Archer

For one thing, spaceships are not submarines. All of the engineering requirements are reversed (negative pressure vs positive pressure, match density to water vs build as light as possible, low-energy propulsion using the medium as reaction mass vs having to carry your own reaction mass, water-cooled everything vs radiant cooling, I could go on). Spacecraft have much more in common with aircraft than they do with submarines in terms of design parameters and space missiles have more in common with air to air missiles than they do with sea-launched torpedoes. For another thing we have built weapons to kill spacecraft. They're still called missiles. The US has built two that I know of; one was a modified sea-launched anti-aircraft missile used to shoot down USA 193, a malfunctioning spy satellite (or deliberately planned demonstration target, take your pick). Before that there was a multi-stage missile which was to be launched from an F-15 in a zoom climb to shoot down satellites. It succeeded in a few test runs but it never entered production. China shot down one of their own satellites out of nowhere as a demonstration a few years ago with a missile of their own. Also note that the main differences between a surface to orbit launch vehicle and an intercontinental ballistic missile are aim and payload.

I got the difference (just adding the Orion drive as another option), the point is that a NSWR could technically be considered a plasma core nuclear rocket since it works by heating its propellant mixture well into plasma territory but is otherwise pretty similar to an open cycle gas core nuclear rocket. Generally a plasma core fusion or antimatter rocket is assumed to heat the propellant to temperatures well beyond what any known materials can withstand, thus requiring magnetic fields to keep it from contacting the chamber walls. A NSWR would probably run into the same limitation and require a similar solution.

Ah of course, that and the good old Orion drive. Though technically a nuclear saltwater rocket could be considered an extreme form of gas core nuclear rocket... Plasma core nuclear rocket?

On the other hand a chair with a three-axis gimbal mount could actually help save you from high G forces. If the chair is placed on the ship's center of mass then the ship can freely spin and tumble around the pilot while the pilot doesn't feel any of it. Even off the center of mass the chair can align to the net acceleration vector, ensuring the pilot only ever has to experience Gs in one direction. Plus the more traditional spinning chairs will let you replicate the effects of some of the Origin-Jumpworks ships in Star Citizen, where the chair faces backwards for easy access then rotates to line you up with the flight controls after you sit down.

You can get a much greater variety of power plants and propulsion systems than you need even if you stick to relatively plausible systems. Personally I would pick anywhere from 1-3 representatives from each broad category, aiming for opposites (or two opposites and a middle-ground option) on whatever scale is applicable, and stick to them. Examples: There are dozens of chemical fuel combinations. I would just use a liquid fuel/oxidizer pair, monopropellant fuel and solid fuel. The fuel/oxy pair would give the best efficiency but also be the most complex, requiring two different feed systems. The monopropellant fuel has only one feed, ideal for RCS rockets and cheap spacecraft, but is less efficient. Solid fuel provides the greatest thrust but once you light the rocket there is no shutdown or throttle control until it either runs out of fuel or explodes. To assign real world names the fuel/ox pair could be methane-oxygen, kerosene-oxygen or hydrogen-oxygen, the monopropellant could be hydrazine or hydroxylammonium nitrate (HAN) and the solid fuel would be ammonium perchlorate composite propellant (APCP). Nuclear thermal rockets can come in solid core or gas core varieties. Both require nuclear fuel as well as propellant, though the propellant can be just about any liquid or gas as long as it doesn't react with any part of the engine. Gas core rockets run much hotter and give better performance but also tend to waste the fissionable fuel since some of it will inevitably find its way into the exhaust without burning. The closed cycle gas core nuclear rocket, AKA nuclear lightbulb, might be available as a tradeoff. It keeps the fissionables isolated from the propellant with its performance falling somewhere in between the other two extremes. It also conveniently avoids spraying radioactive death everywhere. Fusion fuels have several different possible reactions including tritium, helium-3, boron, etc with deuterium. Since they all lead back to a hydrogen isotope it's probably best to just regard fusion fuel as being extracted from raw hydrogen. Fusion rockets can be geared for high thrust or high efficiency depending on the needs of the spacecraft, though either way the craft is likely to be largely dominated by the magnetic confinement system. Antimatter drives would be visually and mechanically similar to fusion drives (and a single engine might be able to accept both deuterium and antimatter as fuel) but require something to use as propellant, similar to NTRs. Of course effective antimatter production or mining would be a huge project. They might also require constant power to the containers. Also, don't shake them. Ever.

