A more intuitive tech tree

[CaptainKipard]

It does force players to play a certain way, starting small and getting bigger. There is definitely room for major improvement in the tree, but that doesn't make it a priority. I would rather see drag physics modeled better than they are before they fix the tech tree, because at the moment the tech tree works bug drag physics are just silly.

I'm the opposite. The aerodynamics don't frustrate me as much as a contrived progression does. I barely even notice the drag issues. Just add more boosters.

Generally, I want both of those, and a lot of other improvements, but I want the tree sooner rather than later.

Edited December 18, 2014 by Cpt. Kipard

[klgraham1013]

It does add to the game experience, just not in the way you want it to. It does force players to play a certain way, starting small and getting bigger. There is definitely room for major improvement in the tree, but that doesn't make it a priority. I would rather see drag physics modeled better than they are before they fix the tech tree, because at the moment the tech tree works bug drag physics are just silly.

The tech tree is a much easier fix. We've nearly done all the work for them. No complicated mathematics involved. Just rearrange the nodes and the parts within.

I understand there's still work involved, but it can't be nearly as complex as redoing the aerodynamics of the game.

[sherkaner]

The tech tree is a much easier fix. We've nearly done all the work for them. No complicated mathematics involved. Just rearrange the nodes and the parts within.

I understand there's still work involved, but it can't be nearly as complex as redoing the aerodynamics of the game.

Agreed. I think part of the reason we're all pretty eager to bring this up now is that technically speaking it's relatively simple -- but gameplay wise I think it might end up being a core element around which everything else gets balanced. Once you settle on a tree like this to maximize flexibility, you can start thinking about how things actually get unlocked (such as my earlier R&D idea in this thread), how much things "cost" (in science, money, time, etc.), and the stats on the parts and parts options themselves. So now that we're in beta, I'd love to see the game start the probably-sooner-rather-than-later gameplay balancing process with a really solid tech tree like what we're discussing.

[Kegereneku]

It might take effort to balance a new tech-tree configuration but on the other hand, the actual Tech-Tree IS unbalanced as well as being frustrating.

The devs can do that in their own time, but it will have to change.

I'm repeating what we all discussed but the actual tech-tree feel like a placeholder.

Edited December 19, 2014 by Kegereneku
correcting typing error

[CaptainKipard]

I hope so. There hasn't been so much as a peep about it from official channels ever since it showed up.

[Kar]

In my opinion is not a matter of re-design the tech tree, even if it could be much more improved in a logical way. For example, I think probes should be the first thing to use, then a manned flight.

The thing is, that science still is a click fest and grind. Science should be a sorta of investment (like the rep system) where we need to develop part to build for example, a engine. Also, it should be time related, I mean, instead of click, click, click, get 300 points, spend 280 points in engines and fuel tanks, it should be done in a way to investigate materials, making experiments with real application in the tech tree.

So if we get a soil sample, lets say from the mun, we then put that soil sample in the lab and do several tests, which could result in other advances like new composites or something like that.

The actual science tech tree, should be called engineer tech tree, where the parts would be available after done x amount of experiments in the labs.

For this, crew reports, soil samples, go cans and so on, are not enough. Why research a solar panel for example, if we do not know if that panel will be strong and efficient enough in the cold vacuum of space? With the research of new materials, experiments could be done to test materials or a combination of materials the player could choose to see what materials would be best to construct a certain part that would be used in the assemble line together with other developed parts to construct the final part that would attach to a rocket or a space ship or whatever.

So, we need science labs both on the ground and in space (orbit), engineer division to manufacture those blueprints that we got after testing different materials and science data gathered by missions. A simple probe measuring pressure, temperature or other things, that data, could be used in the development of experiments in the science labs. A combination of procedural generated experiments would yield a certain result and a blueprint would be "looted". The engineer team would build that item using a the "looted" blueprint and then we could test it on a rocket. If that part works or not, it's anyone guess.

This would make the game more science oriented and with a procedural generation system, every game would be different adding new things to try and discover.

A launch of a telescope for example, to scan other planets or other starts (if that will ever be implemented) could give valuable research to be applied in experiments in the science labs.

If so, we would need several science branches to test different results.

[klgraham1013]

So, Max just tweeted about the upcoming update. Aero and resources. Still no mention of the tech tree. Have they once mentioned the tech tree since it's release?

