KSP 2.0 why we need it and what it needs

[Guest]

20 hours ago, Renae said:

Much of what needs to be done, could technically be done with the existing, but with the limitations of DX9 places on both the game and the PC's running it.  I have a i5 with a 1080Ti and plenty of Ram, yet even I run into part count slow down.

The slowdowns are physics related. DX12 wouldn't help with it at all.

[Renae]

3 hours ago, Brikoleur said:

The slowdowns are physics related. DX12 wouldn't help with it at all.

Actually if the GPU was handling more of the graphics the CPU would have more resources, thus, yes it in fact would.  Besides, most of the more modern (since KPS was first put out there) GPU's do a lot better with physics modeling so..)

3 hours ago, Brikoleur said:

My reactions/votes:

1. abstain
2. no
3. no
4. no
5. abstain
6. no
7. no
8. no
9. no
10. yes
11. yes but not a priority (the current one is passable but stock FAR would be nice)
12. no
13. yes (but already being addressed in Breaking Ground)
14. no
15. yes
16. no
17. no
18. no
19. no
20. strong no, keep Kerbin demilitarised!

What you're describing sounds like a pretty cool game but it wouldn't be KSP anymore.

A visual overhaul with stock atmospheric effects would be nice, better atmospheric physics would be nice, and somewhat more structured campaign missions would be nice. None of these would be high priorities for me however. 

My priorities for KSP 2.0 would include:

1. Optimisation for scalability, e.g.:
   • dynamic autowelding which would limit the number of physics parts per craft to a value the engine can easily handle (I'm sure this could be done without materially affecting gameplay)
   • optimisations for landed craft and bases -- remove them from physics calculations when they're stationary and not interacting with anything so they don't gum up the framerate
2. Better surface interactions, notably:
   • stationary craft stay stationary and don't slide or jump around (fudge it if necessary)
   • ground effect (fudge it if necessary)
3. More procedural parts, notably:
   • fuel tanks -- currently we have so many it's starting to be tricky to find the right one in the set; these could be collapsed into just one tank per diameter allowing you to specify the length (capped by your tech level)
   • wings -- wing assembly from elements is fiddly and clunky, a few simple procedural wings would help a lot
   • parachutes -- would be nice to be able to optimise them for the craft
4. More planets -- a stock OPM basically
5. Some near-future tech way at the end of the tech tree -- spaceborne nuclear reactors for example

Point 20 was just an example of real world Space Programs being heavily influenced by military needs driving the science and tech that later had civilian applications.

Otherwise I like your points, thank you for posting.

 

 

[Kerbinchaser]

16 hours ago, The_Cat_In_Space said:

And for people who run the game on low end PCs, increasing the graphics can make the game nigh unplayable,  causing people to uninstall or just delete the game. I'm fine with KSPs graphics at the moment, they look good. And if people want better graphics, a lot of mods are available that do just that

Exactly. 

[The_Rocketeer]

19 hours ago, The_Cat_In_Space said:

And for people who run the game on low end PCs, increasing the graphics can make the game nigh unplayable,  causing people to uninstall or just delete the game. I'm fine with KSPs graphics at the moment, they look good. And if people want better graphics, a lot of mods are available that do just that

KSP 1.whatever would still exist for those players. KSP 2 would be a sequel.

[pljfpv]

Not sure man. KSP is a simulator above all else. Freedom to simulate is worth much more to me then fancy graphics. I would rather be able to play it on both my gaming rig and my corporate laptop with integrated video, then only on my gaming rig.

All the other stuff you mention seems to integrate mods above anything else .. so is actually a better argument for an add-on combining the most popular, theme grouped mods into an official and supported release, if the mod-owners agree that is.

[kirmie44]

(This is a copy from another thread of an idea I had for a career overhaul, I was thinking DLC but new game works too)

Intro: 
     So hear me out. The best way to get people back in the game is to make something new. Although I'm loving that they are making mods like infernal robotics a supported function of the game, it doesn't change the fact that a lot of us have been using those features for years. Same with the deployable experiments. everything added since the late stages of early access have been mods with the exception of their first dlc (which felt more like a groundwork than an actual feature in my book). With that said, there is one area of the content market that hasn't been touched, career mode and science mode. Besides for the rearrangement of the tec tree and the addition of a few new missions here and there, not much has changed. I've also thought the way these features have been implemented seemed a little shallow. What I want to suggest is a new overhaul for these two modes. 
The first part of this is looking at how it works in the real world. 

Science: 
(Continuing the comment the next day. Thought I would post it up here to avoid confusion) 

Problems with science: 
     In real life, there isn't the direct correlation seen in ksp between science as we see it in the game and the advancements made in technology. The relationship in real life is 90% of the time closer to finding new problems we have to solve. This is never a problem in ksp because we are given all the information about all the star, all the planets and everything in between. what is the point of bringing a barometer to a planet where you already know everything about its atmosphere? In the real world, we had tests to see if space was a vacuum, we had to run many tests to see if we could pressurize a capsule, and we had little idea about any of the stuff we know until we sent something there to test it (Think about our recent visit of Pluto). 

