Jump to content

What is your biggest issue with ksp 1 that you want to see fixed in ksp 2?


Rutabaga22

Recommended Posts

7 hours ago, Bej Kerman said:

It's an issue with the aerodynamics, what kind of question is this. 

I don't know if you intended the sentiment, but this comes across as unnecessarily rude. Pthigrivi asked for clarification on what the specific problem is since you seemed informed, you said "I'm not writing up an academic paper for you" (which was not requested, but whatever), and then they asked for clarification from anyone else to explain the problem. You are not obligated to provide an explanation if you do not wish to, so dismissing someone else's question is unproductive.

Edited by TROPtastic
typo
Link to comment
Share on other sites

7 minutes ago, TROPtastic said:

I don't know if you intended the sentiment, but this comes across as unnecessarily rude. Pthigrivi asked for clarification on what the specific problem is since you seemed informed, you said "I'm not writing up an academic paper for you" (which was not requested, but whatever), and then they asked for clarification from anyone else to explain the problem. You are not obligated to provide an explanation if you do not wish to, so dismissing someone else's question is unproductive.

In either case, they quoted me. Apologies if it came off that way :)

Link to comment
Share on other sites

14 hours ago, Nazalassa said:

Well, that won't solve everthing, but at least:

  • orbit randomely changing during time warp
  • parts randomely rotated
  • all the precision issues
  • all the other precision issues
  • etc.

because they are more precise.

For other stuff (like the bug that when you board a module the Kerbal's body stays outside), there are other ways to fix them.

Actually, no.  It would help, but you need to understand why it happens to really understand why just increasing the precision won't solve the problem.  It would just mask the issue, which is the insane precision needed to do this sort of a space sim.

Link to comment
Share on other sites

10 hours ago, Pthigrivi said:

Right. Im looking for a slightly more in-depth and informative answer, if anyone knows. You already said you weren't prepared to write a paper on it and thats fine. Thats why Im asking if anyone else knows more about this.


It could have to do with the way they've mapped the US Standard Atmosphere onto Kerbin. Atmospheric density looks pretty similar for the first 30km. Temperature also effects the speed of sound, and that doesn't look wildly different except that it cuts off at 70km

https://en.wikipedia.org/wiki/U.S._Standard_Atmosphere#/media/File:Comparison_US_standard_atmosphere_1962.svg

vs

600px-Kerbin_Atmosphere_T&P.png

So more likely it has to do with the way mach effects and drag are modeled for parts? I remember shortly after 1.0 was released they made a number of changes to drag values and other physics and temperature calculations because of problems with reentry and rockets tumbling over. The change log for 1.03 lists "Drag curves modified to lower transonic hump." and 1.05 lists "Drag coefficient changes based on the same factors as turbulent convection (a Pseudo-Reynolds number). This means higher drag high up in the atmosphere, and slightly lower drag when going very fast very low." If I remember correctly there was a whole conversation about unrealistic behavior and they decided to deliberately jigger things to make rockets launches and capsule reentry less frustrating. This isn't my field though so I could be wrong about that. 

Its not a problem with the density profile of the atmosphere since a linear slope on a lin-log chart is the only sensible way for an atmosphere to be modeled, it's just a lot of over simplifications. the game would need to include a lot more niche factors like thrust profiles vs relative wind speeds for engines as well as a more voxel based drag model. For instance, a scramjet cant turn on with an engine at 0 m/s nor even just mach 1, its just not in the cards for that type of engine since it relies on forward motion to compress the air instead of a turbine, yet in KSP  1's world a scram jet would be able to be used at takeoff with the type of modelling KSP 1 does, as I understand it at least. I think something like this could be solved by simply including a thrust profile vs relative wind speed for any air breathing engine that moreso matches real world performance without going into all the CFD bs  to do a genuine simulation. As for other aerodynamic qualities like wing/air interactions the cube drag model is overly simplistic and would require a finer approach with some more sophisticated simulations  including transonic and supersonic regions similar to how FAR works (and proves its method can be effective in real time).