Just a random idea I had a few minutes ago but it's something that might come in handy, particularly for small ships and for VR. Usually in voxel games, particularly Space Engineers, when you sit down in a chair you're locked in place. Your ability to interact with the environment is limited to a standardized control panel which probably involves an annoying menu interface. The idea here is to enable players to interact with anything within arm's reach while seated in basically the same way as they would while standing. This way you can place door controls, electrical panels, status displays, etc in voxel mode and have all of them fully usable while seated. This would make it more practical to build compact interiors that don't generally provide enough room to move around without being limited to pre-made cockpit elements. And yes, this also gives you the ability to spin your chair around.

On the topic of solar farms I'm imagining a solar powered particle accelerator occupying a substantial part of a moon's surface, possibly running around the entire equator. This accelerator has one purpose: Antimatter production. You know you want it. In gameplay terms this would be the ultimate in high performance fuel at the expense of requiring a huge infrastructure to manufacture it. Antimatter propulsion could be geared for high efficiency, making them well suited for long distance exploration, or high thrust, ideal for combat maneuvers. Antimatter power plants could similarly work both ways, either with a long endurance power plant where a little bit of antimatter goes a long way, or a high performance plant, for when you need to recharge your railgun capacitors right now.

I would use a hybrid system for ship weapons. Basically each weapon may or may not have a magazine depending on the nature of the weapon; a giant cannon might only load a single round at a time but a point defense gun might load a drum holding 1,000 rounds. On top of that you can build a conveyor system which will automatically pull ammo from a storage unit and load it into the gun as needed or you can omit the conveyor system and have players load it manually. In the case of the big gun this would mean either the conveyor or the players have to be active full time while the gun is firing and the limiting factor on rate of fire is how long it takes to load each round. In the case of the point defense gun the entire magazine is swapped out to reload in blocks of 1,000, though whether this is done by a player or a machine this takes time and forces a pause to reload. The choice between a conveyor and player is the classic auto-loader dilemma. If you spend the mass to have a conveyor system then you can generally reload more quickly but that mass is good for reloading and nothing else. If you spend the mass on another player (including the character's mass plus whatever additional life support or interior space might be required) then your load times might be slower and less consistent but that player can also fend off boarders, repair battle damage and serve as an extra set of eyes when they're not loading the gun. If you use neither then you can use that gun on a smaller ship but your gunner will have to leave the gun controls to reload the weapon. It should make an interesting design tradeoff.

Another vote for building large nanoformers/constructors, along with refineries (assuming some material processing step is required between mining and building, could be used to make things interesting). Ultimately I would like to be able to build a vehicle factory that can automatically construct anything within its size constraints with the only player inputs being the blueprint for the construct and the raw materials. Pipes and conveyors, possibly as separate elements specialized for transporting solid objects, liquids, gases, power, etc. Space Engineers oddities include the potential to build a functional system which uses a single conveyor pipe to transport ore, ice, liquid hydrogen, computer chips and ammunition through a nuclear power plant. Judging by the power conduits posted on Twitter the devs might be going in this direction already. Handholds. I mentioned these in another thread, basically the idea is to give people something to hold on to and pull themselves along when in freefall. The theory is that you can arrange them all in a row and basically climb along them a bit like a ladder. This would be helpful for when artificial gravity fails or you just don't feel like spending the resources to build a gravity generator. Explosive bolts. I don't know whether multi-stage rockets will be particularly useful at any point in this game but if you want to replicate early spaceflight properly you need some pyrotechnic decouplers. Also useful for rapidly ejecting anything potentially dangerous on the ship, such as a damaged antimatter storage pod, biohazard containment, mysterious alien artifact, etc. Solid fuel rockets. Small, cheap and with high thrust capabilities but only good for one use with a short burn time and no throttle or shutdown capability. These would pair well with the explosive bolts for ditching said hazardous items or becoming the first player to get to space... if you can manage that without blowing yourself up on the launch pad. Some sort of automatic voxel wrap system could make life a lot easier for designers. The idea is that certain components can be placed inside a voxel wall with a hole automatically cut out that is the perfect shape for it. This would make it easier to build partially enclosed rocket motors, internal cannons, pipes and conduits routed through walls, etc. It also means you can build small rooms which intersect with large meshes without creating annoying gaps between the wall and the mesh.

