Jump to content

PDCWolf

Members
  • Posts

    846
  • Joined

  • Last visited

Posts posted by PDCWolf

  1. I find that FAR makes building good planes substantially more difficult. Sure, I can make just about anything fly with enough thrust (KSP Interstellar fusion thermoturbojets could fly a city bus or herd of goats into orbit, with enough struts), but the tricky part for me is making it fly well. I've wrestled spaceplanes into orbit in heated, seat-of-the-pants pilot-versus-aircraft fights, but only rarely can I build something that doesn't wobble horrendously.

    Mostly depends on your definition of "flying well". I'm one of those that believes exposure to stock aerodynamics creates false expectations of how stuff should behave or even how it should work at all. Dihedral, incidence and things like the relative position and angle of main wing vs tail stabilizers are all factors that make for more stable planes, incidentally stock ignores all of this (by ignoring wings except when the Z axis is not exactly vertical), which means at all times you are trying to make a cylinder fly horizontally, so when transitioning to FAR, one still expects wings to have almost no effect, even if not knowing it, it's like a cognitive continuity thing. In FAR, all those factors come into play and yeah, you can't just slap wings anywhere, and this is what most people -i believe- find difficult, but it doesn't come from FAR, it comes from stock aero being bad (or rather nonexistant).

    Recently I spent a while figuring out how to make a very maneuverable fighter jet, and I did eventually succeed for the most part (analysis of the Firehound helped), but it was a challenge. I find that I really have to think on my feet when designing FAR aircraft, particularly super-/hypersonic ones or ones that are designed to maneuver especially well.

    Well, take a look at those that have already invented the wheel. Maneuverable supersonic fighters have short, slender wings because at high speed you don't need that many wing surface area, big control surfaces relative to where they are placed, and arrow look-a-like designs (most of them are delta or a variation of delta). Again, you have to think now because you are no longer trying to fly a cylinder, you now have wings you can rely on.

    I have met an especially large number of people who claim that FAR makes things easier - sure, it's possible to fly much faster at low altitudes (and in general) with FAR, and you can apply more real-world design points to your planes, but in terms of just sticking some wings on a tube full of fuel and getting it to fly nicely, stock aero is substantially simpler.

    Because wings don't matter, you are trying to fly a rocket horizontally, as long as you keep it mostly symmetrical or withing reasonable asymmetry weight wise it will work.

    I still have yet to see any craft in FAR that can remotely compete with the most agile stock planes, for instance, and you'll never end up in a catastrophic stall in stock. Stock aero enables more planes to fly with fewer complications, so in stock, arbitrarily-designed craft are far more likely to fly satisfactorily. If they don't fly well, usually they just flip out and crash; there's not a lot of middle ground like there is in FAR.

    Stock doesn't simulate air or wings, so there are no stalls.

    So does it make the game harder? Yes, in the sense that it makes it more complicated. There's no real way around that; I don't think FAR will ever be easier than stock aero for someone who's used to the latter. Once you get used to airplane design in FAR, though, it's potentially possible to do a lot more with your aircraft than you can in stock.

    "Makes things actually matter" would be the correct wording.

  2. Are you serious?

    Every rocket that has had what I have asked for exhibited a CoL ABOVE the CoM. That is a fundamentally unstable rocket. How am I wrong?

    If you put a light weight payload on top of a small rocket in FAR it is going to want to FLIP. Do the same thing in stock KSP and it won't.

    The one person that I've seen achieve orbit without tail fins to launch one payload used a heavy FUEL TANK; plopped it onto the top of the rocket above a decoupler and launched. How is this cheating? When the fuel drains from the rocket below the decoupler, the CoM will start moving UP towards the heavier part of the rocket making it MORE stable.

    I asked for people to build a probe core payload launching rocket without tail fins using stock KSP and FAR. Only Ippo listened to me and understood.

    Short rocket; light payload = flip. That was the point I was trying to make.

    And Iddo, the tailfin on the nose does shift the CoL above the CoM BUT it does also make the rocket more stable flying through the atmosphere. I bet that when you reached the typical 100 m/s velocity for your turn, you started to very lightly tap the yaw button about 1 degree every few seconds until you reached 87ish degrees and stopped in order to keep the rocket in check. The fins on the nose kept the rocket from rolling and since they don't articulate, shouldn't have been a huge factor in flipping the rocket at higher altitudes.

