Kerbal Space Program

[warpus]

I'm definitely still savouring it! Something else I've noticed - other screwups in the game don't bug me nearly as much after anymore after all *that*.

I spent hours building a large and interesting speed rover last night, and I ended up not being able to build a delivery system for it, because.. the game started doing very strange things whenever the rover was loaded. ctrl-z not only undid the last thing you did anymore, it also left in place the changes as well. Effectively, it duplicated the last thing moved. I'm not sure how this started happening, but my pod in the original design was facing sideways. This made takeoff realy weird in that my markers were all in the wrong places - and various calculations the game did meant that it was impossible to use mechjeb to get into orbit - it would want to fly sideways and it was really confused. Then the game started doing other random things.. and that eventually lead to undo being broken and me accidently duplicating a bunch of parts on my rocket without seeing that they ended up exactly directly ovetop the old parts - leading to more chaos down the road. So I probably shouldn't do that again like that.. I also wasn't able to delete certain things from my rocket, clicking on one of these phantom parts would make it so that I could never unselect it again.. nor could I actually place it down anywhere.

This lead to a scenario in which I was not able to delete "everything but the rover". I had to delete everything. The .craft file was a mess too. And whenever I tried to launch this monstrosity? The rover had all these weird parts sticking out that were cannibalizations of other parts, in seemingly random places.

So after hours of work, centre of mass issues and fixes, and kerbin tests, I have to do everything over. And it doesn't really feel that bad.. I learned a lot spending hours with the rover and I might as well start from scratch, it will make the whole system probably just barely feasible while before it might have been a one in a thousand chance in terms of getting it to the mun and racing it around.

So thanks docking! You were annoying but now everything else is better.

edit: To the guys who got rid of their towers - did you have to do anything after you installed the mod to get rid of it?

 

[hobbsyoyo]

Jool *would* be fun. I went there once and the mission was a total failure

Which moons should I land on? I would prefer to land on one I could take off from, because I don't like to strand astronauts on planetary bodies unless I'm setting up a base.

You need to go to Laythe. It's a beautiful waterworld. I'll edit in some pics I took on my mission there. I think the gravity is manageable, it has an atmosphere and lots of small islands.

Landing is tricky precisely because it's all small islands, but once you're there, man, what a view!

It's also a preferred target for colonization, maybe even moreso than Duna.

I would skip Moho for now - that inclination is such a pain to deal with. Eve is also impossible to get off of until they give us some new engines.
Jool Pics:

Spoiler :

It's hard to line up the landing on Laythe.

So cool....

I continue thinking this game should have a much more completed interface, there is good bunch of raw data which would be nice to have in screen, for instance apo and periapsis times and distances or at the time of docking a way of knowing how you ship is oriented relatively to your objectives since trying to figure it using the chase view is a poor way of doing things (i dont want to be tiresome but they should take a look at Orbiter). I know the game tries to have a funny look like the sims or something, but spacial operations are complex and precise and we need more numbers. Mechjeb does a good job patching some aspects but the game should do it for itself.

Congrats on your docking!

I do not think the developers will be adding a ton of more information to the screen. If they do, it will be in an intuitive, easy to understand format (quite unlike orbiter). They have said over and over again they favor fun and playability over realism and I just can't see them adding more complexity to the game or the interface unless they can present it in an easy-to-use way.

Even the current info they are giving to us on screen will probably be repackaged to make it more useful and intuitive.

I wouldn't be surprised if the devs started talking to some of the mod creators into integrating the mods right into the game, perhaps in an altered state, before the full game is released.. I mean.. There's a lot of cool stuff the would be awesome in the final game and why should they spend time reinventing the wheel? The community has already done a lot of work for them, it'd be pointless to reinvent it all and go through the development process all over again.

They have already incorporated at least one mod that I know of and many of the parts they released in .18 were originally mod parts.

The one complete mod they incorporated was a physics mod that slowed down the time-step that the physics engine used. The original mod was called lagsbane because the intent of the mod was to help solve some serious lag issues that arose with larger rockets. It worked, and the devs put the thing into the official game in .18 and have since improved on it.

The game is considerably less laggy than it used to be if that is even believable!

Also, parts like the Orange fuel tanks came from Nova Silisko, who released mods with many extra parts in them. They have since hired him onto the staff and have officially released some of his parts in .18.