Problems with technological advancements: 
    The other problem with how career mode/science mode was made was almost systemic. I feel like what was implemented wasn't there original gameplan. I've come to this conclusion because of how little their parts actually fit into the tec tree idea they have implemented. The tec in the game was not designed to go into a tec tree. For the most part, the "tec level" is based on the size of a part and not actually how much knowledge was needed to build it. What I want to see is the implementation of iterative designs. 

My solution: 
    Basically, you start out with a lot of junk parts. These parts look bad, don't work well, probably fail in environments outside of Kerbin sea level, and for good measure, have a base fail chance. Along with taking missions to study your solar system (of which you know nothing about), you would also take missions from your science team to help them improve your parts. unlike the test missions given off now. These missions would give you credits towards upgrading our parts. Along with these missions to give you credits. there would be general things like allowing engines to work in a vacuum, water, restart, throttle, etc. (All things we had to learn, and are still learning, in the field). This science would be collected by doing things like the barometric test and temperature tests we see in KSP today. 
A good way to think of this is the first capsule ou send to space should always be empty. It needs to be tested first. Without things like this, this game is missing out on some of the best parts of our job in exploring the world and beyond! 
  
Career: 
Problems with Career & the suggested improvements: 
    The main problem with the career is how lightweight it really is. There is very little to it, and it's hard to actually mess up. This is fine for some, but I think a lot of us were originally hoping for a bit more. A lot of what makes career modes great is the management. You need to be able to see your expenses and take risks. Squad tried to implement this before launch with that one building no one ever clicks on but it really doesn't work that well. a career mode should revolve around time and funds managment. You should see how much it costs to maintain a team of engineers as well as how long it takes to build a rocket. An alarm clock and construction time should be implemented. You should be able to speed up the construction of a craft with part failure chance going up. you should be able to fire staff to lower your costs and hire them as needed. There should be a monthly/yearly expense report. These features would go a long way to making career mode a better mode. A few more ideas would be including things like kerbal konstructs, allowing you to spend money to build more launch pads as well as just more infrastructure.  I don't like that particular mod too much because of how messy it looks but a lighter version of it could go a long way. The main point I would say is to look at a game like X-COM. It has a solid management system that would work beautifully in a game like KSP. even the kerbal upgrade sstem could be improved along the same lines. 

[electricpants]

after reading the OP, i have this to say:

 

overall i'd say that, yes, some improvements should be made to the stock game, but a whole new ksp sequel seems pretty unnecessary in my eyes.

 

also, some advice to the OP:

use mods. just... use mods. FAR, scatterer, EVE, and OPM all seem good for what you want.

 

just throwing my two cents in, i guess.

[Renae]

On 5/25/2019 at 2:30 PM, electricpants said:

after reading the OP, i have this to say:

 

overall i'd say that, yes, some improvements should be made to the stock game, but a whole new ksp sequel seems pretty unnecessary in my eyes.

 

also, some advice to the OP:

use mods. just... use mods. FAR, scatterer, EVE, and OPM all seem good for what you want.

 

just throwing my two cents in, i guess.

The mods make some good, but the engine is sputtering, flailing.  It's time to upgrade.

[pandaman]

On 6/6/2019 at 5:12 PM, Renae said:

...but the engine is sputtering, flailing.  It's time to upgrade.

Aahhhh the old saying - 'The grass is always greener on the other side' - springs to mind.

I don't know much about the different 'game engines' to be fair, but even if they did move it to another engine what's to say it would be any better overall?

The different engines  will each have their own strengths and weaknesses and, naturally, the Unity specific issues would not exist with a different engine, but what shortcomings would the new engine have that Unity doesn't?   And how long before there is a call to ditch the 'new' engine in favour of another because it is not up to the job?

[Incarnation of Chaos]

Everyone keeps asking for KSP to move to another engine; and i would urge these people to look at the other options available and see how they handle complex physics simulations. Because the majority don't do much better; games fake a lot of "Realistic" physics events using baked explosions/sprites, less particle effects and hard-coded flight models precisely because actually doing an accurate physics simulation crushes CPU cycles. Now there are multithreaded physics approaches; but they're difficult to intergrate into an existing program and result in major bugs that are a headache to chase down and squash. DX 12,11,10 won't make a difference; DX is just a graphics API. And KSP is not graphically that intensive; the majority of slowdown is from physics which DX won't be involved in. 

There are likely ways to do general physics on GPU's; but that's the realm of CUDA/Open GL and i doubt there's many people who would code a realistic physics sim for a game. And even if they did; you just move the bottleneck somewhere else. You would end up needing multiple GPU's for the best performance.; personally i don't mind that since i'm using 2X Vega 56. And people on lower-spec machines could use an IGPU as a "Physics Accelerator" with something else rendering graphics; but all of this is very easy to say. The difficulty in coding it and balancing performance vs an accurate simulation would be massive; which is the same situation we're in now. 