Also, ever notice how a plane with 0 AoA flying upside down still gets lift pointing up from the ground and not relative to the aircraft its self?  These kinds of things should be taken care of.

Link to comment
Share on other sites

PERFORMANCE.

First, KSP1 has some kind of nasty memory leak that gradually grinds the game performance until you reset. Second, graphically, you do a lot to make the game look nice, but your framerate will just be awful. Third, the enormous RAM usage, the absurdly long 'render' times when approaching things, terrible performance with 400+ part space stations, etc.

Quote

 

Also, ever notice how a plane with 0 AoA flying upside down still gets lift pointing up from the ground and not relative to the aircraft its self?  These kinds of things should be taken care of.

This is a concession of the wings we have. They can't really know how we have the wings used and they need to be dynamic enough for all sorts of usage, so... that's what we get.

Edited by Frostiken
Link to comment
Share on other sites

5 hours ago, linuxgurugamer said:

Actually, no.  It would help, but you need to understand why it happens to really understand why just increasing the precision won't solve the problem.  It would just mask the issue, which is the insane precision needed to do this sort of a space sim.

Yes... But we have, well, (almost) no other way to fix the precision issues.
For the orbit randomly changing, I admit that the problem isn't entirely due to precision, but rather to the game applying physics to the ship during time warp, therefore losing precision and changing the orbit.

Link to comment
Share on other sites

For me personally, it's the fact that you can go to the tracking station on day 1, and zoom in to any celestial body you want for a full, real-time, 360 view of said body. I did it for fun when I first started playing, and then I realized that it totally killed my motivation to go exploring. So I ultimately never really played much. 

The ResearchBodies mod fixes this issue, but hopefully stock KSP2 does something similar that gives you gradually better views of each planet as you explore them.

Link to comment
Share on other sites

For me it was definitely the lack of information before they added the dV calculator. I could get as far as Duna without it, and then I had to download a mod to show me that information in order to figure out what I was doing more accurately.

Link to comment
Share on other sites

On 8/23/2022 at 11:35 AM, Nazalassa said:

That's with the help of a mod. Even if this can work, we probably also need a stock way to solve the problem, no?
Anyways, it's not a bad idea.

Ok, so I've been messing around with EVA Construction more, including in zero-g, and I'm expanding my opinion somewhat: I'm glad they did EVA Construction;  it's a wonderful concept and absolutely something the game needed/benefited from...but the execution is just atrocious. It's awkward and unwieldy, the building process is inconsistent with other parts of the game, it doesn't feel logical, and it's just generally inconvenient all around. I would love to see it overhauled into something more user-friendly and logical in KSP 2. Hell, it's fine even if they don't necessarily want it to be too easy at first, just have construction arms and the like that you can unlock to add more functionality and increase weight limits. 

Edit: Oh yeah, and all is only if it even lets you do anything (especially trying to move an already placed part) without everything going spaghetti kraken or just exploding, which is what usually happens.

Edited by GigFiz
Link to comment
Share on other sites

  • 2 weeks later...
On 8/23/2022 at 11:12 PM, Pthigrivi said:

Right. Im looking for a slightly more in-depth and informative answer, if anyone knows. You already said you weren't prepared to write a paper on it and thats fine. Thats why Im asking if anyone else knows more about this.

Since anyone else hasn't made any real effort to do so,  I'm gonna help you out on this. The reason why it's so easy to break the sound barrier(in KSP) is that both the planes and the aerodynamics are incredibly unrealistic.

1) Parts are unrealistically dense, which increases their ballistic coefficient(A crunched up ball of paper flies further and faster than a sheet.)