So they definately do incorporate good mods and will probably begin incorporating many more as they come closer to completing the core game. I imagine they will hold off on implementing a bunch of them for now becuase as they implement new planned features, the mods might conflict with those features in unintended ways.

There might be issues with profit sharing or whatever, but I can see a lot of mod creators being content with having their names in the credits and maybe what.. I don't know. Free 10 copies of the game to give to friends.. or whatever.

Yeah I don't know how they'll handle that but since the modders are already doing it for free, I doubt their will be many direct financial payoffs for them.

Either way though, they still seem to have along way to go in terms of delivering a polished product, and I think the community sort of realizes that. Development takes time - we're lucky they're letting us play alpha versions of the game in the first place! And as such should not really be complaining about lack of features.. and no I am not giving you crap here for doing that, after all I am probably the worst offender

Honestly, I think the way they chose to develop this game by raising money selling the alpha is a genius way to go and allows them so much creative freedom to make a product the way they (and the community) want it to be.

I am sure trying to pitch this idea to a big game company would be a nightmare and even if it had been approved, it would be a drastically different game than what we've got. For one, I would expect 5000x more weapons in the game if it were a mainstream release!

What's your next mission? I am thinking of sending a fleet of habitat modules, rovers, and a return vehicle to Duna and landing them right at the edge of one of the caps. That, or since I just realized it's already 9:30pm, I might try to put a giant rover on the Mun. I also wouldn't mind sending some probes to map surfaces of planetary bodies for me, to find interesting places to land and explore. Ahh so many options. I will have to think about this

Well, I'm working on the landers for my Duna Station today. I'm also considering adding a couple of expansions to the station itself and then I'll add quantum struts to it.

Of course i am not complaining. This game is totally brilliant. However i think that the intentions of the developer is to give it a casual look with the most simple interface possible so people do not get scared with tons of data which may not be very intuitive at first, but i think this way the game losses potential. Obviously games like this with accurate newtonian physics are not for casual gamers anyway but for people already interested in the topic so some hard data would not do any harm.

I don't think the game loses potential. I mean, it's hard to even claim that with a straight face given the fact that the game is in alpha stage, and is missing many key, planned features. It's just too soon to tell how much the developers design philosophy has made the game lose potential. Plus, this never was and never will be an Orbiter clone, it's probably never going to rise to that level of detail or realism.

BTW are there interesting places to explore? I mean are the different planets surface homogeneous or are there particular things to discover? That would be great! I can imagine caves, gas and minerals to drill or even lost extraterrestrial artifacts That would get a lot of players hooked. I hope the developers add things like that (i bet they are going to be very busy trying to add all suggestions from the fans) But thinking on it the possibilities are unlimited and maybe even doable!

Well, the planets and moons are kind of hemogenous in that the surfaces are kind of the same all over, with a crater here a crater there and maybe some hills and small mountains. However, the developers have added a lot of easter eggs that you can go and look for and are continually adding in details to the surfaces of the different planets.

They will be adding cities to Kerbin, for instance.

As for currently-existing interesting features:

Minmus has smooth, frozen ice lakes.
Mun has lots of Easter eggs like a couple of giant arches.
Duna has a 'face' on it.
Eve has molten lead leakes.
Jool is just gorgeous with blue waters and a green sky + Jool in the background

Oh yeah that is the greatest feeling in the world. Enjoy that feeling because from here on out, docking will be humdrum and mundane...

And TEDIOUS.

 

[hobbsyoyo]

Would anyone be interested in playing an ICBM game I thought up?
The game is simple:
Someone (a game master or something) picks a target by specifying geospatial coordinates on Kerbin. Next, they specify a payload size and any requirements such as "payload must deliver an RTG to target intact" or "payload must not have any rocket engines attached to it". The game master says which mods are or are not allowed to be used and specifies how many launch attempts players are allowed.

Then, players try and develop a launcher that can actually send the specified payload to the target. They must document each launch attempt with screenshots and show the final coordinates for each crashed rocket. The players who get closest to the target win the game.

What do you guys think? If people are interested in this game, we should start another thread for it.

 

[warpus]

You need to go to Laythe. It's a beautiful waterworld. I'll edit in some pics I took on my mission there. I think the gravity is manageable, it has an atmosphere and lots of small islands.

Laythe seems an intriguing target.. I wonder how easy it'd be to land there and return? I might just do this today, or at least try to.

