rcp27

While I can see a logic from the "realism" point of view that things like thermometer readings should be able to provide as much data via transmission as they do from actual recovery, from a gameplay perspective I can completely see the logic for the current system. It is way easier to land a non-recoverable probe lander on a body than to make a full return mission, so providing a gameplay mechanic that rewards actually coming back again is OK in my book.

The way I see it, contracts are not the driver of my space program. I play in career mode in the same way I do in sandbox and science modes. I have my own objectives that I chose to define for myself. The way I see it, contracts are a way of gathering the resource, funds, that I need to support my personal space program goals. If a contract happens to align with my own objectives, that's great, but if there happens not to be a contract associated with it, I'll do it anyway, or perhaps do something else and wait for a relevant contract to come up. If I'm running low on funds, I do a few profitable contracts to earn some cash, before returning to my own personal objectives. I decided fairly early on that I didn't want to be too tied to contracts, so I have set the custom difficulty in order to increase the funds yielded by contracts.

Personally I don't think SSTOs are overpowered. Let's consider the task of putting a small payload into LKO. A typical reason I want to put a small payload into LKO is as a first step towards a "place satellite in Kerbin/Mun/Minmus orbit" to earn funds. A typical contract will call for a probe core, power, an aerial and some sort of science instrumentation. I obviously need an engine and some fuel to get to the desired orbit from LKO, and some method of getting that payload up to LKO. Here's how I approach such a contract: The satellite itself contains an OKTO (early career) or QBE (once unlocked) probe core, a couple of surface mount solar panels, a Z100 battery, a small reaction wheel for extra controlability, a FL-T100 fuel tank, the soon-to-be nerfed 48-7s, a communotron 16, whatever science instruments the contract requires, and a decoupler to attach it to whatever I use to launch it. From LKO, this has sufficient delta V to get to any of the obrits of Kerbin, Mun or Minmus I have so far been offered (being careful to launch retrograde or polar if appropriate). So, using the "overpowered" SSTO concept, I have a spaceplane built around Mk2 parts. The design I have uses a pair of turbojets, a pair of small rocket motors, a bunch of fuel and a Mk2 cargo bay for the satellites. If I get an efficient ascent and accurate descent so that I use a small fuel load, I generally burn something like 1500 funds worth of fuel for the flight. I could probably get this down if I redesigned my spaceplane. To use this, I need a fully upgraded SPH to unlock action groups, and a level 2 R&D centre to unlock the Mk2 parts. I generally want at least a level 2 runway because the basic runway is a pain to use. Alternatively, I can try to build the cheapest possible disposable rocket. I saw this design on a Scott Manley video, so I can't claim the credit. I use a BACC SRB for the first stage, TR-18A stack decoupler and an RT-10 SRB for the second stage. To work well I set the thrust on the SRBs to about 60% otherwise this rocket reaches a bit too high a speed in the atmosphere, and requires a bit of use of the satellite's own engine to circularise, but it still leaves enough fuel to reach the final orbit. Total cost for launcher, 2000 funds. All of these parts are available without upgrading the R&D centre, and can be launched from the level 1 VAB and launchpad. I use satellite launches as the main way of earning funds to make my space program run. When I'm launching satellites, therefore, my main objective is to maximise profit per hour. Spaceplanes take ages to fly up to orbit and a fair amount of time to bring down safely. Rockets go up, well, like a rocket, and there is no bring-them-down to worry about because the launcher is disposable, and all the parts are discarded on a sub-orbital trajectory. I can complete perhaps three satellite contracts using my disposable launcher in the time it takes to get a single satellite up using a SSTO spaceplane. Finally, when I want to put something heavier up, I use SRBs to get a decent payload up on an otherwise single stage liquid fuel second stage, and give the second stage a probe core and parachutes. I tend to aim to land them back in the sea east of the KSC because it's easier to not destroy the returning stage with a splashdown than a terrestrial landing, and I can usually recover about 85 or 90% of the value of what I bring back. Sure, SSTOs are cheap, but they aren't so much cheaper than rockets that they are entirely unbalanced, in my view.