    Basically, launch: 100m/s tap yaw and hold pitch at 87 degrees until 30km+, THEN start turning for real when the air is thinner. Smart but inefficient.

    Anyways I'm tired of talking about this stuff. FAR won't be stock because stock KSP isn't balance for it.

    Well, yeah, if you go against the laws of rocket design then you aren't going to get stuff working, at least not without difficulties, that's why rockets are top heavy and with the center of pressure as low as possible (seen on that mythbusters episode with the guy with 40 rockets in the back and a skateboard or something like that). To worsen your case you show your lack of knowledge on the matter: Putting tailfins on the top of the rocket is one of the worse things you can do and it won't help you with stability at all, as they will continuously push to put themselves as far behind as they can, which means flipping your rocket so that they end up in the back.

    Javascript is disabled. View full album

    But yea, here it is, Last 2 images are with full and then with empty tanks. It did get hairy but well inside my skills, normal gravity turn, normal ascent to orbit, normal circularization. Just don't send the stars via paypal because they don't like us Argies.

    As far as using fins/wings/tails on the tail of the rocket, I'm trying to not to use them, however, I find that they help when I have instability problems that I can't otherwise seem to figure out, mainly a design issue thing.

    "fins on the bottom" is the easiest way to make something stable. Look at things like darts and arrows for example, weight on the front, pressure on the back.

    a) more difficult

    What are you finding difficult?

  3. Where is your center of lift on that rocket? Show me.

    EDIT: I meant build a rocket and launch a PROBE into orbit. Not launch a rocket with a probe core on it. My bad.

    And with regards to Dangit... and I suppose having 100% reliable parts is super realistic?!

    Javascript is disabled. View full album

    What's the exchange rate for gold stars to Argentine pesos?

    On a more on-topic note, I'm really happy with having both bac9 and ferram in the thread, and also for those guys that voted "I'll probably try FAR now"

  4. FAR is hard – on the processor.
    I don't have my tests at hand, but really, FARs footprint is almost nonexistant. I'll find or either redo my test and add this one to the OP, thanks for reminding me this excuse exists
  5. It was a poorly phrased statement. I oversimplified things when I implied that "fun" and "realism" were diametrically opposed, when I do think that it's more complicated than that. But the opposite idea - that increased realism automatically equals increased fun - is equally fallacious. I'm okay with people enjoying more realism - but the assumption that only complete realism is enjoyable is ridiculous. Should we remove time warp and do all missions in real time? Some of the arguments I hear imply this sort of conclusion.

    The thing is WHAT do you make more realistic: If you make heat or electricity management more realistic, then yeah, stuff ain't getting funnier at all, at least not for the majority that now has to control watts, amperes, volts, Celsius, Fahrenheit or whatever. But aero is something that definitely should be more realistic because everyone deals with it in real life (or knows how it is dealt with at least by looking at pictures or videos) and thus they have an idea of how it should work, otherwise (and as it is right now) you force them into a place where they have to throw all their knowledge away and relearn how planes and rockets work from scratch, which is definitely not fun.

    Im in the 'only ever played in stock' camp so anything they do will be an improvement. Those who like FAR or NEAR or whatever mods are used to them so it wont be the same for them. Whatever they do wont be final I dont think, just the first step in fixing it. So why not wait and see what it does right, does wrong, and how they can fix it. Everyone's definition of fun will be different and subjective.

    If I had to describe it, it is something more akin to "get it right the first time because we don't know how much are you going to leave us hanging before you fix it if you fix it at all"

  6. I thought most did? Huh, the more you know. Or maybe I was mostly thinking of missiles, which generally do have fins, especially cruise missiles.

    Anyways, what's the rule of thumb for where the center of lift should be on a rocket? I thought it was supposed to be under the CoM, or best at the CoM? Which is why I put winglets on all of my rockets.

    In rockets you want a low center of pressure (that's why fins, when they are used, go on the bottom) so that the airflow stabilizes the rocket (in FAR this also allows to move way out the velocity vector while maintaining control).

  7. As far as the adding winglets to rockets, aren't you supposed to do that anyway? I mean, look at ANY real life rocket, it has winglets at or near the bottom. Even the massive Saturn V has winglets, tiny ones, yes, but they're there.

