KSP2 EA: Tech tree progression

[Vl3d]

As there has been ample discussions about how to improve the science system (Kerbalism ongoing experiments system, more parts and experiments, telescopes for initial discovery, getting actually useful information for gameplay, interesting and unique biomes etc.), I think it's very important to talk about the thing that I believe it's the greatest problem with stock KSP1: tech progression and design incentives and limitations.

Because we might be getting life support and new gameplay elements like radiation, resources (materials and fuels etc.) in the future, in my opinion the KSP2 tech tree should mirror this from the start by combining and balancing uncrewed space missions with historical progression.

Because I am experimenting with Kerbalism I tend to prefer the structure of Unkerballed Start and Simplex, but Stock also has it's merits. We need incentives to protect the crew, but also use the crew when tech is limited. We need to risk probes for experiments, but protect crew for prestige gain. We need incentives to strive for more advanced tech instead of strapping on more basic boosters. We should have reasons to build small for probes and big for crewed missions.

It's one of the hardest things to balance and get right. How should KSP2 tackle tech progression? What are the merits of some tech trees compared to others?

Edited November 4, 2022 by Vl3d

[Guest]

I'd  like something ore similar to skill trees in RPGs like Skyrim, Fallout or Cyberpunk 2077, allowing players to pick and choose their classes in a fluid way by assigning points to different skill trees.

One for planes, one for manned, one for unmanned, maybe a couple for different fuels and so on.

You choose your own start and how to advance your space program.

My reply to the manned VS unmanned is almost always: why not spaceplanes?

Immagine the history of space exploration if, after the X-15 they continued with spaceplanes and never even developed capsules until much later.

The tech tree doesn't have to be confined to a single view of progression.

A system like this would also greatly improve the replayability of the game.

[Vl3d]

34 minutes ago, Master39 said:

I'd  like something ore similar to skill trees in RPGs like Skyrim, Fallout or Cyberpunk 2077, allowing players to pick and choose their classes in a fluid way by assigning points to different skill trees.

One for planes, one for manned, one for unmanned, maybe a couple for different fuels and so on.

You choose your own start and how to advance your space program.

My reply to the manned VS unmanned is almost always: why not spaceplanes?

Immagine the history of space exploration if, after the X-15 they continued with spaceplanes and never even developed capsules until much later.

The tech tree doesn't have to be confined to a single view of progression.

A system like this would also greatly improve the replayability of the game.

Profound. I would add: why even try to bring crew back?

Edited November 4, 2022 by Vl3d

[Vl3d]

21 minutes ago, Master39 said:

My reply to the manned VS unmanned is almost always: why not spaceplanes?

I would say because of fuel/mass efficiency and use case. Spaceplanes (which are actually also manned craft) are useful for reuse and going into low orbit (meaning you need to transfer crew and have stations), terrible for high speed atmospheric re-entry needed for manned exploration.

Edited November 4, 2022 by Vl3d

[shdwlrd]

I prefer the unlock as you need and/or discover something approach. This way you still have the "this isn't ideal, but still doable" feel when progressing through the game.

[Vl3d]

16 hours ago, shdwlrd said:

I prefer the unlock as you need and/or discover something approach. This way you still have the "this isn't ideal, but still doable" feel when progressing through the game.

I agree with this, sometimes I just want to keep using an older model of a booster instead of upgrading it right away. It all starts with the incentives to build - if it's for exploration, world firsts, creative projects I'm engaged. If it's only to do some experiment to get 3 more science points to unlock another tech node, I already feel like I'm bored.

19 hours ago, Master39 said:

I'd  like something ore similar to skill trees in RPGs like Skyrim, Fallout or Cyberpunk 2077, allowing players to pick and choose their classes in a fluid way by assigning points to different skill trees.

I've been also thinking about a system where we have a certain base Research Speed, we just prioritize (guide) what to research but we need to do certain experiments for each technology branch to speed things up. I feel like kerbal scientists should be the ones that ask for experiments just to be able to research parts (and maybe improve their reliability), and new discoveries (like a new resource) would unlock the visibility of new tech. Something that mimics real life more.

Edited November 5, 2022 by Vl3d

[magnemoe]

On 11/4/2022 at 2:54 PM, Vl3d said:

I would say because of fuel/mass efficiency and use case. Spaceplanes (which are actually also manned craft) are useful for reuse and going into low orbit (meaning you need to transfer crew and have stations), terrible for high speed atmospheric re-entry needed for manned exploration.

