A more intuitive tech tree

[tater]

Yeah, it goes with my general impression that Squad has done an absolutely incredible job of creating a solid platform for a space exploration game, but at this point somebody really needs to come in and actually design a cohesive game using all the tools in that platform. Sandbox mode works great because you get to just play around all the great parts and physics freely, but creating a cohesive career game that feels meaningful, player-directed, and satisfying requires another layer of consideration that doesn't seem to be there yet.

This.

There needs to be a vision, or even different visions. "Science" could just as well be an alternate "career" mode, for example.

I know this is a "tech tree" thread, but I think it is a fundamental mistake to treat the TT as something that exists outside the context of science, "contracts," and career since they are so tightly woven together.

Any attempt to mess with the tree without messing with everything else WILL fail, IMO. It'll just be differently grinding.

Tech is NOT developed by doing space science, sorry. That disconnect is a fundamental flaw of the entire tree system. It can be partially a function of this, but I think in a very specific way (certain kinds of science required for certain kinds of improvements).

I'd ideally have a time-based system (a budget to spend every XX kerbin days (I was thinking a 50 day minmus month)), and you assign funds to research. Funding generates research points, and buying tech takes those points, plus possibly specific science missions. So a Hitchhiker part might require a certain amount of science (crew reports) from orbit. Maybe science over time like a lab, but keeping track of time on orbit. Some might require part-testing contracts (sensible ones). ISRU parts might require research and sample returns from any world (asteroids might all count as 1 world for this) where ISRU is to function.

You'd then pick a path to research, and that would sort of drive your program for a while (as you need to wring out issues with that tech). Honestly, there needs to be more "upgrades" for parts. Squad was willing to have astronauts buff parts, why not have research buff parts, instead (improved engine with better Isp, or gimbal range, etc)?

[CaptainKipard]

I'd ideally have a time-based system (a budget to spend every XX kerbin days (I was thinking a 50 day minmus month)), and you assign funds to research. Funding generates research points, and buying tech takes those points, plus possibly specific science missions. So a Hitchhiker part might require a certain amount of science (crew reports) from orbit. Maybe science over time like a lab, but keeping track of time on orbit. Some might require part-testing contracts (sensible ones). ISRU parts might require research and sample returns from any world (asteroids might all count as 1 world for this) where ISRU is to function.

You'd then pick a path to research, and that would sort of drive your program for a while (as you need to wring out issues with that tech). Honestly, there needs to be more "upgrades" for parts. Squad was willing to have astronauts buff parts, why not have research buff parts, instead (improved engine with better Isp, or gimbal range, etc)?

I like this.

[passinglurker]

This.

There needs to be a vision, or even different visions. "Science" could just as well be an alternate "career" mode, for example.

I know this is a "tech tree" thread, but I think it is a fundamental mistake to treat the TT as something that exists outside the context of science, "contracts," and career since they are so tightly woven together.

Any attempt to mess with the tree without messing with everything else WILL fail, IMO. It'll just be differently grinding.

Tech is NOT developed by doing space science, sorry. That disconnect is a fundamental flaw of the entire tree system. It can be partially a function of this, but I think in a very specific way (certain kinds of science required for certain kinds of improvements).

I'd ideally have a time-based system (a budget to spend every XX kerbin days (I was thinking a 50 day minmus month)), and you assign funds to research. Funding generates research points, and buying tech takes those points, plus possibly specific science missions. So a Hitchhiker part might require a certain amount of science (crew reports) from orbit. Maybe science over time like a lab, but keeping track of time on orbit. Some might require part-testing contracts (sensible ones). ISRU parts might require research and sample returns from any world (asteroids might all count as 1 world for this) where ISRU is to function.

You'd then pick a path to research, and that would sort of drive your program for a while (as you need to wring out issues with that tech). Honestly, there needs to be more "upgrades" for parts. Squad was willing to have astronauts buff parts, why not have research buff parts, instead (improved engine with better Isp, or gimbal range, etc)?

well if coders can totally convert and overhaul the tech tree system it would be cool but as for me the best I can do is alter configs(well and texture and model but those don't help here). but you make a point that the availability of contracts is influenced by the tech you unlock as well as the feats you accomplish. Perhaps then an alternative to a parts based intuitive tree, or a technology based intuitive tree, is an activity based intuitive tree which should hopefully make integration with career more seamless if contracts can be rigged to appear with the associated activity that a node is balanced to equip you with the basic parts for (for example an aircraft node (cockpit, engine, fuel fuselage, wings, control surfaces, wheels) would bring about visual survey contracts or a space habitation node (crew can, docking port, RCS port, RCS tank) would bring about station/base building contracts.