I like the idea that the good is a bit awkward to mount. There are ways you can deal with it, for example if you mount one of the smaller radial mount monopropellant tanks opposite and adjust the initial fuel load in it, you can balance the mass of the goo.

I think it's not unreasonable that the plane side of the game has had some loving in recent updates, and a few improvements are still needed. The new aerodynamics model is absolutely needed, not just for planes, but also for rockets. The space side of the game is pretty good as it stands, and getting a space plane to work is a fun challenge. To my mind there is one current limitation to the stock game that, I suspect, the new in-situ resource gathering will make a more irritating problem, and could do with some sort of attention. I can build an orbital space station out of a number of modules with docking and RCS in various methods. Building modular surface bases, though, is a much bigger challenge. To dock modules of surface bases together, though, requires vertically-aligned docking ports and the ability to drive modules together. Rover wheels, either built in to the modules, or on separate assembly vehicles, suffer from suspension travel, so that you never really know whether your docking ports will be vertically aligned when you actually get the modules on location. The problem is worse if you are on a not-very-level surface. With resources being added, I imagine building a mining outpost will become an interesting challenge, but assembling larger surface bases, and loading up the finished product into visiting craft seems to me a real challenge. I also have a niggling concern about how the new aerodynamics model will impact my ability to launch some of the more pancake shaped utility craft I use in space. I have a tractor design nuclear engined tug designed with engines on pylons to fit a 2.5 m payload behind it with the engines clear of it. I also prefer my landers to have broader bases, so the landers I use for e.g. base construction, tend to be quite broad and flat.

You will find if you accept such a contract that you get access to that part in the VAB/SPH once you have accepted the contract, and access is removed once you complete/abandon the contract. Seems reasonable to me: a company is working on a new shiny and they want to have their prototype tested before it becomes available more generally.

Just because we are not getting public access to the beta release (what you might call 0.91) does not mean that there is not beta testing. Early access means we get to play around with alpha and beta builds, not that we are alpha or beta testers of the game. Squad have a bunch of people (a few odd comments in Scott Manley videos suggests he is such an individual) who do actual pre-release testing of the software, and they most definitely are beta testing the new features (and are not allowed to talk about it). Open access and the kind of rolling release that common in a lot of software these days means the paradigm of alpha and beta versions doesn't really carry the meaning it used to. Each release, including all the versions we have had to date, will have gone through an internal closed alpha and beta test phase before we got our hands on them. Just because we don't get our hands on these builds doesn't mean they don't exist, and doesn't mean there isn't proper testing of them. In terms of stability and bug fixes, 0.90 is definitely beta in quality, but in terms of gameplay and content, I have played full release games that were far lower in quality over the years. If all they had done was fix bugs and give us 0.90 with a few stability enhancements and bug fixes, it would be a great game. The other thing to appreciate is that software development is mostly under the hood. What is difficult to execute in software, and takes time and effort to debug and get working right is stuff that, from a user perspective, is frankly unimportant and would warrant perhaps a single line in release notes. For example, the hard part in modelling the physics in KSP is being able to assemble a large complex spacecraft out of components, model their interaction under flight conditions, and have the thing stay together and be controllable with user input (and fly through space in a way that deals with a huge range of orders of magnitude of distance and time interactions). That is something that we take for granted because it works now, and it works pretty well. What is not hard is tweaking the values of individual forces, masses and so on of each component so that they make gameplay better. We, the user, see "full parts re-ballance" and think, "this is a huge change to the game". What the developer sees is, "this is a day or two with a spreadsheet to tweak some numbers on an already working system". We see "massive aerodynamics change will fundamentally change gameplay", developer sees, "one function is changed from F = m * v to F = k * A * v^2" (A is cross sectional area, k is arbitrary constant). We see "full resource system", they see "a handful of new parts and some surface maps". The things that will need sorting out to bring the quality of 0.90 up to 1.0 standard are things like memory leaks and stability. There have been statements on Tuesday developer posts indicating these are being worked on, but when we, the forum community see "big effort at stability and bug fixing" on one line and "new aerodynamics model" on another, we think the second one is the huge big deal, but actually, from a software development perspective, and from the point of view of getting the game to 1.0 standard, it's the first one that is way, way more difficult to achieve and way more important. TL;DR: Squad has done a damed good job so far with getting releases out to a high standard so far, I trust them to do it again. If they say "it's ready for 1.0", I am prepared to trust them.

