ability to enforce those laws. So how is that going to happen?

     

    In modern society, the primary means of punishing lawbreakers is incarceration. In a game, that's not really an option - if you lock a character in jail, that's destroying the player's ability to have fun in the game, and will likely just result in them stopping playing - losing a customer. That's bad. So, jail is out.

     

    Most forms of punishment available in a game are either trivial (execution doesn't mean much if you just respawn) or equally harmful to the game. But without some way of punishing those who transgress the rules, society doesn't function.

     

    How do you enforce law and order in a world without consequences?

  6. Yes, people are absolutely want to know what they're buying before they buy it. I'm not disputing that at all. But a simple list of statistics isn't going to work.

     

    The two main suggestions that have come up that are workable are a VR environment to test-fly a ship in before you buy it, and the emergent gameplay of trusted reviewers. Both have pros and cons, as expected. 

     

    A try-before-you-buy VR environment would give you a very good idea of what you're paying for, whether you like how it handles, and so forth. On the other hand, it means you don't really have a clue what you're getting into before you buy, requires more work for NQ, and doesn't really allow any sort of objective ranking.

     

    Trusted reviewers (presumably doing Youtube videos or something) would allow for somewhat more of an actual ratings system, and be able to give you a far better idea of what you're buying in advance, letting you actually plan to buy a specific type of ship. On the down side, they're only realistically going to be able to review mass-produced ships, it's not going to be worth their time to review small runs and unique models.

     

    That said, that's realistic, to be honest. I suspect that what we'll see is more along the reviewer lines, as I'd be surprised if we had VR test functionality at launch. That would encourage mass production and a certain amount of standardisation, as in real life. If you want a custom ship, you'd then have to either build it yourself or work closely with a shipbuilder, paying extra money for the privilege of getting exactly what you want. There, your guarantee of quality would be the shipbuilder staking their reputation on your satisfaction - if you're not happy, you spread that fact around, and the shipbuilder doesn't get any more commissions.

     

    We might even see an organisation of high-end custom ship builders spring up, the equivalent of Bugatti, Rolls-Royce, and the like, who specialise in one-off custom orders for very rich customers. Owning a Quasar-brand ship (for example) might become a status symbol, a sign of wealth and status. That could be very cool.

     

    One way or another, the community will work out a means of making sure that ships can be bought and sold with some understanding of what you're getting. It won't be a simple list of statistics on the market page, though. I suspect it's more likely to be a manufacturer's description which is trusted to be honest, because the manufacturer lives by their reputation.

     

    Hell, the game is all about emergent gameplay and player-driven politics. Maybe whatever nation-orgs become dominant will institute an Office of Trading Standards that you can complain to if a seller rips you off, and they can force compensation.

  7. I don't see why it would be difficult to have a list of at least basic default stats once you construct a ship. Like you build a ship, them you run some kind of scanning function and it outputs a short list of relevant stats. Could be based on the amount of voxel space used, materials used, what modules you used, complexity and type of scripts used.

     

    We already know there's a system in place to determine how a ship works based on how we build it, so having some kind of function to determine some stats that buyerss would find relevant doesn't seem like much of a stretch to me. Its sandbox game after all and tools like that are the Devs job to give us, IMO.

    So you have 5 engines. If they're pointing in different directions, adding the total thrust together doesn't tell you anything useful. But there could be perfectly valid reasons to build it like that - well-scripted, it might use different engines in response to different commands to make it more maneuverable.

     

    It may not seem like much of a stretch to you, but it's actually not possible. Not just difficult, not possible.

  8. Heliomance this forum is for people to discuss thoughts ask questions about the game we are all excited about. Specifically on this topic is to propose people's ideas. All you have done since I made my proposal is shoot me down, tell me it will never work, it can't be done and that basically Imply that I am an idiot. Other people have commented positively and added some realy good ideas too. If your going to go out of your way to make me and my ideas look stupid then please don't comment Unless it's constructive and positive like other people do

    No, I told you that it wouldn't work, I explained exactly why it wouldn't work, and you continued to insist that you wouldn't believe it until someone from NQ weighed in. Posting the video might have been a touch on the mean side, and for that I apologise, but I was getting frustrated at the fact that you seemed to be either ignoring or not understanding what I was saying.There have been plenty of workable ideas posted here, yes, and on the whole the thread has been valuable. but I do know what I'm talking about, you were dismissing what I was saying, and I got irritated and snarky.

     

    Anyone can get something wrong, and that's fine. I'm certainly not immune. I don't think less of someone for incomplete knowledge or for making a mistake - different people know different things. When someone continues to insist that they are right and ignores people trying to improve their knowledge, that's when I get sarcastic.

  9. I know that's what you meant but I was after something set in stone that can't be exploited. Just thought it would make for some important game play for Somone that wants to design an mass produce ships ect. I would like to hear from NQ themselves tho to see if it's possible and if not I guess I'll just make my ships from sticks and wood and people will buy them without knowing how bad they are.