[tater]

Yeah, missions (contracts) need to make more sense.

Right now the reward in KSP is new stuff (tech), and the currency to buy it is science points. The contracts deal out funds (and now seemingly less science), but are the same grindy contracts we are used to now---repetitive contracts that often make no sense whatsoever. Testing parts in ways that are idiotic, for example (jets on the Mun). Or we get survey contracts, but however Fine Print did it, they have a group together (sensible), then usually one outlier that makes mission planning pointlessly annoying. I want career to be decent… but it simply isn't.

[sherkaner]

I'd ideally have a time-based system (a budget to spend every XX kerbin days (I was thinking a 50 day minmus month)), and you assign funds to research. Funding generates research points, and buying tech takes those points, plus possibly specific science missions. So a Hitchhiker part might require a certain amount of science (crew reports) from orbit. Maybe science over time like a lab, but keeping track of time on orbit. Some might require part-testing contracts (sensible ones). ISRU parts might require research and sample returns from any world (asteroids might all count as 1 world for this) where ISRU is to function.

You'd then pick a path to research, and that would sort of drive your program for a while (as you need to wring out issues with that tech). Honestly, there needs to be more "upgrades" for parts. Squad was willing to have astronauts buff parts, why not have research buff parts, instead (improved engine with better Isp, or gimbal range, etc)?

Yeah, I agree with Kipard, this is a great idea. It's unfortunately a pretty serious departure for how things are structured now, but I think it's absolutely the comprehensive approach that needs to happen.

[twisteddman]

I agree that the tech tree makes little sense. In my opinion it should start with aviation ---> basic probes ---> manned space flight ---> advanced probes ---> space stations ,lander. This progression make more sense realistically and it makes more sense gameplay wise.The progression would move slower and make planes and probes a necessary part of development/science. astronauts would game some experience through flying planes also before being sent into space. I dont understand why it is not set up this way already. Maybe they just want people who just started to be able to jump right into space for the fun aspect , I dont know. I have been playing this game since the beginning and I have been hoping that they would make it follow a more realistic progression since the first career mode, but it doesnt seem like its ever going to.

[Francois424]

I was hoping you guys where going to pop-in after the little change the tech tree has gotten so far.

Hopefully with the supposedly easy-to-mod tree, we'll get a starshaped tree or something like that.

Personally, I do not like that they put Rapiers and Ions so late in the tech tree. Ions used to be a nice substitute for Neva and orange tanks, now I really will almost never use them O.o;;

[tntristan12]

What I want to know is why does the tech tree need to start at exactly one point? Why can't we pick a starting position that allows us to begin with a focus on either aircraft, manned rocketry, or unmanned rocketry?

[CaptainKipard]

astronauts would game some experience through flying planes also before being sent into space.

This is a good idea.

What I want to know is why does the tech tree need to start at exactly one point? Why can't we pick a starting position that allows us to begin with a focus on either aircraft, manned rocketry, or unmanned rocketry?

Technically it doesn't. It's a design decision by Squad. Sherkaner made a tree for the previous version with multiple starting points using the Techmanager mod.

I'd like more starting points like in the image in the OP.

[basic.syntax]

Squad has selected one path, I think intending to reduce complication for first-time players: the box says this is a rocket building game. So, here are the absolute minimum parts you need to get started in rocketry. With kerbal characters. I think it starts in a reasonable place, and 1.0 has improved things a little with some new branching. Nerfs to science points and new branching have helped with the .90 and earlier problem, where you didn't need to go farther than Kerbin's moons to fill out the tech tree. (I haven't really gotten into a career game yet, so I don't know by how much that might still be true.)

I'm just explaining what I think happened... I like tater's ideas.

[sherkaner]

Squad has selected one path, I think intending to reduce complication for first-time players: the box says this is a rocket building game. So, here are the absolute minimum parts you need to get started in rocketry. With kerbal characters. I think it starts in a reasonable place, and 1.0 has improved things a little with some new branching. Nerfs to science points and new branching have helped with the .90 and earlier problem, where you didn't need to go farther than Kerbin's moons to fill out the tech tree. (I haven't really gotten into a career game yet, so I don't know by how much that might still be true.)

I'm just explaining what I think happened... I like tater's ideas.

Well, I think the point many of us are making is: why not offer both rather than making the decision for the player?