[Esme]

The only thing that's crossed my mind that I haven't seen proposed here is that maybe Science could also accumulate at one point per day above and beyond anything garnered by one's activities in the game (or, rather, there could perhaps be an option to enable that). This to represent Science inexorably progressing in other ways away from KSC, and also to give those who are struggling another option to help themselves out (ie: worst case, they could just fast forward through time to gain a few extra science points they need. The drawback would be that if they're playing career mode, some contracts may have expired). But that's just a thought, and one I'm not too bothered about one way or the other. I am VERY much in favour of Cpt.Kipard's 'tech-star'. On the subject of different tech trees, ISTR when I played with teh KSP Interstellar mod, it allowed the loading of different tech trees. I can imagine thet it might not be simple implementing a tech-star if the system is intended to deal with tech-tree's, but if the tech-progression system could be made flexible enough to allow for tech stars as well as tech trees, then that's the main problem sorted for everyone, I'd think. Total ability to choose technological progression through the game.

[CaptainKipard]

This is why I suggested allowing modders so append nodes dynamically. There would be much less need for modded tech trees (which at this point can't coexist with eachother). Every mod would just define their own nodes and roots/prerequisites, which would just sit on top of the existing tech-star.

Let's say TACLS needed a place for its parts. The modder would create a node definition somewhere, telling it to branch off of the command pod nodes for example. The nodes would probably have to dynamically rearrange in the tree viewer, but I'm not sure that's too big of a hurdle.

[Kegereneku]

So, Max just tweeted about the upcoming update. Aero and resources. Still no mention of the tech tree. Have they once mentioned the tech tree since it's release?

Considering how HEAVY and IMPORTANT aerodynamic is. (expect a lot of "They ruined it, I quit KSP!") I think it's okay to rework the tech-tree to later.

[Ziff]

So, Max just tweeted about the upcoming update. Aero and resources. Still no mention of the tech tree. Have they once mentioned the tech tree since it's release?

The tech tree is working for now, so I don't think they are too concerned yet. Refining it will be easy and won't take a lot of time once everything else is sorted out. They're going to add deep space refueling resources in the next update as well, so I imagine there is going to be some kind of addition or changes to the tech tree anyway.

[katateochi]

I like the ideas in here! The less-linear suggestion from Cpt. Kipard is a great idea. I feel that the current tree forces you to play out the same game-play with each career and that's not what I want from KSP.

I'd really like to get away from linear progression completely. I'm coming into this thread a bit late and I've not read it all in detail so sorry if this has already been covered before but;

What I'd like to see is tech "slots" for each tech type and you can add any part you like into the slots to unlock it. To start with the size of the slots is very small and you have to spend sci to increase their size. So to start with you have a all the slots available but they're only able to hold 1 or 2 parts, ie; the command pod slot starts with just space for a single 1 man command pod, the engines slot just has space for 1 small engine and one small booster.

If you want to start with the Mk1 Cockpit cockpit you can, but you then won't have space in the command pod slot for the Command Pod Mk1. Or rather than starting with the LV-T30 and RT-10 booster you could start with a LV-909 and a basic jet engine. More advanced or larger parts should take up more space, so to start without you couldn't fit a Mk1-2 pod in the command pod slot, or you could start with a TurboJet but that would take up as much space in the slot as two engines so that's all you'd have, until you pay sci to increase the slot size.

Maybe you don't start with all the slots available, but you get to choose which ones you start with so your start to a career is totally flexible, if you want to start down the aircraft path or even start with rovers then you can.

[Ziff]

The tech tree is a much easier fix.....I understand there's still work involved, but it can't be nearly as complex as redoing the aerodynamics of the game.

1: We have no idea what the plans for the Tech Tree are. Perhaps they want to add more nodes and breakdown parts even more then just rearranging them.

2: That's not how they work. They work on the things that need the most improvements, rather than things that need minor improvements. Otherwise they would have a few key polished features and a bunch of completely broken features and mechanics.

[CaptainKipard]

less-linear ... get away from linear progression completely.

I don't think I've seen these particular words here yet, and it's a very good point.

What I'd like to see is tech "slots" for each tech type and you can add any part you like into the slots to unlock it. To start with the size of the slots is very small and you have to spend sci to increase their size. So to start with you have a all the slots available but they're only able to hold 1 or 2 parts, ie; the command pod slot starts with just space for a single 1 man command pod, the engines slot just has space for 1 small engine and one small booster.