So i'm not against a KSP 2.0; but the idea that Unity is the primary issue is a red herring. Unreal, Crytek, Lumberyard or an in-house engine would not handle it much better; the only realistic way to speed physics up signifigantly would be using CUDA/OpenGL plugins to make the GPU handle it. Which would need to be basically custom code;  also all other major engines have their own jank and bugs; you won't see many existing bugs go away with another engine. 

[sturmhauke]

On 5/21/2019 at 12:44 PM, Renae said:

16.  Put a Ceti Alpha Star with planets as end game.  Something 4-6 LY out, that should take time to reach in the science tree.  That, would be uber cool and allow for amazing sci-fi exploration.

THIS IS CETI ALPHA V!!!

https://i.imgur.com/cLUZpa6.mp4

[IncongruousGoat]

7 hours ago, Incarnation of Chaos said:

Unreal, Crytek, Lumberyard or an in-house engine would not handle it much better; the only realistic way to speed physics up signifigantly would be using CUDA/OpenGL plugins to make the GPU handle it.

Except the rigid-body physics KSP does can't be parallelized. It's strictly single-thread. Trying to run it on a GPU is going to make things slower (a lot slower), not faster. Not to mention the fact that using CUDA for anything is a huge pain, as well as only working on newer Nvidia chips.

To address the OP: We do not need KSP 2.0. There are plenty of things we would all like to see in the game, but it's fun as it is. I know I've certainly sunk untold thousands of hours into it, going all the way back to 0.23.5. Plus, it wouldn't sell very well. We already have KSP, after all, and it's hard to see a way to improve it enough to get millions of people to buy a sequel that would end up boiling down to a more polished version of the base game. Except it wouldn't be, because development is hard even when you're not abusing the engine. There's no reason to think that KSP 2.0 will be any less buggy or any more polished than KSP 1.0. So let's all stop griping about all the things we wish KSP did or had and get back to enjoying the game, alright? And if you're really dissatisfied with the way things are, there are always mods. No, I'm not being flippant; I'm dead serious. That's what they're there for.

Unless you're a console player. In that case, complain away. You've got legitimate grievances, not this "oh, it doesn't look as perfectly shiny as it could, boo hoo" nonsense. For pity's sake, people.

[Incarnation of Chaos]

59 minutes ago, IncongruousGoat said:

Except the rigid-body physics KSP does can't be parallelized. It's strictly single-thread. Trying to run it on a GPU is going to make things slower (a lot slower), not faster. Not to mention the fact that using CUDA for anything is a huge pain, as well as only working on newer Nvidia chips.

To address the OP: We do not need KSP 2.0. There are plenty of things we would all like to see in the game, but it's fun as it is. I know I've certainly sunk untold thousands of hours into it, going all the way back to 0.23.5. Plus, it wouldn't sell very well. We already have KSP, after all, and it's hard to see a way to improve it enough to get millions of people to buy a sequel that would end up boiling down to a more polished version of the base game. Except it wouldn't be, because development is hard even when you're not abusing the engine. There's no reason to think that KSP 2.0 will be any less buggy or any more polished than KSP 1.0. So let's all stop griping about all the things we wish KSP did or had and get back to enjoying the game, alright? And if you're really dissatisfied with the way things are, there are always mods. No, I'm not being flippant; I'm dead serious. That's what they're there for.

Unless you're a console player. In that case, complain away. You've got legitimate grievances, not this "oh, it doesn't look as perfectly shiny as it could, boo hoo" nonsense. For pity's sake, people.

Then it's not even the engine or physics at fault; it's the implmentation of the physics in game. Also i only mentioned CUDA because i didn't want to catch too much flak; OpenGL or some other platform agnostic solution would be the "Best" in terms of compatiblity. Still would be a major pain to code though; as with any of this. 

[kerbiloid]

Spoiler

The Mozilla/Firefox release history:

2    Early versions

• 2.1    Phoenix and Firebird
• 2.2    Firefox
• 2.3    Firefox 1.5
• 2.4    Firefox 2
• 2.5    Firefox 3
• 2.6    Firefox 3.5
• 2.7    Firefox 3.6
• 2.8    Firefox 4

3    Rapid releases

• 3.1    Firefox 5 through 9
• 3.2    Firefox 10 through 16
• 3.3    Firefox 17 through 23
• 3.4    Firefox 24 through 30
• 3.5    Firefox 31 through 37
• 3.6    Firefox 38 through 44
• 3.7    Firefox 45 through 51
• 3.8    Firefox 52 through 59
• 3.9    Firefox 60 through 67
• 3.10    Firefox 68 through 73

Squad should just remove "1.", and start the "rapid" numeration. Then we already have KSP 7.1.

Unlikely Squad will rewrite such specialized game from scratch, and definitely there is no need in several years of players frustration while the KSP 2.0 further debugging.