Realistically, probably not. A volatile fuel and oxidizer is an unstable combination; if you get a situation where one spark could rapidly change everything then chances are that spark will happen at some point in the millions of years the planet existed long before both oxygen and methane built up to the point where they could actually ignite. One could build up in the absence of the other but I doubt there are any natural processes that could account for the presence of both in significant concentrations. On the other hand this could be an interesting way to optimize propulsion to the environment. Conventional thinking would have you carry a fuel and use ambient oxidizer for an atmospheric engine. In an atmosphere with an abundance of hydrogen or methane you'll want to think backwards; carry an oxidizer and use the atmosphere as fuel. It also means that your O2 tanks are suddenly a little more volatile than they were otherwise.

Personally I think it should be possible to make some assets disappear when you log out, particularly since it sounds like the devs plan to make it take a long time to become space-capable. Suppose you start off at the ark, secure enough resources to build an aircraft and fly off to explore the rest of the planet carrying basically everything you own between the aircraft and your backpack inventory. Flying off into space to hide in the middle of nowhere isn't an option for an atmospheric craft; the best you can do is either fly all the way back to the safe zone or try your luck with some secluded place on the planet. Here is an approach I would take: Presumably FTL drives will require limitations for similar reasons. Start by figuring out what works here and apply the same restrictions to vanishing ships on logout. The obvious system here would be a charge-up delay that can be interrupted by an attack or the ship attempting any action. Set a permission flag to determine who is and isn't authorized to take the ship with them when they log out. This way you can set any ship you build yourself as a personal asset and take it with you when you log out, though when you fly a ship for an organization when you log out the ship is still available to other members as needed. This would be a situation where an autopilot would be useful, programming a guild ship to either fly back home or fly off into hiding if its crew logs off. If someone does borrow an organization ship for an extended mission then special permission might be granted. Don't allow a ship to vanish when any other players are on board, hostile or friendly. This way you can't just jump away from a boarding party or leave your allies stranded in space (unless of course you can airlock them all first). If, however, a multiplayer ship is flagged for a single owner then it can still vanish if the entire crew disconnects to make sure the ship isn't lost to some regional internet issues, server problems or very bad luck. Also don't allow it to vanish if someone else's assets are stored on board, for example if a crew member has their own shuttle sitting in the hangar. Finally, set each ship to check for all of the above conditions whenever the owner force logs or disconnects. Basically the ship will automatically attempt to vanish if all of the requirements are satisfied with all of the same limitations. If the pilot of a single seater disconnects then it can basically try to vanish immediately; if it is interrupted by combat then it will try again repeatedly until it is successful or destroyed. If the owner of a multiplayer ship disconnects then the crew can keep on fighting and/or try to retreat. Once the immediate fight is over the crew doesn't have to worry about keeping someone else's ship safe; they can split up and the ship will vanish on its own.