If you want to start with the Mk1 Cockpit cockpit you can, but you then won't have space in the command pod slot for the Command Pod Mk1. Or rather than starting with the LV-T30 and RT-10 booster you could start with a LV-909 and a basic jet engine. More advanced or larger parts should take up more space, so to start without you couldn't fit a Mk1-2 pod in the command pod slot, or you could start with a TurboJet but that would take up as much space in the slot as two engines so that's all you'd have, until you pay sci to increase the slot size.

Maybe you don't start with all the slots available, but you get to choose which ones you start with so your start to a career is totally flexible, if you want to start down the aircraft path or even start with rovers then you can.

Could you maybe add some diagrams, or some bullet point lists. I'm having a hard time figuring out the usefulness of this. Why "slots"?

[katateochi]

Could you maybe add some diagrams, or some bullet point lists. I'm having a hard time figuring out the usefulness of this. Why "slots"?

"slots" is probably the wrong word. What I mean is that for each tech type you have a limited space/area into which you can drag parts to unlock, to start with that space is only big enough to fit 2 small parts. You then spend sci to increase the unlock-space (yeah, I'm struggling to find the right term here!).

The idea is not to have a progression of parts that requires part X to be unlocked before part Y, but to have all the parts potentially available and only limit you by how much unlock-space is available.

So I just roughed this out in balsamiq, I don't mean that the interface should look just like that with tabs and sliders, balsamiq is meant for web interface design so just using what it has.

You have different unlock-spaces for each tech type, command pod, engines, tanks etc and to start with that space is small, but you can select what ever you want so long as it fits. Some parts take up more space (if they are really big or really shiny) so in this example you could either select the LV-T30 + RT-10 booster as your starting engines, OR you can start with a basic jet, but that takes more space so you can only have one of the tiny engines with it. Then you spend sci to increase the unlock-space so you can unlock further parts, and I think that it should get more and more expensive to increase the size of the unlock-space.

Anyway, hope that makes my ramblings clearer!

If anyone wants to mess around with that diagram you can use the demo version of balsamiq. Click on Mockup at the top and hit Clear Mockup, then download this txt file and copy its contents, hit Mockup again and goto import and paste it in.

[CaptainKipard]

The way I see it this kind progression is already possible with tech nodes and funds. The funds and science are the limiting factor instead of this ^.

You could have e.g. a tech node with both LV-T30 and LV-T45, and buy either power or gimballing ability if you haven't enough funds for both. After that node you'd just continue research further nodes for bigger engines. It's a bit of a compromise between your idea and what's already available.

Unrelated idea

Have one branch with gimballed engines, and one branch with regular ones.

[sherkaner]

The way I see it this kind progression is already possible with tech nodes and funds. The funds and science are the limiting factor instead of this ^.

You could have e.g. a tech node with both LV-T30 and LV-T45, and buy either power or gimballing ability if you haven't enough funds for both. After that node you'd just continue research further nodes for bigger engines. It's a bit of a compromise between your idea and what's already available.

Katateochi's idea does offer an interesting way of gently having technology "eras" like I was talking about a bit earlier in this thread. Basically our general scheme here gives you a lot of flexibility in what you research, but the limited number of "slots" means you can't go *too* far in any particular area. Although to do that, I think you'd want to have the "slot" unlocks be across the board, rather than per-tech-type. Otherwise, as Kipard says, it's just a different name for spending science to get the next part you want.

In fact a pretty obvious idea would be to have one's general technology level -- whether implemented in slots or some other way -- tied to the R&D center building's upgrade level. (In in-game terms, once you upgrade your R&D generally, you now have better materials, design techniques, etc. that allow you to go beyond certain limits for all of your part types.) That's sort of what that building upgrade does today, but it makes a lot more sense with the techsplosion tree since the current tech tree sort of limits how far you can go in any area already by "virtue" of its somewhat nonsensical groupings. With a fully flexible tree, having that limit on technology era makes more sense.

Unrelated idea

Have one branch with gimballed engines, and one branch with regular ones.

That would be sweet -- good and interesting gameplay choices with that. You can pick un-gimballed engines, which would be perhaps cheaper to obtain as well as build, for an earlier advantage but later on in a career you'd end up having to start investing in more fins or reaction wheels to provide stability.

[CaptainKipard]

The tech tree nodes are already tied to the R&D building. You can't research higher tiers without improving the building.

Right now there are 9 node tiers split up equally between 3 building tiers. You can replicate this in the tech-star with concentric circles with nodes inside each representing each of the tiers.