2) The Jet Engine mechanics... where do I even start. They have incredibly unrealistic thrust curves, and almost every engine can survive Mach 2 speeds. In real life, You need to build an engine specifically for supersonic flight. A high-bypass turbine such as the Goliath will flame out(compressor stall) at anything over Mach 0.9; and will probably not even survive in supersonic flight(as just dead weight)

3) There is no consideration whatsoever on wing sweep/ geometries. A biplane will break the sound barrier just as easily as a highly swept delta wing, and this is simply not the case IRL. Similarly, phenomenon like Area Ruling, and Wave drag is not considered whatsoever in KSP.

4) Air intakes. Yes, even intakes need to be fine tuned for breaking the sound barriers. If not, they will cause supersonic air to enter the compressor, and cause the engine to die.

5) Shockwaves are actually incredibly violent, if your plane is flimsy it will for sure get destroyed in IRL. KSP takes no account of this issue.

There are several more issues with KSP's aero model, but these are the more major ones.  If you want to learn more, play with FAR and AJE to actually learn how hard breaking the sound barrier is.

___________________________________________________________________________________________________________________________________________________________________________________________________________________________

As for the question in the OP, I have two man problems with KSP as it stands

1) The aerodynamics are incredibly hacky. Like you can design the most sleek looking plane in existence, and then someone will come along, using cheats or hacks, and have a 10X better L/D ratio and 10X better performance. I can deal with all of the inadequacies listed above, but this is where I draw the line. I believe in the quote, "For a plane to fly well, it must be beautiful." and KSP just doesn't care. Why? Because KSP doesn't actually see the plane, it just sees a list of parts. If you are going to have a simplified aero model, then at least make sure its complete, and doesn't allow for hacking away the game's mechanics. KSP's aero just feels incomplete. I play with FAR almost exclusively now for this very reason.

2) The Wheels. I. Hate. The. Wheels. They literally do nothing that real wheels are supposed to do. IMO, they are by far the worst part of KSP.  KSP 2 needs to have better wheel mechanics.

Link to comment
Share on other sites

4 hours ago, DA299 said:

Since anyone else hasn't made any real effort to do so,  I'm gonna help you out on this. The reason why it's so easy to break the sound barrier(in KSP) is that both the planes and the aerodynamics are incredibly unrealistic.

1) Parts are unrealistically dense, which increases their ballistic coefficient(A crunched up ball of paper flies further and faster than a sheet.)

2) The Jet Engine mechanics... where do I even start. They have incredibly unrealistic thrust curves, and almost every engine can survive Mach 2 speeds. In real life, You need to build an engine specifically for supersonic flight. A high-bypass turbine such as the Goliath will flame out(compressor stall) at anything over Mach 0.9; and will probably not even survive in supersonic flight(as just dead weight)

3) There is no consideration whatsoever on wing sweep/ geometries. A biplane will break the sound barrier just as easily as a highly swept delta wing, and this is simply not the case IRL. Similarly, phenomenon like Area Ruling, and Wave drag is not considered whatsoever in KSP.

4) Air intakes. Yes, even intakes need to be fine tuned for breaking the sound barriers. If not, they will cause supersonic air to enter the compressor, and cause the engine to die.

5) Shockwaves are actually incredibly violent, if your plane is flimsy it will for sure get destroyed in IRL. KSP takes no account of this issue.

There are several more issues with KSP's aero model, but these are the more major ones.  If you want to learn more, play with FAR and AJE to actually learn how hard breaking the sound barrier is.

___________________________________________________________________________________________________________________________________________________________________________________________________________________________

As for the question in the OP, I have two man problems with KSP as it stands

1) The aerodynamics are incredibly hacky. Like you can design the most sleek looking plane in existence, and then someone will come along, using cheats or hacks, and have a 10X better L/D ratio and 10X better performance. I can deal with all of the inadequacies listed above, but this is where I draw the line. I believe in the quote, "For a plane to fly well, it must be beautiful." and KSP just doesn't care. Why? Because KSP doesn't actually see the plane, it just sees a list of parts. If you are going to have a simplified aero model, then at least make sure its complete, and doesn't allow for hacking away the game's mechanics. KSP's aero just feels incomplete. I play with FAR almost exclusively now for this very reason.