    I haven't tried building any planes with FAR though.

    I named a lot actually and I can name more:

    Proton, Zenith, Ariane, Angara, Delta, Dnepr, Taurus, Athena, Antares, Atlas, Falcon, Minotaur, Ares, SLS, Peacekeeper, Minuteman, H2. None of them has winglets or fins at all. Falcon 9 has them for the flyback but they do not deploy during launch.

  8. In fairness, FAR used to be fairly difficult, due to the combination of aero failures and ultralight eggshell wings. Low altitude manoeuvres were things to be approached very, very carefully.

    Now, however, with tweakable wing mass, that issue is almost entirely gone. Unless you deliberately build an ultra-fragile ship, it requires actively suicidal piloting to snap a wing. Build something designed for high-G aerobatics and you're more likely to break the fuselage before the wings go:

    http://i1378.photobucket.com/albums/ah120/craigmotbey/Kerbal/Challenges/Kerbodyne%20Aerobatics%20Championship/Dragonfly%20Aero%201/screenshot1122_zpsd062b9f6.jpg

    And, again, it required actively suicidal piloting (low altitude supersonic >15G aerobatics) in order to make that happen.

    I have a conflict with this, I don't know about spaceplanes because well, we have none, but I do know most aerobatic planes are made to withstand loads of about 7 to 9G and going above that means a wing snapping off or both "clapping" as we call it here, so I'm pretty much ok with the default max load. For example, in that picture you are pulling above 15 gs and making a very sharp change in course (look at how far you are from the prograde marker) while going at mach1+

  9. Ever since I tried FAR for the first time, those three words that almost make a rhyme have been bothering me: Every time someone mentions FAR (or even NEAR, which I'll talk about too) these words appear.

    I don't really know what people find hard in FAR, so I'll address a multitude of points. Don't take this as "FAR is easy" either, this is obviously my view and opinion and I may perceive things as easy when other people find them hard so I'll just going to take the main arguments given for "FAR is hard" that I believe are unfounded or plain incorrect and debunk them.

    •The aerodynamic windows confuse me

    They are completely optional, you can totally do stuff without looking at them. That's actually something ferram wanted to demonstrate, along with other stuff, with NEAR and he created a nice placebo effect where players perceive NEAR as easier. Those windows are helpers and timesavers and as things like mechjeb and kerbal engineer, you don't NEED them. You can just resort to the trial and error play-style that stock has.

    •Flying with FAR is hard

    This one sounds subjective at first and, if it doesn't refer specifically to the aerodynamic failures the mod introduces, can be considered to be wrong too. Stock requires very specific parameters for planes to fly and so does FAR too BUT stock doesn't take things like wings into consideration so every time you fly, you are pretty much trying to push a cylinder against the air and get some lift out of it (protip: you won't unless you go really fast, even in real life). FAR on the other hand correctly computes wings and the lift they should provide. The easiest way to notice this is that in FAR you can actually fly forward without going up or down whilst pointing your plane at nearly 0º, this is impossible in stock, you need your nose to aim a bit up for it to work unless you give your wings some pretty obvious incidence.

    Now, add the following factors against stock:

    -If it doesn't have canards (or other dirty tricks like hiding control surfaces in the fuselage), you are in for eating up all the runway and a very difficult landing because you won't have enough pitch authority

    -Infiniglide: Even if it has been somewhat reduced, it's still noticeable on slow flying craft and can definitely screw up an approach by giving you more speed and making you unable to land unless you wait it out without touching anything and watch the runway's end getting closer and closer in the meantime.

    -No benefit from fairings, nosecones and cargo bays: Parts inside other parts still produce drag and can throw your craft off-balance or simply kill its efectiveness

    -Unintuitiveness: This is pretty much the consequence of what I have mentioned until now. Stuff just doesn't behave as one would expect it to until you get used to the fact that stock doesn't actually reflect how things behave when flying. Most people see the SPH and try to create a simple plane and it -won't- work for a reason or another (messed up CoM, too little control, won't take off, etc) and it stays like that until they throw away all they know about how stuff flies and relearn it again. You know, most non-airplane-savvy people probably would try something they saw on the internet, a picture of a fighter or a small cessna or something like that, and those won't work.