It looks like the game will give us an in-universe simulator where you can design a ship without having to actually expend any resources to build anything. From here you can figure out your ship's physical arrangement and put it through a flight simulator to see how well it handles. Of course if you want to actually build the ship you'll need to gather the required materials first but this means you can have a blueprint before you pick up a single gram of titanium... and possibly revise the design a few times after realizing just how long it would take to gather enough materials for version 1. It might be possible to make this available separately from the main game, though unlike the BDO character creator the flight simulator would have to load most of the game's physics and assets anyway so the biggest advantage over the full game would be that it might be usable offline. https://devblog.dualthegame.com/2016/02/20/builder-gameplay-voxel-tools-elements/

I have a few ideas for things I would like to be able to script my ship to do that would require a reference to a specific target location. Examples include calling in an airstrike, dropping a ground vehicle or supplies, picking up a big rock that's in the way or just parking at a specific location. A laser target designator could come in handy for that; instead of having to remotely pilot the ship and manually target a location from there you can pre-program various attack/deployment actions in Lua that reference a laser location. Once that's in place all you have to do is pick an action and point the laser at the target area. Naturally you can also give your friends permission to use their own lasers to borrow some of this functionality for a bit without turning over full control so you can design a ship to support an entire deployed squad. As an added bonus if you manage to integrate a working mech into this system you'll be about two clicks away from Titanfall.

The point I'm trying to make here is that if you have to worry about orbital mechanics then you will not have dogfights with visual range weapons. Imagine being on a wide open plain with no effective places to hide armed with nothing but a sword. There is someone else on that plain with you, also armed with a sword. If you try to chase him down you'll be able to close part of the distance, since there will be a delay before he sees you and decides what to do, but if he doesn't want to fight then he has plenty of time to run away. If he has the same endurance as you do then you'll never catch him; if you're both equally stubborn then you'll both collapse from exhaustion at the same time and you'll never actually get a swordfight. Now replace the sword with a gun. You still have to approach him but you don't have to get as close as you did before, just close enough to take the shot. That gives your target a much smaller window to start running before he finds himself in your gun range. Of course you still don't get into a swordfight; now it's a gun fight with all the tactics that involves. It also doesn't guarantee that you'll be able to get close enough but it does become a lot more likely. Likewise, with orbital mechanics and remotely plausible distances and speeds you'll never be able to use visual range weaponry effectively, you'll need something that can hit a target from thousands of kilometers away. Throw out orbital mechanics, set all movement relative to a stationary reference and confine the playing area to a smaller region (either by having some stationary point of interest players are likely to gather around or limiting the playing field to a small area) and you might be able to get away with shorter range weapons. Another difference between orbital vs stationary, if things can be stationary then you can have a lot of clutter to hide behind or have to maneuver around. This means asteroids, debris, defunct stations, etc., all as densely packed as you want. If orbital mechanics are a concern then nothing is stationary and these things cannot be densely packed in any kind of stable configuration.

Here are a few ideas that might work for both diversity and sufficient performance. Z-pinch Fusion Rocket: A stream of your preferred fuel type is magnetically compressed until it undergoes fusion. The nozzle in this case is an open cage-like design; exhaust temperature is so high that a physical material could not withstand direct contact with it so you need to contain and direct the exhaust using magnetic fields instead. This system would likely have high efficiency but not the best thrust to weight ratio (at least out of the systems on this list). Electrical energy can be produced by placing an MHD generator in line with the exhaust, conveniently requiring no moving parts. Gas Core Nuclear Thermal Rocket: A nuclear rocket run hot enough to vaporize its fissile material. The exhaust temperature is theoretically within the limits of certain materials so a magnetic nozzle is unnecessary; this one can use the more conventional bell nozzle. It is likely to have a superior thrust to weight ratio compared to a fusion rocket but inferior efficiency. Electrical energy can be produced by diverting some of the exhaust to drive a turbine connected to a generator. Nuclear Pulse Propulsion (AKA Orion Drive): Nuclear shaped-charge warheads are rapidly detonated below a pusher plate to propel the space ship. This is probably the least flexible of the designs; the only throttle control involves swapping out the nuclear warheads, precision is very limited given the bursts of thrust rather than continuous thrust and it is very difficult to use this drive to produce electrical energy... but when you need high acceleration there is really no substitute. Of course all three of these designs present a significant radiation hazard to both the crew and anyone nearby. The Orion obviously spreads radioactive material with each activation and the GCNTR produces radioactive byproducts which may or may not be expelled depending on whether an open-cycle or closed-cycle design is used but I don't know whether we'll have to worry about that in the game.