2) The Wheels. I. Hate. The. Wheels. They literally do nothing that real wheels are supposed to do. IMO, they are by far the worst part of KSP.  KSP 2 needs to have better wheel mechanics.

thats pretty much what i have an issue with in general of ksp, the list of reasons, and then what you stated. there is no challenge getting rockets/planes past mach and there isn't really no "evolution" of machines getting better... if colonies are coming into the game and resource gathering i would love to be able to build runways near my operations, be able to take off and land planes with props that don't go past mach and the machines have "weight" to them, the unrealistically high performance of everything in total is somewhat a shut off, they could make everything "harder" for people that want it harder, but people will complain "if it isn't going to be the "majority" of players using it why implement it"... my simple reasoning is, once its implemented, its "one and done" they don't really have to worry about much about it once it is finished, and the option of having it without a mod is a huge plus..

 

There is no real sense of progression overall in the side of "planes" (besides the wheels getting worse, god the wheels) i understand its called, ksp but the lack of real aerodynamics for planes, or things that don't really go into space (or machines that just work in atmospheres) is quite boring and i would love to see improvement upon it give more of a challenging in a way, i would love to start with capabilities of building horrible cargo planes things such as DC-3, slowly work up the C-53, then a while later with later technology a cargo plane like the An-22, and THEN, being able to use jet engines to build mammoth sized C-5's, or An-225, for resources

like even with a new "career mode" you get a juno jet engine before a prop engine, you slap 3-4 of those engines on a plane, and you can carry quite a bit going well excess of mach... So hearing that we will do resources and what not not having "low tech" in the terms of flight prop engines in space gmae, and alongside aerodynamics is quite frustrating..

i would rather have "weight" on an aircraft if that makes sense, not mach capable props that can do 30g's with ease without any issues, wish there is an option for more difficult options without mods.

 

Grammerly... please

Edited by Stephensan
Grammarly jumbling up my paragraphs
Link to comment
Share on other sites

Why did it get so argumentative over such a pedestrian non-controversial topic? O.o

Anyway, I've played since the early beta versions, and there are some issues that have continually marred the experience:

1. The binary "stick or slide"  mesh friction physics (see: rover wheels, landing legs, etc.)

2. The "broken fuel crossfeed" bug

3. The "craft stuck in landed state even after taking off" bug

4. Bonkers mesh collision physics (ever tried to do anything but the SIMPLEST of EVAs?)

5. Maneuver nodes. Do I need to say any more?

6. The terrible scaling of performance with part number.

I would disagree with some of the other comments regarding the inaccuracy of KSP1's aerodynamic physics. Aerodynamics is incredibly difficult to efficiently simulate on consumer-grade computer hardware, and what KSP1 has done is quite remarkable IMHO. Accurate aerodynamics, can actually run on your PC - pick one.

Also, what's awesome is that almost all of the above things i've mentioned, the KSP2 team have already directly mentioned in the interviews and dev blogs,. for example the "Collisions" dev blog.

Link to comment
Share on other sites

52 minutes ago, Rutabaga22 said:

Please stopa arguing about this stuff, if you want to talk about it make your own thread, this was supposed to be about your ksp1 issues not an argument about others

the last few comments have been in-depth explanations of specific elements of KSP that people want to see fixed in KSP 2, neither of them were invalidating or discounting anyone's complaint. Discouraging it wouldn't be helpful. 

10 minutes ago, samhuk said:

The "craft stuck in landed state even after taking off" bug

Bugs like these are the reason I can't play KSP for more than a few hours at a time or make consistent progress on my saves, having to revert launches, load craft many times (jumping bug), or spend time changing the save file to stop things from breaking is sometimes too much for casual play. 

Link to comment
Share on other sites

58 minutes ago, Rutabaga22 said:

Please stopa arguing about this stuff, if you want to talk about it make your own thread, this was supposed to be about your ksp1 issues not an argument about others

I'm... somewhat confused. If the thread isn't for discussing issues with KSP 1 we want to see the devs avoid while creating KSP 2, then what is it for?