The only probe core that used to have SAS in 0.25 to have lost it is the Stayputnik. All other probe cores have at least basic level SAS, and can be used exactly as they could in 0.25. All pilot Kerbals, even totally inexperienced level 0 ones, have SAS ability with the same capabilities as 0.25. You can add SAS to any ship by including a probe core that has SAS on it. That is, any probe core other than the Staputnik. Neither the physics models nor the aerodynamics models were altered in the transition from 0.25 to 0.90. If your craft is unstable now, it would have been unstable before. I don't use FAR, so I don't know whether it has been altered meaningfully. Not sure where you are coming from here. Stayputnik is the only probe core that does not have 0.25 style SAS. Any craft using another sort of core should work exactly like it did before. It can be a bit of a grind to get the cash to upgrade the various facilities, but the custom difficulty settings let you tweak starting cash and cash rewards for contracts to give yourself an easier time. Or you can edit the save file to give yourself enough cash to upgrade the facilities. Alternatively, just play in sandbox or science mode, where upgrading and contracts aren't a feature.

I'm not suggesting it has to be done in career mode, just that this measure is a useful one for establishing a "return to base" criterion. Of course you can always start a career mode with costs turned way down, rewards way up and a huge pile of starting cash so that you can get access to things like kerbal experience and working through the tech tree without having to worry too much about money. That said, if you are going for 100% reusable, it will probably work out quite cheap because your only expense will be fuel. - - - Updated - - - While the mechanic is different in stock, the effect is basically the same. Taking the "parts in stock" model, I build a space plane once, and pay full price. I fly it, return with it and recover the parts to stock. I launch the space plane a second time, and all the parts are drawn from stock. I therefore only pay for the fuel. In stock, I launch my space plane once, paying full price. I return to the KSC and recover the vessel and am refunded the purchase price for the parts, but not the fuel I burned. If I launch the same space plane a second time, I pay full price, but apart from the fuel cost, all of that money was the same as the money I got back for recovering the vessel. The effect on my bank balance is therefore identical.

Personally I would go with the principle that any vessel that is recovered from the KSC itself (preferably the runway or launchpad) counts. In the career mode that gets you 100% recovery value, so that seems a good measure.

Apart from the "its more realistic" argument, I have not seen any explanation of why construction time is actually a beneficial feature. What does it actually do to improve the gameplay? If build times are short (like a few days) then there is effectively no change to the current game. You just replace the "launch now" button with a "launch now (but with the game clock advanced a bit)" button, and the game is no different to how it is now. If you want construction time to actually matter, then it has to be in a way that they are long relative to either contract completion deadlines or relative to the arrival of orbital transfer windows. In each of these cases, I can see no clear benefit from the system, just a way of making the already cliff like learning curve even steeper for newcomers. If you take the view, "it should be in there but an option that only experience players will turn on", I would suggest the better option there is "download the mod for it". I see a similar sort of situation with life support. At present, life support is something that is built in to any capsule that can support Kerbals. Where's the O2, CO2 scrubbers, H2O supply etc? it's in the capsule. Suggesting that it needs to be an extra part to contain it is like the old SAS system. Every craft needs it, so all it ends up being is one more part to forget that increases part count and adds no additional gameplay options. If you make life support some sort of consumable resource, then you have the problem of resupply. Perhaps the first resupply mission to your cool space station is fun. Will it still be fun the 5th time? How about the 15th time? If it is consumed slowly enough for the resupply problem not to become game breaking, then it'll just be a matter of loading up your space craft with supplies and forget about it. Or just include loads in the capsules to begin with (which is pretty much where we are today). Of course then there is the worst of all situation where you combine these two questionable ideas. Imagine a situation with long build times and quickly consumed life support. As your space outposts in various places get larger and more numerous, the build time consumed by one cycle of resupplying each of them grows. There will come a point that, having sent a supply mission to each of your LKO station, your Mun orbital station, your Minmus orbital station, your Mun surface base and your Minmus surface base, the time taken to build each of these supply missions is such that you have to immediately start work on the next round of supply missions. Want to send an ion probe to the moons of Jool? Sorry, the VAB is fully booked with supply missions for the next 10 years. That sounds like a game I will stop playing.