Remember that if you have that much delta V then your target probably does too. If the target's acceleration and delta V are comparable to your own and they are trying to avoid you then you'll never be able to catch up to them quickly since they can just match every move you make to keep their distance. Improved propulsion performance just gives them the flexibility to run away in whatever direction they want without having to worry about accidentally putting themselves on a trajectory to nowhere and no way to change course to return home. Forcing an intercept under these conditions becomes basically impossible unless you either have a huge advantage in propulsion or can start shooting from well beyond visual range. Either way a proper dogfight is basically impossible unless both players agree to match trajectories first. Just take a look at the difficulty of an orbital rendezvous in Kerbal Space Program (with or without infinite fuel and/or modded super engines, just assume the target has the same things going for it) and imagine that the station you're trying to approach is also trying to avoid you.

Lua could probably be used to put something in a sort of "fake orbit" but I don't really see the point; whatever it is will constantly be expending energy, propellant or both just to continuously change its trajectory (assuming Star Citizen-style Newtonian physics). Might as well program a defense platform to sit stationary until a hostile shows up, at which point it either opens fire from its current position or chases it down to either shoot or kamikaze it. As for actual orbital mechanics that would be... difficult. Never mind the computational power, in a realistic system you'll have a 90 minute orbit at a few hundred kilometers above the planet's surface. If your target is on an opposite orbit then you'll have an encounter every 45 minutes. Assuming a 3 km gun range (which is pushing it for the sort of aim-down-sights dogfighting most space fighter games use) you'll have a firing window of less than half a second in a best case scenario. Expand that to allow high orbits, high eccentricity, inclination changes, other planets and moons, etc and you're likely to end up chasing your target around for the better part of a year before you get a firing window lasting a few seconds. I don't really see a good way to implement this in an MMO regardless of how much processing power you have available. Note: I would really like to be proven wrong about this, if someone has a good way to pull it off please do so. Yes, I know about Children of a Dead Earth, multiplayer isn't on the drawing board there. What I would really like to see is an aerodynamics model in atmosphere and a Newtonian model in space, basically with something like Kerbal's aerodynamics in atmo and Space Engineers/Star Citizen maneuvering in space. This would still allow the sort of aim-down-sights dogfighting most space fighters are known for in both settings while still giving the two environments very different movement mechanics and design considerations.

Personally I would like to see the good old Casaba-Howitzer. It's essentially a nuclear warhead wrapped in a depleted uranium jar with the open side capped by a tungsten plate (same basic principle as the Orion drive for those familiar with it). When the nuke detonates most of the energy is directed into the tungsten plate, turning it into a relativistic jet of plasma. I imagine it would be resource-intensive to build and pretty heavy to carry around but when you absolutely have to cut that space battleship in half there is really no substitute.

Well if you think about the current respawn mechanics death already results in two copies of yourself; one of them just happens to be dead. In terms of physics there shouldn't be any difference with regards to whether your current reality version is alive or dead when the system activates unless the game subscribes to the idea of a soul. In this scenario instead of leaving behind a dead body you leave behind an NPC prisoner. Of course this would require a lot of additional game mechanics, ranging from giving prisoner NPCs a useful purpose to preventing people from creating a quantum clone army, but I don't think the basic idea is any more paradoxical than the current system.