Link to comment
Share on other sites

Also important to recognize Bej and I have bantered for many years now and while many perceive his writing style as abrupt or rude he doesn’t mean it that way. He’s just being direct. If my responses seem rude its because Im also being direct so that we can talk plainly.  This aero issue is real and not being an expert on the subject I super appreciate DA299’s breakdown of the specifics. I don’t think anyone is actually angry here. 

Edited by Pthigrivi
Link to comment
Share on other sites

4 hours ago, samhuk said:

I would disagree with some of the other comments regarding the inaccuracy of KSP1's aerodynamic physics. Aerodynamics is incredibly difficult to efficiently simulate on consumer-grade computer hardware, and what KSP1 has done is quite remarkable IMHO. Accurate aerodynamics, can actually run on your PC - pick one.

You can disagree, but accurate aero is not really hard to simulate. Nobody is asking for a full-fledged CFD simulation.

You can play with FAR if you want to know what I'm talking about. It has no performance penalty whatsoever that I can notice, and actually even runs better for large part count crafts, due to its voxel based calculations. And it is actually a reasonably accurate model of real life aero(~85%) and most importantly, is not hacky.

I have no issue with KSP aero not being realistic, but it should not be hacky. Rest, you can agree to disagree.

Link to comment
Share on other sites

As others have mentioned, I'd really like to be able to do things once you get somewhere. Also being able to set up functional outposts that allow you to do more things once you get somewhere. With all of the stuff we've seen with colonies and such I feel pretty good about this being adequately addressed. I'm also excited to see how mods will expand even more on this.

Something that I don't think anyone else has mentioned: rate based joystick control mode, as opposed to impulse based control that KSP (and orbiter back in the day) used. I'm fairly certain that this is how historical spacecraft fly-by-wire hand rotational controllers worked. I know I've seen doc's that support the use of that mode in Gemini. I don't know about Apollo and Shuttle but I'd guess they also had that mode.

For those wondering, it works like this: Joystick (Rotational Hand Controller) pitch back 50% corresponds to a pitch rate of 3 deg/sec. Those numbers are made up but you get the idea. Returning the stick to 0 would null your rotation rate. It's a little thing, but would be fun to "fly" a spacecraft using joystick in that way. 

I have zero expectations that such a mode would actually be implemented.

As a side note, I'd love to see 3D mouse support carried over to KSP2. It's a real niche thing, but is kinda fun. I think it'd be extra fun to use with rate based controls; really give you that feeling of "flying" a spacecraft.

Link to comment
Share on other sites

18 hours ago, Bej Kerman said:

I'm... somewhat confused. If the thread isn't for discussing issues with KSP 1 we want to see the devs avoid while creating KSP 2, then what is it for?

 

18 hours ago, t_v said:

the last few comments have been in-depth explanations of specific elements of KSP that people want to see fixed in KSP 2, neither of them were invalidating or discounting anyone's complaint. Discouraging it wouldn't be helpful. 

Bugs like these are the reason I can't play KSP for more than a few hours at a time or make consistent progress on my saves, having to revert launches, load craft many times (jumping bug), or spend time changing the save file to stop things from breaking is sometimes too much for casual play. 

As @Pthigrivi said the discussion seemed argumentative to me, but people were just being direct, sorry.

Link to comment
Share on other sites

I would second everyone who said aerodynamics!... It would be nice to see something like Ferram Aerospace mod but with a better User Interface integration :).

I would add that a DLC with N-Body dynamics with a better User Interface than Principa mod would be really nice too!.

Edited by Dinlink
Link to comment
Share on other sites

Please. Decent propellers. I want them to automatically adjust pitch and whatnot. We don't manually have to assign elevators or ailerons to KAL control groups and try to get all the orientations and whatnot correct, so why propellers? That's the reason I never use them

Link to comment
Share on other sites

×
×
  • Create New...