I think having the rocketry basics as well as aviation basics from the start is the right solution. If you want to dig in setting speed records and training your pilots with aircraft missions, go for it! If you want nothing but to launch rockets and get to orbit ASAP, do that! I can't think of a good gameplay reason why there shouldn't be much more choice in technology development path directly from the start, and much more granularity thereafter. The beginning tutorials should lead players through the basics of what makes a good rocket or aircraft, and beyond that, turn them loose.

One of these days I need to dig into the more-moddable tech tree. I wonder if they've actually allowed players to do the things I couldn't do with my previous tree, like directly define what R&D facility level is required for each node (rather than have it based on hard-coded node cost thresholds). If they've really done that right, maybe I'll get off my ass and try another tree.

[CaptainKipard]

I can't think of a good gameplay reason why there shouldn't be much more choice in technology development path directly from the start, and much more granularity thereafter. The beginning tutorials should lead players through the basics of what makes a good rocket or aircraft, and beyond that, turn them loose.

Yeah it's like Squad don't expect people to know their butts from their elbows. I think everyone understands that a ladder is not an engine. The lines between varieties of parts in the game are not blurred much at all. To me there doesn't seem to be any potential for confusion. The only confusing parts may be the various resources and engines using them, but games are meant to be a little cerebral like this.

I'm so sick of this hand-holding. I started playing when there was no career mode. I had access to all parts from the beginning and I didn't have trouble figuring out what engine to use with what tank. This is just silly.

[Kegereneku]

Give some credit to SQUAD,

I don't think they take anybody for moron or don't wish themselves the Tech-tree to feel something fulfilling. But they have their own imperative and more at risk.

First, they can only work on so much at once (AERODYNAMIC WAS QUITE IMPORTANT DON'T YOU THINK ?).

Then...

- The tech-tree must progress at a reasonable rate (hence the classic (but horrible) log curve on the RP cost)

- It mustn't break or allow part-exploit too easily (which can happen if you could focus on a few parts)

- The system must be failsafe regardless of how badly you spend RP. (hard to imagine I suppose but non-Kerbal value prudence first)

And it must work on the first pass (1.0 RELEASE OBLIGE) since what we ought to replace is a system that ultimately "work", even if badly.

Thing is, even though I support the idea we suggested here would take multiple balance path.

The only thing I blame SQUAD for is not being bold enough (like Kerbal) to have tried, even for 1.0.

Now I hope the 1.0 Release didn't cemented that ABOMINATION or a tech-tree. Tree can't grow on cement anyway...

[Murph]

Playing through 1.0 career on "normal" (plus stock craft enabled). I'm not really convinced by the current tree. The game feels extremely grindy to get access to aircraft. Sure, you can do a very basic slow and low altitude jet fairly early on, but I feel like the turbojet and Aerospike are much too deep into the tree. As someone with a real world hands-on background in private and military aviation, it seems quite absurd that fast and high altitude jets are pushed to a quite late stage of the game, especially once you consider needing to buy the R&D, SPH, and runway upgrades, as well as grinding a hell of a lot of science. Ladders deep into the tree, ...?!? The old basic gear bay quite deep as well (no, the new simple gear doesn't excuse that)?

I think we need to get a much broader set of aviation much earlier in the tree (mark 1 bits, turbojet, and fast intakes all by the 90 science level at the latest, ladders down with the earliest aviation bits). Fully enable the Aeris 3A, Aeris 4A, and Albatross 3 by the 90 science level, so that the visual and temperature survey contracts on Kerbin are achievable with planes. In the real world, we had the A-12 in the early 1960s, SR-71 in the late 1960s. I.e. we had the fast high altitude jets in parallel with the Apollo program, with NASA having their YF-12 variant of the A-12/SR-71 to play with.

Right now, the early game feels like a huge catch 22 of needing a ton of cash and science to be able to do anything much at all, but not being able to grind the cash and science at a reasonable rate (without being quite artificial and science bombing Mun very early). Pushing people to spam the Mun (and maybe Minimus too) for science at a very early stage just doesn't feel right to me, particularly as you're spamming them with poor equipment, and then lose some of the opportunity to do them properly with the 2.5m rockets, lab, etc.