(Unless some SimpleRockets makes them do this, but this looks unlikely, as probably all interested nerds are already here).

Edited June 11, 2019 by kerbiloid

[Psycho_zs]

Optimizations in general and physics overhaul in particular.

I'm hesitant to buy a computer with over the top last gen CPU just to play KSP.

And my i7-6500U can barely cope with it.

[ILoveStars]

New star system

[Chel]

55 minutes ago, ILoveStars said:

New star system

So for two? Kerbol and another one? Or just replace Kerbol with something else

[ILoveStars]

On 6/18/2019 at 9:53 AM, The_Cat_In_Space said:

So for two? Kerbol and another one? Or just replace Kerbol with something else

And another one.

 

[Yeet_TheDinosaur]

Hi I'm a noob wandering around the forum and this is my opinion as a KSP player.

This is my idea about KSP 2.0. I generally agree with your idea, and here it is.

1. I believe that the KSP's physics engine is not the best. The physics engine is probably the main reason fo crashes, krakens and slow loading times. If squad creates or changes the KSP physics engine by 2.0, it would be an extreme benefit to both us and Squad.

2. More interesting stuff. Most of your opinions about changing is related with how KSP feels like a half-made game. It's a great game, but it could've been better. More science. Make the structures useful. More structures on the planets. Mini games. There could have been tonnes of improvements done, but only a few has been ada[ted by squad. 

3. Cooperation with modders. All the modders of KSP creates BREATHTAKING program for free. But to me, it feels like squad is believing that the whole community is a duck laying goldeneggs. No, there should be partnershi[

I think that sums up my idea about KSP. It's sucha great game, bujt it is limited by the physics engine, unity and the lack of more interesting stuff. Also, the 8th anniversary of KSP's release is 4 days away right now. I expect Squad to have something up their sleeves that day(24th of June, 2019). 

[ZooNamedGames]

As I say with most "we need KSP2" threads I'll push the same message I have since 2016 since despite 3 years passing, nothing has changed in this regard-

There is nothing that a sequel can bring that KSP1 can't have done. KSP2 would suggest it's a sequel but a sequel is a progression in the story or plot... there isn't one with KSP. KSP2 would just be an overhauling update. Which, like all KSP updates, would be buried and forgotten as new content and releases makes the previous obsolete.