  10. That's the kind of stats I want but I want them to be real stats as the ship will perform when in flight. Now I'm no computer programmer but I'm fairly sure the game needs to calculate all these stats like thrust over mass for example before being able to execute the action. I'm sure it can be done and should be easy. Maybe Somone with knowledge with programming physics in sims can correct me if I'm wrong

    This is your problem. If you don't understand a field, you cannot say whether a given thing will be easy or hard.

     

    A simulation is essentially "try it and find out". With the building tools as we currently understand them, there are functionally infinite possible ship designs. A computer is not going to understand how to pilot a user-built ship. It's not even going to understand the concept of "forward". What the computer knows is "This signal means to fire this set of engines. That produces thrust in this direction. Taking into account the weight of the ship, and gravity, that will result in a net acceleration of X in Y direction."

     

    If you mount the engines sideways on your ship, pressing forwards will result in your ship going sideways. The computer doesn't know that, before you press the button. It works it out as it goes along. How, then, can it predict the 1-1000 time if it doesn't know in advance which direction your ship is going to move in?

     

    You can build your ship in any configuration. Some of them will work, some of them won't. You can then program your ship to respond in any way to particular button presses. If you re-code it so that the normal "forward" button actually fires the weapons, and thrust is achieved by pressing a different button entirely, how is the computer supposed to know how manoeuvrable your ship is? It's not. It can't.

     

    This is not a thing that is possible.

  11. The tech tree will definitely be hidden, if it exists at all. The research is intended to be infinite with expansions adding a lot of hidden content. But technology research will have no defined path, it would be more like how technology works in Stellaris; however, more obscure and difficult to track.

     

    For instance, when they add warpgates to the game, no one will know how to build them until the research system actually finds out how. Then the massive amount of reasources will have to be gathered, and finally the warpgate drone will take weeks to get to the next star system.

    Fantastic! This is pretty much exactly what I was hoping for.

     

    That said, not sure how they're going to stop the alpha/beta testers learning the tech tree and just rushing it at go-live.

  12. Do we have any idea how the tech tree, research and development is going to work? For ages, I've wanted a game with a hidden tech tree, paralleling real life, where you don't know if something's possible until you make it work, and DU could be the game that actually implements it. Obviously the mystery wouldn't last long, because the possible research would rapidly get documented by the player base, but that initial exploration of what's possible and what's a dead end could be fascinating to be part of.

  13. I think this is a decent idea, but the main issue I see is that it is decently complex but not necessary enough to get development attention for quite a little while. Long enough, that what would be in the game until then as a "placeholder" (assuming this was a planned feature) would be well-established and used uniformly, and to change such a fundamental concept at that point would piss off a lot of people. Perhaps as an alternative to components later on, still keeping the originals in the game but allowing the new ones to be more efficient for the same cost. But just to restate, I think if this wasn't present in the initial release, it would be difficult to implement at all.

    Yeah, this is the major problem with it. Something like this, while amazing, would have to be in the game from launch to work well, and it simply isn't going to be. Which is a shame, because it would massively increase the value of R&D, which seems like a really good thing.

  14. It would be best if we were able to engineer components out of subcomponents each of which have particular properties, methods and interfaces. For example, a rocket engine could consist of a fuel tank, fuel pumps, high pressure piping, fuel injectors, a combustion chamber, an afterburner assembly, a nozzle and control actuators. Without each of these sub components, you can't build an engine. But if you developed a superior fuel injector, you'd have a superior rocket engine, perhaps one which is more fuel efficient, while a superior nozzle could improve manuverability, a superior combustion chamber could improve acceleration, superior tanks and pumps could increase the pressure at which you store fuel and that increases range, and superior piping could reduce weight. That way you'd have more than the Mark 1, Mark 2 and Mark 3 engines. You could have a company that makes rocket engines that are lightweight and high maneuverability for combat craft while another company makes engines that are fuel efficient and reliable for long haul cargo ships. The new Zykos-V fighter is equipped with a Motokrafwerks F-300 rocket motor with the highest vector-thrust capability of any commercially available engine! Subcomponents for major components could greatly enhance the individuality of the designs produced by players. It wouldn't be the coolest looking starship is the best, it would be deeper than that.

    That would be very cool, and would really help make R&D be an actual thing. Probably quite complicated to implement, though.

  15. But the flight characteristics will all be calculated on-the-fly by the physics engine. I'm fairly sure there's no way that you can make an analytical program to assess how well a certain ship design will handle and give it a numerical rating - the only way to know how a ship handles will be to test-fly it.

  16. creators might have a little idea on what they offer, so creators should give a complete mechanical description, otherwise will be difficult to trust

    If it's set by the creator rather than being calculated, though, it would be really easy to lie about. That's not great.

×
×
  • Create New...