I am sure trying to pitch this idea to a big game company would be a nightmare and even if it had been approved, it would be a drastically different game than what we've got. For one, I would expect 5000x more weapons in the game if it were a mainstream release!

Yeah, the model these guys have adopted is definitely unique.. I wonder if anyone's paying attention in terms of this being a success with an established community upon game release - seems like it could be a viable game development model for games that could even be very financially successful. And I think you're right - a rocket/space mission building and execution game would have probably been a lot different if it was released by EA or something.

---

I went back to the mun today to do some speed racing on its surface. I built a rather oversized rover and attempted to land it on the surface using a very small stage 1 to get me there. The rover was almost as tall as the launcher part of the rocket! It was a test launch, really, to see how many solid fuel boosters I would need to attach to be able to get there and land.. Turned out to be good enough to get me there, so I stuck with it. But uhh.. Landing wasn't so easy.

Munracer has 8 wheels, is incredibly stable, and is capable of turning slowly at high speeds. The wings help stabilize it and the outer wheels prevent it from rolling over. The solar panels provide extra balance and a ton of electricity for no reason whatsoever. Of course if you can manage to land it..

Spoiler :

I had to basically use up all the rest of my launcher's fuel (orange tanks) to slow down as much as I could, then detach it, spin around using RCS, then burn 4 small engines I had underneath the centre of gravity of my rover. This made the navball markers not line up properly and disoriented me - but I used the occasion to practice and hone my orientation/RCS skills, which as I've realized during docking, kinda sucked. The landing was challenging because the rover's centre of gravity and the 4 small landing engines did not fully match up, so it would slightly drift. That, and the engines did not give me much thrust at all, so I had to continue burning to keep the speed slow enough so that I would have enough time to land. Both those things with the navball disorientation made for many things to keep track of at once, but I guess all those docking attempts really paid off.. That, or something clicked in my brain. I have no idea what happened, but every single keystroke had a specific purpose. No trial and error.

Why didn't I attach more landing engines at the edges of my rover as opposed to keeping it so close to the centre of mass? Oh I should have totally done that

Almost there. "Nope, this isn't gonna work, what the hell was I thinking" going through my head

Spoiler :

Surprisingly enough though, I didn't have to load the game too many times before I got pretty good at keeping it stable, moving the camera around, and keeping a good grasp on the vectors involved and staying balanced wrt to the horizon, not moving laterally too much, all while my navball markers were all messed up due to the horizontal position of my cockpit. Anyway, it was hard but it wasn't nearly as hard as I thought it was gong to be. It just required "that magic touch", really. and I seemed to tap into my mojo for that magic touch on my 2nd try.

Touchdown moving quite a bit laterally at about 11m/s, facing diagonally, totally thinking at the time that I would have to reload.

Spoiler :

My rover's wings and supporting structure (with wheels) worked out VERY WELL in terms of giving me amazing balance. I thought landing was going to flip my rover this way and that - but nope, the design worked beautifully.

Let's test this thing out

Spoiler :

Hangtime = fun. Optimal and almost maximum cruising velocity turned out to be about 60-65 m/s (230-235 km/h). The thing was so stable that I could set my thrust as high as possible and leave it there, while I focused on turning. I could also hold down J or L and slowly turn while going at full speed and this did not lead to any problems at all with balance, even if I happened to be going downhill or bouncing around.. for the most part. The rover would bump around on the uneven terrain at those speeds and would get at times amazing amounts of hangtime, but it would almost never crash. This was the best jump of them all.

Spoiler :

Many cheers were heard at mission control when the resulting landing didn't lead to an explosion and the rover kept on going.

I did crasht 3 times, once I got to a significantly more steeper start of a crater with a bit of a "launchpad" I found, all 3 times in the same spot or very close to it. Basically what'd happen is I'd start going at speeds approaching 80m/s. I reloaded and tried again, and as long as I kept my speed to 70m/s or less, I was fine... almost no matter what.

Awesome. So much balance and I was totally not expecting it. I feel very comfortable placing and operating the RCS thrusters now too, which will really help with docking in the future.

Next up.. Laythe? I'd do your game, but I really wanna set up this mission. I feel inspired!