I name my spaceplanes after types of bird. Once I'd run out of the more usual names, I had to use some less common birds. In my fleet, I have the "Ptarmigan", "Plover" and "Wigeon".

As the originator of the thread, I would like to add one other aspect of this particular challenge I set myself (and have now managed to pull off). There are lots of relatively easy solutions to the problem "land something on Minmus". That's easy, I've done it tons of times with built-for-the-job craft. I also agree that landing tail-first with a space plane is probably the easiest way to solve the problem of getting a space-plane onto the surface of Minmus. These are all well and good, but the challenge I set myself was landing a spaceplane on Minmus, using the flats as a nice open flat runway, in a way that resembles a plane landing on a runway. I wanted to do this not because it's the best way to get a space plane onto the surface of Minmus, but basically because I thought it would be cool. In achieving it, I can confirm it is definitely cool, and Jeb approves. My advice to others trying to achieve the same thing is to keep landing speed to below about 80 m/s, and vertical speed at touchdown below about 10 m/s, and to make absolutely 100% certain that you are absolutely pointed in the direction of travel when you touchdown, because any slight yaw angle will cause your destruction. Also, bring lots of monopropellant or a retro engine to slow down with, because brakes don't work well on Minmus. When taking off again, it is possible to lift off the surface simply by burning your engines on a straight horizontal takeoff roll, but unless you pitch up, chances are from the flats, you will not clear the higher terrain beyond. I thought about driving along the surface to find a nice slope to use as a ramp, but in the end just used RCS to lift the nose and get a trajectory clear of terrain.

I started from the premise that those big flats on Minmus would make a great runway. I wanted to use one of the regular space plane designs that I normally use for things like crew transfer to my LKO space station, so I didn't want to add extras like landing legs or VTOL engines, hence the need for a creative approach. Anyway, the visit has been accomplished, and the plane is now on its way back to Kerbin, with a load of science on board.

Well, a few more attempts and I got it. Landing speed was about 80 m/s, and I used a silly amount of RCS in retro to stop after something like a 5km landing roll, but I'm down. Now to figure out how to get up again.

... on minmus. So I set myself the challenge to build a space plane that I can take off and fly to orbit, where I refuel in LKO, then fly it to Minmus and attempt a landing. My approach goes as follows: circularise in a low minmus orbit, about 15,000m AP. Line up for one of the flats, (I went for the greater flats). start with the plane pointing retrograde in orbit, upside-down (landing gear towards the sky). Burn retro so that my trajectory meets the flats just after the near end. Pitch down so that the nose is pointing 45 degrees up/backwards, still upside down. Deploy landing gear. at about 150m altitude, burn until my vertical velocity is between -5 and -10 m/s. At this point I am doing about 120 m/s. Pitch "down" (nose goes over the top) so that I am in a slight nose-up landing attitude, use RCS to try to keep vertical velocity below 10 m/s, and hold down RCS in retro direction to try to kill as much speed as possible. Touchdown takes place at about 105 m/s. So far, I've attempted it 4 times, but I haven't managed to keep control after touchdown, each time I yaw a bit, a wing tip goes down, makes contact and provides a spectacular fireworks display, but I'm sure it can be done. Anyone else managed this? Is there a better approach? (I nearly wrote "is this a sensible approach", but I don't think anything about this particular endeavour is sensible).

Sometimes when I play KSP I just want to build cool rockets and stuff, and don't want to worry about contracts and funds. While sandbox is great for scratching that itch sometimes, I still quite like the "science mode" gameplay option because, while it takes the stress about making a launch as cheep as possible out of the equation, it still has a feeling that what I do is important to the progress of the space program as a whole, in a way that simple sandbox mode kind of gives up. In some ways it's better in that you don't get science from just completing contracts, you have to actually go out there and make scientific discoveries to unlock the tech tree. It gives the sense of achievement that you get with career mode without all of the funds related drawbacks (particularly with the feeling that you need to grind funds to upgrade KSC facilities). When I play in science mode now, though, one feature I really miss from full career mode is kerbal experience and levelling up. Can we have an option (or make it default) in "science mode" that kerbal experience, and the unlocking of abilities in the way that they do in career mode? At the moment I have a career save where I have set all the funds sliders to maximum to sort of give this kind of gameplay, but it would be nice to have it as an option, but it's not quite the same.