Yes, spaceplanes will start suffer in the range between reaching LKO and having infrastructure in space. 
An X-15 is in the ballpark of getting into LKO but its not something you can do an Mun landing with and return. 
Yes people has build spaceplanes for getting huge payloads into LKO but this is way more advanced than slapping together an heavy lift rocket. 

One very obvious thing is that the tech tree will be much deeper than KSP 1, the endgame engines are nuclear thermal, ion and the rapier. 
In KSP 2 hybrid nuclear thermal,  metalic hydrogen, pulse nuclear and fusion is confirmed. 
We seen one new and unusual engine in the engine view, no idea that it is, might be an advanced ion engine. Pretty small and no bell. 
We also has colonies and resources down the line who will have their own long tech trees.
Not sure who engines will be released as early access, probably not fusion but hope for pulse nuclear. 

[mcwaffles2003]

On 11/4/2022 at 9:38 AM, Master39 said:

I'd  like something ore similar to skill trees in RPGs like Skyrim, Fallout or Cyberpunk 2077, allowing players to pick and choose their classes in a fluid way by assigning points to different skill trees.

One for planes, one for manned, one for unmanned, maybe a couple for different fuels and so on.

You choose your own start and how to advance your space program.

My reply to the manned VS unmanned is almost always: why not spaceplanes?

Immagine the history of space exploration if, after the X-15 they continued with spaceplanes and never even developed capsules until much later.

The tech tree doesn't have to be confined to a single view of progression.

A system like this would also greatly improve the replayability of the game.

RP-1 really felt like it did this with it's tech tree where a lot of techs didn't cross over to other techs much, essentially making separate skill trees to go along. To make something like this work that wasn't available in KSP is adding drastically more parts to choose from. In KSP 1 when looking at part selection aside from structural parts, the amount of engines, propellants, etc... were fairly limited. Having more engines with a greater variety of fuel types and procedural parts in the game would really make this kind of system more possible I think. It just requires providing a greater variety of ways to solve a problem.

 

Perhaps stats of procedural parts could vary based on how people go through the tech tree, allowing different maxes and mins to specific variables.

Edited November 5, 2022 by mcwaffles2003

[MarcAbaddon]

I feel discussing the tech tree as a separate topic from science progress (see the other thread) might already be a mistake. Preferably tech tree and science points should be linked together more tightly, rather than have science points as a generic currency. Have specific science experiments performed under specific circumstances allow progress towards specific tech tree nodes. 

Regarding the order of technology, I think a lot of the "not historically accurate stuff" such as rockets before planes and manned before unmanned flight should stay since it's part of the appeal, and this is primarily a game about rockets with planes being a valuable but secondary part. Ideally, now that technology goes a lot further than it did in the past, I'd like to unlock a solid foundation of standard rocketry quite early - maybe up to the current 3rd tier or so. Put fuel cells before solar panels.

Mainly for educational purposes I also would like if it is very transparent when we transition from actual known technology into speculative sci-fi with engine concepts like metallic hydrogen and interstellar travel in general.

[t_v]

For your concerns about splitting threads, we are many threads into the science discussion. I would have linked them in the “science progress” thread to save space and time, but then I realized there was some original stuff with the discussion of real science experiments. I really hope it doesn’t turn into yet another thread discussing the tech tree or science points. 

And this is in the first 1/5 of the total amount of threads (over 50 pages, wow!) so I don’t know how many more there are beyond that. I do remember seeing a few… 

[Vl3d]

I've been reading some old interviews and I believe each colony will have it's own tech level (tech tree tier). I think that's great - you can only build and launch ships for which you have or are able to manufacture components. This couples great with needing material resources for everything.

Edited November 10, 2022 by Vl3d

[theJesuit]

A new perspective would be to have science based on achievements with science experiments as bonuses, a little like the testing contracts now.  So after a certain number of launches or burns with an engine, another gets unlocked.  You can choose to emphasise different branches (like a SIMPLEX model) but be given direction (like contracts) that tell you what to do do unlock the next node, rather than click and repeat science.  This makes science more fluid.