    RHDUd1H.png

    a RATO "paperplane" flying with FAR

    •Now I need to add winglets to my rockets

    Not really. Otherwise rockets like falcon, zenith, atlas, delta, dnepr, and most of them to be honest wouldn't be able to fly, and we certainly know they can fly so there's got to be a way. For rockets, I believe this is the only difficulty FAR introduces: The need for a proper gravity turn. As long as your turn is inside the normal parameters, which means keeping your rocket pointing at the velocity vector, nothing bad will happen to it (you know, like flipping and snapping in half).

    Just take a look at this Dnepr replica

    •Rockets snap with the slightest bending

    Well, no words needed for this one, just look at this watch your ears, this one is loud:

    It gets some speed in a direction but keeps turning, once it's almost perpendicular to the direction of travel it snaps.

    You can also experiment at home: Take a page from a periodic or pretty much any other paper, roll it, grab it by the base and use it as a bat, sooner or later it'll bend (it wont snap because it's paper but you get the point)

    •The wings do the same thing

    I'll show you something from a personal friend of mine, Dino Moline

    Here's a bigger plane with bigger wings. This one is flying at normal parameters:

    •The changes in physics with mach number make it hard

    This is pretty much an inverse placebo. Most people have no idea how the shock cone forms and reacts after going from transonic to hypersonic so they don't actually know what's happening at those speeds (or what would FAR change). In fact, knowing so little, you pretty much won't notice anything strange. You will notice those changes however if you go down from FAR to NEAR and really really pay attention at moments like reentry or when flying SSTOs and going from high speed atmospheric flight to the hypersonic phase and then the final climb into space.

    As a closer, have some more FAR awesomeness

    A x-15 like experiment in stock aerodynamics: a counter for both weight and drag is needed on the opposite side of the plane.

    The same thing using NEAR

    I have to admit I really like wingflex, too bad it also means there's wobble in the game in places where it shouldn't exist.

    An SSTO in tests coming in REALLY hot, saved by a lot of speed bleed-off techniques

    DreamChaser like "thing" being drop-tested.

    Since we are here, I also invite you to post some FAR awesomeness while we discuss the topic at hand.

  10. HarvesteR updated the article, it might help you to understand what he means :)

    And he still managed to miss the point.

    When he talks about backwards compatibility, most people (me included) get the feeling he's talking about leaving some sort of leeway for flying bricks to fly as they already do, and/or some kind of superglue'd hack is being implemented to keep the dV-to-orbit, and/or something similar to that. Not about file formats or keeping the old stock craft around (and making the new atmosphere in a way so that they work easily, which taking into account some designs, is a pretty bad idea)

    For me the scary part wasn't the backward compatibility itself, but the fact that it implied compatibility with broken design from a broken model

    here, he said it better than me

  11. Coming from someone who likes to provide assistance in the Gameplay and Questions forum, if we are expanding the realm of possibilities to 4 aerodynamic systems I may have to say "you're on your own". I hate to do that, I like being helpful when I can but it's bad enough trying to figure out stock vs FAR (since people seldom say which one they are using), then there is NEAR which is mostly the same except the whole wing strength thing. Let's just complicate matters more.

    NEAR doesn't simulate the supersonic/hypersonic shockwave which has some certain consequences on how stuff flies at very high speeds (it also makes things hotter on reentry because of that)

  12. Well duh, how would they even maintain backwards compatibility?

    I guess that they are not overhauling the aerodynamics but rather tweaking the already existent model. Probably including a (like they did with reentry or in fact just using the same) camera that sees the craft from the direction of the airflow to analyze the shape, overlaying some form of map on the craft to see airflow exposure and then model the aerodynamics from that data, this means "crazy stuff" flies too but rocket-like rockets and plane-like planes get a buff.

  13. Re-entry, and weather are to be added when the aerodynamics gets its overhaul. Squad wants to do all air based fixes in one go.

    Resources were cancelled for a reason. It wasn't fun, it wasn't balanced and it wasn't comparable to the rest of the game. Squad explained why it didn't come in the 0.19 update.

    fun is subjective, balance is not a factor here, just look at the rest of the game, and I don't even know what do you mean by not comparable to the rest of the game.