But to address the original post-

• New engine for KSP would be nice but KSP has already upgraded once to Unity 5 (from 4). Why couldn't it do it again? Not to mention, no game engine will ever be enough for KSP. KSP is a very unique game. Trying to render whole planets while in orbit with other moons rendered, following on rail paths while also having to calculate orbital paths and navigation as both the targetted vessel and every other in the game is following calculated trajectories through the solar system. There's a reason why there's only 1 game like KSP (Simple Rockets 1/2 doesn't nearly match the scale of KSP as they limit the game to 2D making inclined orbits and those fun orbits impossible).
• I think Squad is naturally trying to keep to a routine update schedule, but hiccups in development, QnA, and other background issues keep pushing it off routine.
• A Squad-built CKAN is a lot of work. Maintaining the database, making it functional, making it safe (it is downloading from the internet, they don't want to be responsible for someone uploading a virus and getting blamed for sharing it), making it capable of identifying KSP versions, which OS, making sure mods are for the matching OS, matching version. Plus this would be a massive tease for console players who don't have any mods to choose from.
• MJ stock/not stock is a debate that's been around since KSP first became available to the internet. Ain't touchin that topic, but personally I like it in it's mod status.
• There's a mod for first person view already, though I wouldn't mind it being absorbed into stock. Though this doesn't warrant an entirely new game.
• An option to spawn from the Astronaut Complex would suffice this suggestion well. Alongside a new launch tower part that allows you to be warped to the top of exit upon selection or approach (leave how that would work to Squad) Spawn with a prebuilt Astronaut Shuttle Bus and drive up to your rocket, step up the launch tower, reach the top. Whamo. Suggestion met without a new game.
• Though I've got no information to support my claim- my experience tells me that the scene you're referring to was actually Hollywood'd as usual since I doubt NASA would've knowingly stuck crews in a rocket with another one launching. The crews were likely getting ready for their mission (since the Gemini 8 launch was nearly 7 hours after their Agena TV launched) and not aboard the Gemini spacecraft as depicted. But addressing your request- this could easily be met by having multiple launchpads. Rocket 1 on Launchpad A and Rocket 2 on Launchpad B. Again, not needing a new game.
• You can save subassemblies and use them. Though a sorting or catagorization system of subassemblies would be nice to sort probes, launch vehicles, tugs, landers, etc.
• Improved career has been a necessity since it was first introduced as many people have noticed. It was a great way for Squad to test the waters and get their feet wet but by now Squad's feet have become wrinkly from the water... and it's starting to stink. I would pay to make sure that KSP1's career system was fixed rather than giving up and starting fresh with a new game.
• KSP flight mechanics are a tricky matter. If it's too realistic, it'll be a turn off for more casual players as they can't master the learning curve, but meanwhile it is very problematic. Hence why it's constantly being tuned and adjusted... granted KSP can't change the aerodynamic properties either without problems since that can upset the community who now have broken planes or planes that are vastly inferior. Something for KSP2 sure, but not beyond KSP1's scope. Especially since FAR has been fixing KSP's problems since very early on.
• Go figure that one of the most complicate spacecraft ever built by mankind (making the lunar lander seem like a walk in the part by comparison) would also be the most difficult to replicate in KSP. They've already thrown us a massive bone with the Vector. A better OMS style part (the puffs are nice but their ISP/thrust makes navigation a challenge with a piddly ~200m/s with my designs, though maybe someone else can make it work better) and flat sided variants to the Mk3 parts (since that's clearly an option as we see with the new lander/rover variant of the lander can) would also help, not to mention larger mk3 wings but other than that, we've already got quite a huge leg up all things considered.
• I've been talking about things to do on planets for a few years now. So I agree. Though this isn't something to be left to a whole new game. Plus Squad appears to be considering that problem, as they attempt to give us more to do with Breaking Ground. There's a good possibility they'll continue to add more to do as time goes on (Thank you Squad).
• Squad used to have prereleases of updates for players and mod developers but it created some problems so they stopped doing it. Though I think it would be fine if they had an optional prerelease build would do the job just fine. Especially since there's a lot of smaller mod makers that would suffer if they too didn't get include (a mod is a mod, no matter how small).
• Clouds are another age old request. Handled well by mods but always offering Squad the challenge of making it scalable by settings, togglable, and functional with the stock game which is a taller order than it sounds.
• Even at maximum timewarp (100000x) and at the speed of 17km/s would takes ages to reach. Not to mention then you're reaching into far future developments and that is far better suited by other games.
• Funding does matter. Resources? Well I've always wanted the revival of the "limited number of parts" mechanic. But that's just me. As to multiple launches a day could simply by replicated by a maximum number of launches per day mechanic. That said all of this doesn't necessitate an entirely new game.
• To be fair, there's nothing stopping you from launching vehicles, or assembling them from space stations right now. Just launch and dock stuff in segments.
• A more realistic and developed Kerbin is great, and I'd love it. But structures and things on the terrain hog computer resources. Adding another hurdle for lower end PC players (which I used to be, so I definitely sympathize). A new game would only make that slightly more feasible since even if built from the ground up, you're still having to render an entire planet when in orbit along with either a texture of those structures, sprites of them, or min models, or something else. It still taxes computer power.
• Perhaps that hitting things could be better fit into a less militaristic approach. You're not wrong about space history being very militarized, but KSP isn't supposed to be based off of any one nation's history or backstory. KSP is entirely it's own history. One that can be peaceful or militarized based on the playstyle and interests of the one playing it.

 

That took a lot of words. But most of those things don't require an entirely new game to solve.

Edited June 24, 2019 by ZooNamedGames

[Renae]

On 6/24/2019 at 10:20 AM, ZooNamedGames said:

As I say with most "we need KSP2" threads I'll push the same message I have since 2016 since despite 3 years passing, nothing has changed in this regard-

There is nothing that a sequel can bring that KSP1 can't have done. KSP2 would suggest it's a sequel but a sequel is a progression in the story or plot... there isn't one with KSP. KSP2 would just be an overhauling update. Which, like all KSP updates, would be buried and forgotten as new content and releases makes the previous obsolete.