The perfect setup for me for early game would be Aeris 3A and 4A, and Albatross 3 levels of aviation early, enabling comprehensive atmospheric science exploration of Kerbin, in parallel with the beginnings of sub-orbital and LKO rocketry. That would nicely lead into 2.5m missions to Mun and Minimus with labs. It would probably also be a good thing to keep the atmospheric Kerbin science exploration contracts below 12,500m until the turbojet is enabled (the ones that have you chasing down targets on the map). Science from Mun and Minimus would move to enabling 3.75m, interplanetary, and high end stuff, instead if enabling atmospheric exploration of Kerbin.

[CaptainKipard]

Playing through 1.0 career on "normal" (plus stock craft enabled). I'm not really convinced by the current tree.

Did you see our proposed tree in the OP? What do you think about it?

The tech-tree must progress at a reasonable rate (hence the classic (but horrible) log curve on the RP cost)

I don't actually have a problem with increasing costs like this. Every game does it.

It mustn't break or allow part-exploit too easily (which can happen if you could focus on a few parts)

Like what kinds of exploits?

The system must be failsafe regardless of how badly you spend RP. (hard to imagine I suppose but non-Kerbal value prudence first)

If you design a game where the player can't fail by making stupid choices then it's not a game anymore. This goes to the hand-holding I hate most.

Edited April 30, 2015 by Cpt. Kipard

[Murph]

Did you see our proposed tree in the OP? What do you think about it?

It's interesting, clearly has had some thought put into it. It's a bit much to process quickly, so this is not comprehensive feedback on it, more some immediate and relatively quick thoughts.

1. Turbojet and basic ram intake needs to be the top end of level 1, not level 2. This is to enable atmospheric exploration of Kerbin early, as an aid to getting started on science and reputation. Probably the current 90 point level. Aerospike should be level 2, it's not really that fancy, but is handy for a simple and low profile LKO crew shuttle type thing. It's not really high end tech. RAPIER is fine as level 3, it's genuinely bleeding edge high tech.
   
2. Similarly some of the aerodynamic stuff should probably move down to level 1.
   
3. Mark 2 & 3 can stay in levels 2 & 3, only need mark 1 and probably all the wings and related simple aerodynamic stuff in level 1.
   
4. Some more of the science equipment needs to be in level 1, otherwise it's too slow and grindy to gain science and open up the tree and R&D levels. Thermometer and barometer are the obvious ones that should be down at a very low level. They are ancient tech, pre-date basic aviation!
   
5. Batteries should move down a notch, so that the 1.25m disc battery is in level 1, and the 2.5m disc in level 2. Having a level 3 something that's equivalent to just spamming a few level 1 devices inside a service bay doesn't make a whole lot of sense to me. Just make the batteries available early and be done with it, there's no fun aspect to not having the right battery module available when you need it (i.e. 1.25m when you're building lots of 1.25m rockets, and 2.5m when you're building 2.5m), it's just annoying.
   
6. Probes early is good, but the new 1.0 mechanic where there's different levels of SAS capability for the probes probably means you shouldn't have so many in level 1. Level 1 should probably have a basic no-SAS probe, simple SAS probe, and pro/retro-grade-hold SAS probe, as you move up the level. Levels 2 and 3 can add the remaining features.
   
7. Strong basic aviation (including fast and high), and solid basic probes are things were absolutely should have early.
   

[GusTurbo]

Definitely agree that the tech tree needs an overhaul. I don't understand the "rockets first" reasoning. If people want to jump in and build rockets, they can play sandbox. Career doesn't feel like a progression at all, it just feels like randomly unlocking parts.

[sherkaner]

Career doesn't feel like a progression at all, it just feels like randomly unlocking parts.

Yeah, that's a good concise way of putting it. Even totally aside from gameplay issues, it's just not fun when there doesn't seem to be any sense to the technology development, and very few meaningful choices to make on how I direct it.

[basic.syntax]

...I don't understand the "rockets first" reasoning...

I just wrote on this topic elsewhere

Squad clearly wants to put the focus on Kerbals first: they give the game its character, and make it a game (as in: an entertainment.)

Seeing their goofy expressions gives the broader audience additional reasons to play, much more than a dry simulation would.

KSP would not have gained nearly the following it has, without these interactive mascots.

It's very unlikely that Squad would change the career game to "unmanned first."

...after reading this

Here's my philosophy on this and why I actually understand why manned (kerballed?) is first.

This is less a simulation, and more a story where the Kerbals - these awesome, incredibly brave, and endlessly optimistic little dudes are the focal point. They are the protagonists in the story, and the player is the hero, spooling out the narrative. And in that scenario, starting without them is just.. weird. It's Kerbal Space Program, and I think the thing to remember is that it's all about the Kerbals, and there just happens to be some rocket parts lying around.