Sounds like a really good idea. I find planning planetary transfers quite awkward because there is no straightforward way to "plan ahead" to get a feel for when a good transfer window might be. Equally, I might want to just get there ASAP, but need to figure out just how overpowered I need to build my craft to get there. Equally, this could work for moons, for example planning a Mun to Minmus transfer, or transferring between moons of Jool.

You must be really cool! Thank you, thank you, I'll be here all week

If you ask the question, "are there any [member of arbitrary subgrouping of humanity] who [do activity]?" the answer will, almost certainly be "yes" for just about any way of filling in those blanks.

If you want to draw a distinction between "fun" simulators and "science" (what you call Industrial, but are as much used in academic research as in industry) simulators, that's fine. In that case, though, "fun simulator" is just one genre of computer game, alongside others like FPS, RTS, platform, beat 'em up and so on. That being the case, asking whether KSP is a game or a simulator is equivalent of asking whether an SUV is a "car"* or a "4 wheel drive". The answer is "both" * car in the general sense, I am aware that legally a lot of them are "trucks"

I have to say I disagree with this basic concept of where the split between game and simulator is drawn. There are tons of games out there that are essentially sandboxes, where you are given a load of basic elements with which to construct something, and it is up to your own creativity to have fun with them as you see fit. Games like Minecraft, The Sims, Sim City, Elite, Transport Tycoon all are definitely games rather than simulators, but are also sandboxes in which you as a player are given minimal structure to follow, and just do what you feel like. Equally, a computer program can be considered a simulator where the purpose is to run a business, follow a particular defined path and achieve carefully architected goals. The difference between a game and a simulator is down to the objective of the program. If the point is to provide a method for the user to entertain himself, then the program is a game. If the objective is to model a real-world phenomenon as accurately as the computational power available can manage, then it is a simulator. There are simulators that are fun to use and there are games that are highly realistic. My day-job is as an aerodynamics engineer, and one of the key tools I use to do it is computational fluid dynamics. I use powerful computer hardware/software to simulate, as accurately as possible, real world physics problems. KSP is absolutely not at all like proper engineering simulation tools. It is a game, and that's why I play it. I could produce a long list of shortcomings of KSP as a simulator that, if "fixed", would give it much greater physical accuracy, and in the process entirely ruin it as a fun pastime for anyone other than actual rocket scientists. KSP is like lego. You can build a house out of lego, but lego is not a building material, it is a toy. In the same way, with the right sorts of mods installed, you can plan an actual mission to the Moon with KSP, but it is not an actual spaceflight simulator.

Two of the most successful computer games franchises in the history of computer games are Sim City and The Sims, neither of which have much "actual game", and have "nothing to strive for other than goals [the player] sets for [himself]".

The announcement about 1.0 states that there will be a full part rebalancing as part of 0.90 -> 1.0 With the editor gizmos introduced in 0.90, placing the existing drogue chute in a radial alignment is not difficult, although a radial mounted part would be useful. Not having one isn't the end of the world, though. We already have an air-augmented rocket: the RAPIER engine. That feature was added in 0.90, using the R key. More planets would make a fine addition for a post-launch upgrade, but to be honest, the system as it exists now is big enough to keep a newcomer to the game occupied for tons of gameplay. In terms of what is needed to make a decent 1.0 launch, there is enough content now. Of course players with tons of game time will have been there and done that, but for newcomers, what we have is plenty. It took me nearly 200 hours of gameplay before I actually landed a Kerbonaut anywhere further away than Minmus, and I am still yet to send anything bigger than an ion probe to the Jool system.

A generator that uses LF + O to generate electricity would be a nice addition. Such devices exist in the real world, and are used extensively in real world space explorations, in the form of fuel cells.