But to address the original post-

• New engine for KSP would be nice but KSP has already upgraded once to Unity 5 (from 4). Why couldn't it do it again? Not to mention, no game engine will ever be enough for KSP. KSP is a very unique game. Trying to render whole planets while in orbit with other moons rendered, following on rail paths while also having to calculate orbital paths and navigation as both the targetted vessel and every other in the game is following calculated trajectories through the solar system. There's a reason why there's only 1 game like KSP (Simple Rockets 1/2 doesn't nearly match the scale of KSP as they limit the game to 2D making inclined orbits and those fun orbits impossible).
• I think Squad is naturally trying to keep to a routine update schedule, but hiccups in development, QnA, and other background issues keep pushing it off routine.
• A Squad-built CKAN is a lot of work. Maintaining the database, making it functional, making it safe (it is downloading from the internet, they don't want to be responsible for someone uploading a virus and getting blamed for sharing it), making it capable of identifying KSP versions, which OS, making sure mods are for the matching OS, matching version. Plus this would be a massive tease for console players who don't have any mods to choose from.
• MJ stock/not stock is a debate that's been around since KSP first became available to the internet. Ain't touchin that topic, but personally I like it in it's mod status.
• There's a mod for first person view already, though I wouldn't mind it being absorbed into stock. Though this doesn't warrant an entirely new game.
• An option to spawn from the Astronaut Complex would suffice this suggestion well. Alongside a new launch tower part that allows you to be warped to the top of exit upon selection or approach (leave how that would work to Squad) Spawn with a prebuilt Astronaut Shuttle Bus and drive up to your rocket, step up the launch tower, reach the top. Whamo. Suggestion met without a new game.
• Though I've got no information to support my claim- my experience tells me that the scene you're referring to was actually Hollywood'd as usual since I doubt NASA would've knowingly stuck crews in a rocket with another one launching. The crews were likely getting ready for their mission (since the Gemini 8 launch was nearly 7 hours after their Agena TV launched) and not aboard the Gemini spacecraft as depicted. But addressing your request- this could easily be met by having multiple launchpads. Rocket 1 on Launchpad A and Rocket 2 on Launchpad B. Again, not needing a new game.
• You can save subassemblies and use them. Though a sorting or catagorization system of subassemblies would be nice to sort probes, launch vehicles, tugs, landers, etc.
• Improved career has been a necessity since it was first introduced as many people have noticed. It was a great way for Squad to test the waters and get their feet wet but by now Squad's feet have become wrinkly from the water... and it's starting to stink. I would pay to make sure that KSP1's career system was fixed rather than giving up and starting fresh with a new game.
• KSP flight mechanics are a tricky matter. If it's too realistic, it'll be a turn off for more casual players as they can't master the learning curve, but meanwhile it is very problematic. Hence why it's constantly being tuned and adjusted... granted KSP can't change the aerodynamic properties either without problems since that can upset the community who now have broken planes or planes that are vastly inferior. Something for KSP2 sure, but not beyond KSP1's scope. Especially since FAR has been fixing KSP's problems since very early on.
• Go figure that one of the most complicate spacecraft ever built by mankind (making the lunar lander seem like a walk in the part by comparison) would also be the most difficult to replicate in KSP. They've already thrown us a massive bone with the Vector. A better OMS style part (the puffs are nice but their ISP/thrust makes navigation a challenge with a piddly ~200m/s with my designs, though maybe someone else can make it work better) and flat sided variants to the Mk3 parts (since that's clearly an option as we see with the new lander/rover variant of the lander can) would also help, not to mention larger mk3 wings but other than that, we've already got quite a huge leg up all things considered.
• I've been talking about things to do on planets for a few years now. So I agree. Though this isn't something to be left to a whole new game. Plus Squad appears to be considering that problem, as they attempt to give us more to do with Breaking Ground. There's a good possibility they'll continue to add more to do as time goes on (Thank you Squad).
• Squad used to have prereleases of updates for players and mod developers but it created some problems so they stopped doing it. Though I think it would be fine if they had an optional prerelease build would do the job just fine. Especially since there's a lot of smaller mod makers that would suffer if they too didn't get include (a mod is a mod, no matter how small).
• Clouds are another age old request. Handled well by mods but always offering Squad the challenge of making it scalable by settings, togglable, and functional with the stock game which is a taller order than it sounds.
• Even at maximum timewarp (100000x) and at the speed of 17km/s would takes ages to reach. Not to mention then you're reaching into far future developments and that is far better suited by other games.
• Funding does matter. Resources? Well I've always wanted the revival of the "limited number of parts" mechanic. But that's just me. As to multiple launches a day could simply by replicated by a maximum number of launches per day mechanic. That said all of this doesn't necessitate an entirely new game.
• To be fair, there's nothing stopping you from launching vehicles, or assembling them from space stations right now. Just launch and dock stuff in segments.
• A more realistic and developed Kerbin is great, and I'd love it. But structures and things on the terrain hog computer resources. Adding another hurdle for lower end PC players (which I used to be, so I definitely sympathize). A new game would only make that slightly more feasible since even if built from the ground up, you're still having to render an entire planet when in orbit along with either a texture of those structures, sprites of them, or min models, or something else. It still taxes computer power.
• Perhaps that hitting things could be better fit into a less militaristic approach. You're not wrong about space history being very militarized, but KSP isn't supposed to be based off of any one nation's history or backstory. KSP is entirely it's own history. One that can be peaceful or militarized based on the playstyle and interests of the one playing it.

 

That took a lot of words. But most of those things don't require an entirely new game to solve.

I'm aware Hollywood and reality aren't often in alignment, but since this is a game... "Wouldn't it be cool..?"

And yes I know that Kerbal isn't "specific" to, any countries military, Rocket Science started as a military program, acknowledging that as mission steps would be... proper IMHO.  "Launch inert payload on a suborbital trajectory to land within this circle, the closer the better.  This is to aid... space navigation, and that Army General has plans too..."  Just saying.

I do agree that a "REAL" flight sim model would be a powercurve some could not handle, but the one we have is just... broken.  

 

 

 

[Incarnation of Chaos]

On 6/27/2019 at 10:54 AM, Renae said:

I

I do agree that a "REAL" flight sim model would be a powercurve some could not handle, but the one we have is just... broken.  