 

[cardgame]

Would anyone be interested in playing an ICBM game I thought up?
The game is simple:
Someone (a game master or something) picks a target by specifying geospatial coordinates on Kerbin. Next, they specify a payload size and any requirements such as "payload must deliver an RTG to target intact" or "payload must not have any rocket engines attached to it". The game master says which mods are or are not allowed to be used and specifies how many launch attempts players are allowed.

Then, players try and develop a launcher that can actually send the specified payload to the target. They must document each launch attempt with screenshots and show the final coordinates for each crashed rocket. The players who get closest to the target win the game.

What do you guys think? If people are interested in this game, we should start another thread for it.

I am totally game, I have landed on a tiny antipodean island barely visible from the map mode from a polar orbit

 

[GAGA Extrem]

Laythe seems an intriguing target.. I wonder how easy it'd be to land there and return? I might just do this today, or at least try to.

I have sent two missions to Laythe, but due to suboptimal approaches I was not able to return home.

The planet itself is a beautiful ocean world mainly covered by water with some larger islands. Lots of sandy beaches, probably one of the best places to get stuck for your Kerbanauts.

The atmosphere seems to be a lot less dense than on Kerbal, so it shouldn't be too hard to lift off again. My problem is more about the Jool transfer, which requires a lot more delta V compared to, lets say, Duna.

 

[cardgame]

You can practically get a free return via Tylo, though.

 

[warpus]

I have sent two missions to Laythe, but due to suboptimal approaches I was not able to return home.

The planet itself is a beautiful ocean world mainly covered by water with some larger islands. Lots of sandy beaches, probably one of the best places to get stuck for your Kerbanauts.

The atmosphere seems to be a lot less dense than on Kerbal, so it shouldn't be too hard to lift off again. My problem is more about the Jool transfer, which requires a lot more delta V compared to, lets say, Duna.

I remember having issues with getting intercepts and figuring out how to navigate in the system efficiently, but I'm going to give it another try. 1. Dinner.. 2. Mission design & execution.

Goodbye Saturday

 

[hobbsyoyo]

I'm testing out a torpedo system. My original sepatron idea didn't work out, so I've switched over to SRB's. The advantage of this system is that they can be operated either in dumb-fire mode or in guided flight.

Spoiler :

I am testing them out now, I'm going to launch two of them and use one to blow the other one up. XD

Why?

Because AMERICA F*&% YEAH!!!

lol

Hey, so for the ICBM game, does anyone know of any mods that will display the final coordinates of a crashed rocket? I just realized that the little pop up that shows up at mission end/crash doesn't show your coordinates so we'd have no way to track where our rockets landed.


One other thing: I don't think Laythe's atmosphere is as thick as Kerbin's, just FYI.

 

[Thorgalaeg]

Congrats on your docking!

I do not think the developers will be adding a ton of more information to the screen. If they do, it will be in an intuitive, easy to understand format (quite unlike orbiter). They have said over and over again they favor fun and playability over realism and I just can't see them adding more complexity to the game or the interface unless they can present it in an easy-to-use way.

Even the current info they are giving to us on screen will probably be repackaged to make it more useful and intuitive.

I don't think the game loses potential. I mean, it's hard to even claim that with a straight face given the fact that the game is in alpha stage, and is missing many key, planned features. It's just too soon to tell how much the developers design philosophy has made the game lose potential. Plus, this never was and never will be an Orbiter clone, it's probably never going to rise to that level of detail or realism.

You said it yourself, the info is not going to be increased in the final version, so we can comment on it with some basis seeing the alpha only. Well, it is logical seeing the comic funny style of the game. But in many ways the game is already as realistic as orbiter, the main difference is that we have almost no way to know what exactly we are doing. For instance without mechjeb at launching you must change to map screen continuosly to know where the apoapsis is or if you have circulatized your orbit yet. And sometimes it is a matter of seconds. Same for most operations. IMO it would be nice and would make for a better playability to have that info in the main view while you are piloting. If they dont want to resemble orbiter in any way or put in any complicated numbers at least could add a mini-map at a corner of the main view or something similar. That would make the trick for me.

Anyway it is the freedom we have building and developing things that has me hooked so any decision they take on the interface and such would not be that important for me and could live with it, only thinking outloud.

 

[cardgame]

Hey, so for the ICBM game, does anyone know of any mods that will display the final coordinates of a crashed rocket? I just realized that the little pop up that shows up at mission end/crash doesn't show your coordinates so we'd have no way to track where our rockets landed.