Here's an idea: If you are captured enable respawn without suicide. This would mean whoever captured you still gets an NPC version of your character that they can imprison/ransom/enslave/etc while whoever was captured can still continue playing with the same penalties normally associated with death. If the player ever returns and rescues their NPC clone (which may be done by trade, effectively paying a ransom, or by force) then maybe the quantum reality trickery that they used to respawn in the first place can also be used to re-combine the two duplicates and reverse the original death penalty. Of course this only works if the penalty for capture/death is severe enough to discourage people from allowing themselves to be "captured" and cloned as NPC prisoners so I'm not sure what kind of penalty that would require.

It could also be interesting to use the ship's acceleration as a form of gravity if artificial gravity is disabled/not installed. In this case the direction of "down" will be the direction of the engine exhaust. If the main engines, gyros or maneuvering rockets are able to compensate for the center of mass shifting slightly (or if the game ignores the current position of people in a ship when calculating its CM) then you should be able to walk around normally as long as your ship is designed with the floor in this direction with the exact gravity strength dependent on the ship's acceleration. Cut thrust and you're back to artificial gravity, mag boots or floating even if the ship still has some speed. For a good example of how this arrangement would work watch any episode of The Expanse. Yes, I'm really hoping gravity generators will be optional. As far as the programming side is concerned Star Citizen had a way of pulling this off. From what I understand each ship generates its own coordinate system which governs the position of all players in, and possibly near, a given space ship. In this way the player's position is referenced relative to the ship, not the world. This way the movement of the ship is irrelevant with regards to the movement of people inside the ship... at least until you crash into something. DU could do something similar, using that base block as a reference for any players in, on or near a given space ship. With the game treating the ship as a stationary world it should be relatively simple to apply a gravity vector in the mini-coordinate system. Overall the arrangement would basically be what Fitorion said; the game first treats the ship as a stationary world then determines the visuals of the universe moving around it based on the ship's movement in the global coordinate system.

Atlantis. Because the opportunity to build a flying city is not something to pass up.

On the topic of character customization there are two things I would *really* like to see that very few games actually use: Transparent visors and customizable space suits. I don't know whether players will typically spend much time outside of their space suits but it seems that a lot of games completely remove any individuality as soon as you put on a space suit regardless of how deep the customization is otherwise. A transparent visor can go a long way towards making those options mean something and changing colors, parts or base models on the suit itself can add more options. Decals are another option I would like to see. Ideally I would want to see multiple decal slots on a given outfit (and on ships). One example I was talking about with AlexWright set one logo for the Cerberus Expeditionary Fleet and one logo for individual ships or players, so any individual in the organization might have a Cerberus patch on the right shoulder and a ship-specific patch on the left shoulder.

Some of the early screenshots suggest that ships will rely on artificial gravity, though I was hoping it would be practical to design a large ship with no artificial gravity. I would imagine that artificial gravity generators would probably be pretty time and resource intensive; even if they are relatively cheap it would still be nice to have the style choice. One thing that bugs me in Space Engineers is that you can't move around without either gravity or a jetpack, though using a jetpack inside a space ship is probably a bad idea. Here are some of my ideas on this: First thing's first, don't lock the player character in a standing position. Space Engineers (among others) tend to keep the character in an upright position facing forward as if they were standing up even when there is nothing to stand on. This tends to cause the character's legs to bump into door frames and things unless either the doors are oversize or the player lines up perfectly. If the character were set to something closer to a swimming position it would be much easier to get around and the interior corridors could be made smaller. Next up, handholds. At a basic level they can be made to work like a ladder in most games, though it would also be nice to see something resembling a simplified variant of Assassin's Creed's free climbing system, possibly with other context-sensitive movements depending on how difficult this system would be to program. With somewhat narrow corridors and regularly spaced handholds you should be able to pull yourself along fairly easily without having to use a jetpack. Finally, magnetic boots. This could be used as an easy way out, though they could also be used to interesting effect if you can switch them off and jump across the room at any time, with or without a jetpack. Also, if someone tries to dictate which surface is the floor you do not have to agree with them.