 

 

 

We don't have a flight sim model at all lul; the way stock KSP computes drag/lift is practically fantasy. And it's not like the "Old" model just couldn't become a toggle in difficulty settings; for those who would rather stick with it.

[Lord Aurelius]

I would love to see a KSP2.0, although most of the things in the OP have already been accomplished via mods.

If I'm thinking of major systems changes that would justify an entirely new game instead of just continued updates, here's what I would like to see:

• Proper game engine for the job
  • I totally agree with this part of the OP. Unity was a reasonable choice when the game first got started, but today the engine has boatloads of problems and hasn't been maintained well at all. Would like to see them make a clean break and use an engine better suited to the game.
• Vastly improved mission planning
  • In RL, missions and maneuvers are often planned out years in advance, and are often planned out before the craft is designed. Would love to see a mission planner where you can pick the date/time of the launch, and plan out all the nodes to get a full dV estimate for the mission even before entering the VAB, along with a mission flight plan that saves the maneuver nodes so you don't have to recreate them manually during flight.
• Improved flight automation
  • Would love to see something like MechJeb made stock, along with kOS (albeit with a simpler interface, maybe like what LEGO Robotics uses), including the ability to run scripts on staging. RL rockets are entirely automated, and a big part of rocket design is also writing these launch programs.
• Ability to control multiple craft simultaneously
  • This is one of the biggest gameplay weaknesses of the current game, and the lack of this feature makes many RL missions (especially involving SpaceX style booster recoveries) just about impossible.
  • Could be solved with the improved flight automation where pilots/command pods can be given autopilot instructions to execute on staging
• Easier vehicle reuse
  • Maybe have a payload part that be given an arbitrary size/shape (or automatically take on the size of the fairing/cargo bay its inside)
  • When testing the rocket, you can put a mass simulator in the payload part to demonstrate the weight capacity of the rocket
  • Once a rocket has been verified for a given payload volume/mass, you can easily click launch rocket, choose a payload that fits in the payload part, and go
• Limited locally hosted coop multiplayer
  • Would be awesome to have the ability to have friends (or just a second PC so you can fly 2 craft at the same time) connect to the game with the ability to control in-flight craft or potentially launch new ones. I envision this as being hosted locally on one PC, with non-host players being more limited in what they can do (i.e. host has to provide permission for them to fly/launch craft).
• Create custom parts in-game
  • Similar idea to procedural parts, but more robust including the ability to order a new engine, probe core or crew capsule design
  • Parts would follow the laws of physics and would take lots of time/funds to research
• Ability to randomly generate solar systems with a seed (and store them per-save)
  • Would provide even more replayability, could always use the default seed to get the default solar system
  • Also have support for custom star systems so something like RSS could be stock
• More RL gameplay considerations
  • Life Support
  • Part reliability
  • Ability to construct additional launch sites and tracking stations
• Fully implemented part upgrades
  • Keep each part relevant for the entire game
  • Makes game balancing better without parts that are either OP at the beginning or useless after the first flight (looking at you, Flea).
• Upgraded Kerbal Inventory System
  • We've got it now in Breaking Ground, want to see a lot more done with it like carrying parts and tools
• Tutorial Overhaul
  • With all the proposed changes, would want a robust tutorial explaining all the parts of the game, especially given the current sad state of tutorials in most games today (including current KSP)
• Overhauled career with greatly improved management considerations
  • New tech tree starting with probes, and with much better part progression (current tech tree looks like it was generated via dartboard or RNG)
  • New science system that decouples science points from parts research, and overhauls experiments to fix many of the head-scratching inconsistencies (crew reports aren't per-biome, but EVA fly-over reports are?!? same for silly transmission science point losses for what should just be a number value from a sensor), and adds some new experiments (where are our cameras?)
  • Make something like Rocket Construction Time stock (no more rockets appearing out of thin air)
  • Better Kerbonaut hiring (pay money to train them instead of them magically learning things in space, major rep losses if they die)
  • Better funding (regular budget, varies based on rep, ability to get fired if rep drops too low)
  • Story missions similar to what NovaSilisko proposed a long time ago and started on with some of the easter eggs, and also the anomaly surveyor contracts (with some neat rewards as well)
  • Much better and more realistic contracts (i.e. you're given a prebuilt payload, bring it to the specified coordinates or orbit, or a set of requirements to design and build a craft that can do X).
• General graphics/model/texture/audio/UI improvements and polish
  • Especially on the audio side, would love to have per-planet music
  • Procedurally generated cities/towns on Kerbin (with buildings to crash into, and negative rep if you do so make sure your rocket has a range safety system)
  • Clouds and weather on appropriate planets
  • Probe IVA (camera view, or in some cases just telemetry
  • Mission control scene with lots of big screens (one shows camera footage of the launch, another shows telemetry and orbit info, all are customizable, etc)
  • More interesting things to find in general

Anyways, just some random thoughts I've had for awhile regarding KSP and how to make it better

[Renae]

On 6/30/2019 at 12:40 PM, Lord Aurelius said:

I would love to see a KSP2.0, although most of the things in the OP have already been accomplished via mods.