    Flying inactive ships and building ships in space was never on the table

    [citation needed]

    FTL and other star systems has been denied by Harvester as ever being part of the game scope. Don't lie, they were never coming stock

    Not lying, just quoting our old community manager on his words.

    Life-support was going to be added when they found a working resource system. No point adding it before. It's lack of existence is due to the resources not being added yet, so you can't count that as double.

    [citation needed]

  14. Mostly he talked about the resources system. That was cancelled based on internal playtesting, and looking at that overcomplicated chart of made-up substances I say thank Kod it was. I don't thing I'd have enjoyed KSP as much if it was full of interplanetary fetch quests like some grindy MMORPG, which is what I fear that resources system would have resulted in.

    As for other stuff mentioned, well let's have a rundown

    *list*

    Overall, not much that Rayne was talking about nearly two years ago hasn't made it into the game.

    I'd look at it differently:

    Will happen in 0.19 but didn't

    •Reentry heat

    Will happen in the next two to four updates but didn't:

    •Resources

    •Discovery system

    •Flight model engine rebuilt making you able to control inactive ships

    •Build ships in flight (definitely doesn't mean docking or related to that, it was added on the previous update)

    •Buoyancy update

    •Telescopes

    •Proper tweakables (only a small part did, not as it was mentioned on the resource chart)

    "Still on the table"/Discussed

    •FTL

    •Other planets/star systems

    •Weather

    •Eva First person

    •Life Support (supposed to tie into resources according to chart)

    A lot has happened, sure, but it's either incomplete or totally unrelated, and a lot of good ideas were said to be thrown out the window such as resources and other star systems. Bad things have also happened (curse)

    As for training kerbals, what was talked about was closer to this:

    Training them to do stuff implies that they can actually learn to do stuff (like execute maneuver nodes, or follow a simple flight plan)

    and not what we have now, but we still have some open tiers for that stuff on the kxp system.

    All of this shows that there's nothing to expect except for that what they tell you two days before releasing an update, given that with every update they may do a 180º turn like what they did with multiplayer and resources or delay things or just keep them from happening because god knows why (reentry heat was supposedly ready according to the interview). So yeah, talking about expectations or "what we should expect" is pretty much out of discussion.

    Also, resources was cancelled based on the same internal playtesting that left us with sticky launchpads, buggy claws, buildings with visible seams both in texture and mesh, etc. So yeah, keep that in mind.

  15. Well, I would vote all of the things you named if I could, but you made it a one option only so I picked planets. Even so, I still believe KSP should -with a well though of difficulty level- correctly display the dangers of spaceflight and how complex it is. And yes, it can do that while remaining a game, that's what abstractions are for, aren't they?

    It's been known since the beginning that what we have now was merely a placeholder.

    Planets and overall universe size were also placeholders because there was no timewarp, engine effects are placeholders, stats are placeholders until they get around balancing stuff, NERVAs using LF/O was supposedly a placeholder that would tie into the resource system and then resources would also work with the planned life support system, if I remember correctly, sounds are also placeholders and haven't been updated either. The only placeholders that have been properly updated are the KSC, and the space-plane parts.

    KSP suffered the same thing minecraft went through on the indev-alpha transition: They used placeholder assets but the community grew attached to them in a form or another, or rather got accustomed and now is afraid of change and puts up a lot of bull excuses to avoid it, and this makes the devs afraid of change too, even though they already have the money, they still fear the "hate" they are going to get (Protip: they won't).

  16. Question here, I started toying around with this and found out that the regolith definitions for Oxygen makes it a type 0 resource (crustal according to the comments in the CFG), is it ok for it to be like that? I mean, Oxygen is supposed to be mainly in atmospheres. Shouldn't it be type 2?

  17. Like already alluded to above I suspect what were seeing is a first draft, maybe rushed in for an 0.90 christmas release. Even from the distance the new buildings seem distinctly off in some way compared than the now tier 3 building. Looking at all the models in the rest of the game I wouldn't be surprised squad already knows the new buildings are a bit out of line and some touch ups were already in the planning.