Parachute a target capsule down and we'll see our missile's distance from the target. Alternatively, use the second KSC as a target, or even the first, or the airfield.

One other thing: I don't think Laythe's atmosphere is as thick as Kerbin's, just FYI.

Laythe's atmo is roughly 2/3 that of Kerbin's with 3/4 the gravity. Roughly.

 

[Civ'ed]

So, for my first actual attempt, I tried to send a satellite into orbit around Kerbin.
I kinda failed...

Spoiler :

Luckily the satellite is solar powered.

Spoiler :

Here's a picture just after the decoupling of the remainders of the rocket

Spoiler :

 

[GAGA Extrem]

Well, orbiting around Kerbal, orbiting with Kerbal - whats the difference?

Do you have an Ion engine? In that case you can just re-enter Kerbals gravity well and slow down to bring it back into planetary orbit.

 

[cardgame]

Finally figured out fairings!

They still don't actually work properly (decouple) but they're sexy.

Spoiler :

 

[warpus]

Somebody on reddit was kind enough to summarize the extent of upcoming changes in the game, mostly what'll be new in 0.19:

2:31 - Unity 4 support, which means linux support.
2:33 - 0.18.3 is coming out in the next weeks, includes unity 4.
2:35 - flag planting will be in 0.19, expression AI on Kerbals should be updated.
2:36 - flags may be customizable in-game, game database is restructured (in a good way) changing how mods can use resources for things like compound models in a single physics part.
2:40 - memory usage may be improved
(gotta go get some lunch, I'll be back soon)
2:49 - 'kerbal docking' allows kerbals to 'dock' to seats on rovers, becoming a driver of the craft. Can be ejected from the seat by bumps, could also get a first person camera for EVA.
2:51 - demo is updated to current version of the game, but missing several features.
2:53 - login usernames may be consolidated, meaning you have one account for all things squad.
2:55 - visual overhauls of forums, spaceport and website, as well as a re-write of the patcher in Python.
2:56 - "Kerbal Knowledge Base", 'you can see the things you scan'.
2:58 - Merch is coming soon-ish, is of very good quality

 

[hobbsyoyo]

Somebody on reddit was kind enough to summarize the extent of upcoming changes in the game, mostly what'll be new in 0.19:

2:31 - Unity 4 support, which means linux support.
2:33 - 0.18.3 is coming out in the next weeks, includes unity 4.
2:35 - flag planting will be in 0.19, expression AI on Kerbals should be updated.
2:36 - flags may be customizable in-game, game database is restructured (in a good way) changing how mods can use resources for things like compound models in a single physics part.
2:40 - memory usage may be improved
(gotta go get some lunch, I'll be back soon)
2:49 - 'kerbal docking' allows kerbals to 'dock' to seats on rovers, becoming a driver of the craft. Can be ejected from the seat by bumps, could also get a first person camera for EVA.
2:51 - demo is updated to current version of the game, but missing several features.
2:53 - login usernames may be consolidated, meaning you have one account for all things squad.
2:55 - visual overhauls of forums, spaceport and website, as well as a re-write of the patcher in Python.
2:56 - "Kerbal Knowledge Base", 'you can see the things you scan'.2:58 - Merch is coming soon-ish, is of very good quality

So does the bolded thingy mean we'll be able to scan planets in .19 and have that scanned data saved?

Currently, you can only scan with mods.

Parachute a target capsule down and we'll see our missile's distance from the target. Alternatively, use the second KSC as a target, or even the first, or the airfield.

Can you explain how the target capsule would work?

Do you mean to use a target capsule and then upload a save file that we can all play off of?

I don't know how to do that.

 

[cardgame]

Just drop a capsule with a chute on it, nothing complicated.

 

[hobbsyoyo]

So are you talking about then uploading that savefile?

How would anyone know where that pod is otherwise?

 

[cardgame]

We could do a savefile I suppose, for greatest equality

I was thinking specifically of dropping it in an easily describable location such as an island, taking a pic from map view, and then a final with how close you could get to it.

 

[hobbsyoyo]

Ok I like the 'describe it' method. Not sure how accurate it can be, but it should still be fun.

Though if you are able to get the capsule to the spot, I guess your ICBM can easily hit it again so the only issue that remains is your aiming ability. Maybe the game master shouldn't play along.