If I'm thinking of major systems changes that would justify an entirely new game instead of just continued updates, here's what I would like to see:

• Proper game engine for the job
  • I totally agree with this part of the OP. Unity was a reasonable choice when the game first got started, but today the engine has boatloads of problems and hasn't been maintained well at all. Would like to see them make a clean break and use an engine better suited to the game.
• Vastly improved mission planning
  • In RL, missions and maneuvers are often planned out years in advance, and are often planned out before the craft is designed. Would love to see a mission planner where you can pick the date/time of the launch, and plan out all the nodes to get a full dV estimate for the mission even before entering the VAB, along with a mission flight plan that saves the maneuver nodes so you don't have to recreate them manually during flight.
• Improved flight automation
  • Would love to see something like MechJeb made stock, along with kOS (albeit with a simpler interface, maybe like what LEGO Robotics uses), including the ability to run scripts on staging. RL rockets are entirely automated, and a big part of rocket design is also writing these launch programs.
• Ability to control multiple craft simultaneously
  • This is one of the biggest gameplay weaknesses of the current game, and the lack of this feature makes many RL missions (especially involving SpaceX style booster recoveries) just about impossible.
  • Could be solved with the improved flight automation where pilots/command pods can be given autopilot instructions to execute on staging
• Easier vehicle reuse
  • Maybe have a payload part that be given an arbitrary size/shape (or automatically take on the size of the fairing/cargo bay its inside)
  • When testing the rocket, you can put a mass simulator in the payload part to demonstrate the weight capacity of the rocket
  • Once a rocket has been verified for a given payload volume/mass, you can easily click launch rocket, choose a payload that fits in the payload part, and go
• Limited locally hosted coop multiplayer
  • Would be awesome to have the ability to have friends (or just a second PC so you can fly 2 craft at the same time) connect to the game with the ability to control in-flight craft or potentially launch new ones. I envision this as being hosted locally on one PC, with non-host players being more limited in what they can do (i.e. host has to provide permission for them to fly/launch craft).
• Create custom parts in-game
  • Similar idea to procedural parts, but more robust including the ability to order a new engine, probe core or crew capsule design
  • Parts would follow the laws of physics and would take lots of time/funds to research
• Ability to randomly generate solar systems with a seed (and store them per-save)
  • Would provide even more replayability, could always use the default seed to get the default solar system
  • Also have support for custom star systems so something like RSS could be stock
• More RL gameplay considerations
  • Life Support
  • Part reliability
  • Ability to construct additional launch sites and tracking stations
• Fully implemented part upgrades
  • Keep each part relevant for the entire game
  • Makes game balancing better without parts that are either OP at the beginning or useless after the first flight (looking at you, Flea).
• Upgraded Kerbal Inventory System
  • We've got it now in Breaking Ground, want to see a lot more done with it like carrying parts and tools
• Tutorial Overhaul
  • With all the proposed changes, would want a robust tutorial explaining all the parts of the game, especially given the current sad state of tutorials in most games today (including current KSP)
• Overhauled career with greatly improved management considerations
  • New tech tree starting with probes, and with much better part progression (current tech tree looks like it was generated via dartboard or RNG)
  • New science system that decouples science points from parts research, and overhauls experiments to fix many of the head-scratching inconsistencies (crew reports aren't per-biome, but EVA fly-over reports are?!? same for silly transmission science point losses for what should just be a number value from a sensor), and adds some new experiments (where are our cameras?)
  • Make something like Rocket Construction Time stock (no more rockets appearing out of thin air)
  • Better Kerbonaut hiring (pay money to train them instead of them magically learning things in space, major rep losses if they die)
  • Better funding (regular budget, varies based on rep, ability to get fired if rep drops too low)
  • Story missions similar to what NovaSilisko proposed a long time ago and started on with some of the easter eggs, and also the anomaly surveyor contracts (with some neat rewards as well)
  • Much better and more realistic contracts (i.e. you're given a prebuilt payload, bring it to the specified coordinates or orbit, or a set of requirements to design and build a craft that can do X).
• General graphics/model/texture/audio/UI improvements and polish
  • Especially on the audio side, would love to have per-planet music
  • Procedurally generated cities/towns on Kerbin (with buildings to crash into, and negative rep if you do so make sure your rocket has a range safety system)
  • Clouds and weather on appropriate planets
  • Probe IVA (camera view, or in some cases just telemetry
  • Mission control scene with lots of big screens (one shows camera footage of the launch, another shows telemetry and orbit info, all are customizable, etc)
  • More interesting things to find in general

Anyways, just some random thoughts I've had for awhile regarding KSP and how to make it better

SQUWEEEEEE they did it! THEY WENT AND DID IT!!!!

[kirmie44]

ITS HAPPENNING OMG LITTLERALLY SO EXCITED