I was also reading with interest - Flowerchild's thoughts on these subjects, in the BTSM thread. There's a good exchange starting about here, with Arsonide responding on the next page... contracts, tech, science - its all tied together, and a lot of work to build something cohesive (echoing tater discussion,) while only editing one part of it.

[CaptainKipard]

I just wrote on this topic elsewhere

I think this question derails the topic. No one here is advocating unmanned first. We're advocating choice. Let's not get into this again. Sherkaner's tree includes the little pod from the start.

Edited April 30, 2015 by Cpt. Kipard

[Kegereneku]

Did you see our proposed tree in the OP? What do you think about it?

I don't actually have a problem with increasing costs like this. Every game does it.

Meh, Grinding time staying constant is ok.

Grinding time constantly increasing is bad design usually used by bad game which confuse it with "increasing play time" or game which business model involve you paying for every second playing and thus want to make you waste as much time as possible. (I'm looking at you Eve Online and every MMO in existence)

Of course, this frustration is increased in that you need to GRIND for part-bundle of 80% uninteresting parts just to get the 3 you needed to make a good rocket.

Otherwise it MIGHT feel comfortable.

Like what kinds of exploits?

I guess the right therm isn't 'exploit', more like "accidentally make the game too easy".

As analogy : the way parachute make heat-shield practically useless (on Kerbin) but with the "right technology" rather than "OP/bugged tech"

If you design a game where the player can't fail by making stupid choices then it's not a game anymore. This goes to the hand-holding I hate most.

I recognize it would be improbable considering "more booster" is a valid solution to every problem.

But I meant failsafe in that you cannot do tech-tree choice that turn a "easy game" into something very very hard because you choose the wrong tech. Ex : investing in the wrong technology making every RP point harder to attain.

Since the only way to solve it is to do it the very hard way or restart the career mode (or cheat) you don't want it to happen.

THE POINT IS,

Even if SQUAD disappointed us and wasn't bold enough to make an really enjoyable Tech-tree (some player are still satisfied), they still delivered a system that "work" and there's many way it could have not if they tried different approach (including different model from the one suggested first-post here, other like budget-based or stuff).

They even had this self-imposed time constraint (Alpha beta release...etc), so I can't blame them (too much). Even more since I wouldn't trust most Suggester here to design a video game themselves.

[kcs123]

I think having the rocketry basics as well as aviation basics from the start is the right solution. If you want to dig in setting speed records and training your pilots with aircraft missions, go for it! If you want nothing but to launch rockets and get to orbit ASAP, do that! I can't think of a good gameplay reason why there shouldn't be much more choice in technology development path directly from the start, and much more granularity thereafter. The beginning tutorials should lead players through the basics of what makes a good rocket or aircraft, and beyond that, turn them loose.

One of these days I need to dig into the more-moddable tech tree. I wonder if they've actually allowed players to do the things I couldn't do with my previous tree, like directly define what R&D facility level is required for each node (rather than have it based on hard-coded node cost thresholds). If they've really done that right, maybe I'll get off my ass and try another tree.

Second this. Especialy for us that loves FAR and Deadly re-entry. I would prefer building basic aircrafts prior rockets, just to train pilots more before going to space. It is shame on SQUAD that they didn't use almost anything suggested in this thread. They were having most of stuff ordered and more-less balanced, but they didn't even a try.

I hope that they were alow to tech tree be more modable or at least easier modable than in 0.90. Also sorry that I have not been of much help with feedback on tech tree topic, quite busy elsewhere lately.

[CaptainKipard]

I hope that they were alow to tech tree be more modable or at least easier modable than in 0.90. Also sorry that I have not been of much help with feedback on tech tree topic, quite busy elsewhere lately.

It's somewhat easier. There's still no easy way that I can see to organise parts in bulk. Any modded tree will still need to place parts from mods individually.

[kcs123]

It seems that Yemo already make something usefull. SETI balance mode covers some of things that was discussed in this thread. It is based on Comunity Tech Tree mod. Although I suggested wider root of tech tree, that will alow even more customization for various moders. But even without more nodes in career start looks promissing.

I thought that many people missed those threads and could find it usefull.

[basic.syntax]

Over 700 posts in less than 24 hrs, on a reddit post asking to "work on the tech tree." Wow. I just glanced. Moving parts around, "it should started unmanned," "No, it's the Kerbal space program," etc.