    Let's remember those 0.25 dev diaries where they mentioned they were already working on assets for "the big feature in 0.90". That's months and months of work. With those months and months of work, they came up with a barn which was even worse than what we have now, it was criticized by almost everybody on the texture/mesh areas, and that's just for the barn, the rest of the buildings received almost the same criticism. That's why I don't trust we can get anything better than this in a reasonable timeframe, and for this argument I also quote the old buildings, the vfx, the lighting fix, etc, which took years before they were even looked at. Add on top of that the announcement for the next update: "Aerodynamics and Resources", which doesn't even come close to "fixing our ugly buildings".

  18. Regarding aerodynamics: I'm a big fan of FAR, but it's probably too difficult for new players ("what's a derivative?"). If such an advanced aerodynamics model will be implemented, it needs to be optional.

    Knowing what a derivative is is totally optional with FAR. Most intuitively-made designs won't fly in stock KSP, but they do in FAR. If it looks like a plane and it is balanced, it will probably fly under FAR (as it would in real life).

    I really fail to see a downside with far, given the advanced features such as charts and math formulas are optional and you don't even need to look at them.

    This is a game. Theoretically we could mine water from the moon, it would just take inordinate* amounts of time. Thus, we could get "rocket fuel" (hydrogen + oxygen) from nearly any rock in the kerbin system.

    There's a problem with that: Squad seems reluctant to de-abstract stuff, and explaining getting fuel from water to specifically make hydrogen and oxygen ends up in having to de-abstract the already existing "LF/O" system. If you decide to do that, then all of the current tank mass ratios and consumption ratios are wrong. They know this already though and that's why they invented "propellium" for example (which was only obtainable on EVE's oceans) back when the old resource system was being shown off, to avoid this kind of abstraction conflict.

    TL;DR You can't get fuel from "anywhere" with a rational -canon- explanation, or you end up killing the LF/O abstraction and having to explain/fix a lot of stuff.

  19. I really hope they go with something like FAR, physics changing with mach numbers is a must (reentry profiles change between FAR and NEAR because of that, so does the heating if you have DR). The charts and the specialized guys are not a -need-, given the aerodynamic model is pretty intuitive: if it looks like a plane and follows the basic laws of plane design, then its probably going to fly too.

    Deep space refueling? I don't like the sound of that. I want resources, not just GET MORE FUEL FROM THE SURFACE OF WHEREVER YOU ARE FOR INFINITE SPACE EXPLORATION, that's just pretty much slaughtering the entire feature and reducing it to the lowest common denominator possible. Give me something to maintain extraplanetary bases, give me something to get funds with, give me stuff that holds a scientific value. You killed the previous resources arguing it was not fun, yet you implemented the clickfest science system we have now, such hypocrisy. Now you have an opportunity to redeem yourselves with a meaningful mid-endgame feature that expands into the rest of the already existing game like the three currencies, the science system, life support in the future, etc. Don't blow it.

  20. Well, not sure why everybody is trying to run away from the topic, given the atmospheric work needed is pretty obvious (take the soup away for starters). If you take the soup away, then there's a lot less difficulty in reaching orbit. Both NEAR and Stock Drag Fix come in at about 3/3.5 km/s ratio so really, trying to avoid this discussion arguing we don't know what's going to happen to the atmosphere seems like a rather toxic behavior to me.

    Personally, I would prefer making the planets/moons bigger to match their previous dV-to-orbit requirements, that way you kill 2 birds with a single stone: You get the accustomed dV to orbit requirements, and you also get planets that make a bit more sense that the ones we already have.

  21. KSP most certainly did look like someone's first Unity game originally. It was one of the reasons I delayed buying it, graphics being one of the few readily visible indicators of the production quality of a game before purchasing it.

    I got the game after they removed the lakes, palms and roads (.13 being my first version) and although the shapes and such didn't make any sense except for the launch tower, it definitely didn't look like someone's first game. In fact, if you look at the first picture, you'll see that only the lakes and roads seem out of place, and while the rest of the models are not the best, they show consistence to a point. One of the first things that broke consistency and wasn't fixed until some months ago was the spaceplane parts, and that's really frightening because they have been in the game since 0.15 and they were already inconsistent, so one can only think that the same will happen with these buildings.

    I'll also add that the admin building added not long ago already sticks like a sore thumb, and it doesn't follow the existent style to certain extents like not keeping the hexagon (triangle?) based grid for building layout, inconsistencies in the roads, textures being used in ways they are not supposed to, etc.

×
×
  • Create New...