You can set up a science experiment that gives a bonus for the EVA science in a place. such as the Mun, but also, you channel more effort into unlocking certain branches or nodes based on what they are.

 

[Turbo Ben]

I like the tech tree idea, but I never really liked the idea of using experiments to collect science points that are spent like currency.

My suggestion would be for the R&D department at the KSC to generate, well, R&D. You then decide which tech you want to focus this on, and the tech is gradually researched over time. The rate drops to zero over time (to avoid just time warping a few years). Rate is increased by completing experiments. I'd maybe have 2 ways of doing this:

First, one shot experiments such as temperature or pressure readings which provide a rapid rate for a short time.

Second, more long term experiments such as space station labs, orbital scanning satellites or space telescopes providing a lower rate but over a much longer time (think hubble providing data for decades).

R&D could also be allocated to development of unlocked tech, such as improving an engines thrust / ISP or reducing it's mass / resource cost. SpaceX didn't build the first raptor and then call it a day, development is ongoing. This would make it so earlier tech doesn't become as irrelevant late game.

I think this would mimic how technological advancement happens in real life while encouraging the use of stations, outposts and satellites earlier in the game.

[Pthigrivi]

10 hours ago, Turbo Ben said:

I like the tech tree idea, but I never really liked the idea of using experiments to collect science points that are spent like currency.

My suggestion would be for the R&D department at the KSC to generate, well, R&D. You then decide which tech you want to focus this on, and the tech is gradually researched over time. The rate drops to zero over time (to avoid just time warping a few years). Rate is increased by completing experiments. I'd maybe have 2 ways of doing this:

First, one shot experiments such as temperature or pressure readings which provide a rapid rate for a short time.

Second, more long term experiments such as space station labs, orbital scanning satellites or space telescopes providing a lower rate but over a much longer time (think hubble providing data for decades).

R&D could also be allocated to development of unlocked tech, such as improving an engines thrust / ISP or reducing it's mass / resource cost. SpaceX didn't build the first raptor and then call it a day, development is ongoing. This would make it so earlier tech doesn't become as irrelevant late game.

I think this would mimic how technological advancement happens in real life while encouraging the use of stations, outposts and satellites earlier in the game.

I do like this, especially the idea of data processing taking time, but in a way you still have science points, you've just hidden them from the player. The game is still collecting data 'points', and then converting those points into new part unlocks. It would be best to let players know how much a given experimental result was worth in this process so they could plan for their next mission and have a rough idea what the return would be, which is exactly where science points come from. Sometimes I think if they'd named it "data" or "research grants" folks wouldn't have nearly as much trouble with the concept. 

Another thing that came to me is if they're planning to use resources to unlock some technologies (presumably something like processing X tons of metallic hydrogen unlocks the first MH drive, etc) then you kind of already do have some separate science branches. Does it make sense to have "Atmospheric science" and "Surface Science" and other branches that unlock various categories of parts? It's a popular idea, but it does come with some limitations. 
 

[Xtra]

Okay... the Tech Tree... Here is an area that REALLY needs better thought!

I realize this is a ROCKET/Space simulation, but as we all have seen by the many mods and even many of the Science contracts... there are things to do ON Kerbin and in the Air.

So... How is it that we are Flying Rockets before we have basic Wheels to build Rovers?  Basic Bi-plane type aircraft parts?  AND BATTERIES!  OMG... I started all over again today and went to launch... Power was dead in seconds without engines running... We can launch a rocket but we don't have AA batteries???  I'm just saying... 

Can we get Rover and Plane parts early in the Tree before we launch into Space?  Of course if you want to put a few Rocket parts at the start that's cool too.

[pss88]

6 hours ago, Xtra said:

We can launch a rocket but we don't have AA batteries???

"The Jatravartids are small blue creatures with more than fifty arms each, who are therefore unique in being the only race in history to have invented the aerosol deodorant before the wheel."

[Xtra]

7 hours ago, pss88 said:

"The Jatravartids are small blue creatures with more than fifty arms each, who are therefore unique in being the only race in history to have invented the aerosol deodorant before the wheel."

@pss88 That was out of Necessity... The combined body odor from even TWO( Jatravartids would be enough to make a Kerbal passout in 50 seconds or less. 

[pss88]

Well, maybe little green Kerbals wanted to feel pure sunlight (undisturbed by atmosphere) so much they invented rockets before batteries?