Edited December 22, 2014 by Cpt. Kipard

[sherkaner]

The tech tree nodes are already tied to the R&D building. You can't research higher tiers without improving the building.

Right now there are 9 node tiers split up equally between 3 building tiers. You can replicate this in the tech-star with concentric circles with nodes inside each representing each of the tiers.

http://i.imgur.com/eN6s5fn.png

Precisely what I was poorly trying to describe. I think that's perfect.

[katateochi]

yes I guess however you cook it there will still be a progression requiring that certain parts must precede other parts. Your system looks very good Cpt. Kipard. Question; the way you envisage it, would it be possible to start off with the basic jet engine and some basic wing parts? or if you wanted, start off with some wheels and go driving on Kerbin? That's the kind of flexibility that I'd like to see in the tech-tree wheel.

[Baythan]

Question; the way you envisage it, would it be possible to start off with the basic jet engine and some basic wing parts? or if you wanted, start off with some wheels and go driving on Kerbin? That's the kind of flexibility that I'd like to see in the tech-tree wheel.

This is pretty much what we are making the argument for. There are basically three(most likely more, but these three are the general idea) philosophies oh how people want to start their career games:

The Wright Way: Start with plane parts like wings, landing gear, air intakes and air breathing engines. Fly around gathering science and funds to unlock more parts.

The Sputnik Method: Unmanned Probes! Rocket engines, probe cores, fuel, and maybe some batteries/solar arrays.

The Full Kerbal: Why test stuff first? Put Jebediah on a firecracker and send him up! Rocket engines, fuel, and command pods.

While the unmanned and manned rocket starts can be very similar, the unmanned probes can be much lighter and therefore you start with smaller engines.

So in the end we've all pretty much determined that we need brand-new career games to start with either nothing at all and we buy/research whatever parts we want with a starting allotment of research/funds that will allow us to get just enough parts to fulfill one of those three ideas or we get to pick a starting node that has one of those three sets in it. Mostly what we all want is the CHOICE to pick what parts we begin with so we can customize our career path as we choose.

The Tech-Star/Wheel is just a visual representation of being able to go in whatever direction we want. The layout of the tech tree doesn't HAVE to be a circle as long as we get to pick our path and actually get all the parts needed to do simple versions of whatever mission type we decide to start with.

The trickiest part of making a fully player-choice driven tech tree is making sure you cant lock yourself into a position where you picked parts that can't actually DO anything together, like getting engines, lights, batteries, science equipment, and probe cores.. but no fuel. The tree has to start relatively limited, but should not force us into doing manned rockets first if we don't want to. My biggest complaint with the techtree as it stands in how HARD it is to get all the parts needed for a simple(not rocket-powered) plane.

[CaptainKipard]

yes I guess however you cook it there will still be a progression requiring that certain parts must precede other parts. Your system looks very good Cpt. Kipard. Question; the way you envisage it, would it be possible to start off with the basic jet engine and some basic wing parts? or if you wanted, start off with some wheels and go driving on Kerbin? That's the kind of flexibility that I'd like to see in the tech-tree wheel.

It's not really up to me what you actually start with in the first node, but all other nodes in the diagram are available right after that, so you can go with jets first if you want.

I'm not sure myself what to put in the Start node, except a basic science part and some wheels so you have something to collect science with. Why don't you suggest what to put in it? Squad already gave us the option to start with some science available so really even these parts might not be necessary in the start node, since you'll be able to research in the direction you want. IMO the start node should have very few parts that in no way suggest to you what direction you should take.

The layout of the tech tree doesn't HAVE to be a circle as long as we get to pick our path and actually get all the parts needed to do simple versions of whatever mission type we decide to start with.

Yeah. I only went with the radial layout because it's efficient with monitor space.

Edited December 22, 2014 by Cpt. Kipard

[Baythan]

Yeah. I only went with the radial layout because it's efficient with monitor space.

Yes, the radial layout is the more space-efficient one, and I happen to like it. The point behind saying it doesn't have to be a circle is because any way you shape it, it does the same thing. At least, it does until you start requiring certain parts of one tech line before you can get parts of another tech line like the current tree does (gotta have one or both 'these two' researched before you can research 'this one').

[Superfluous J]

Random thought that goes along with Kipard's radial diagram (which I quite like. I was fiddling with a similar one myself a while back) and Baythan's "big 3" starting plans. It'd be a bit kludgey, but what if you started with - say - 1000 science points and each of those 3 nodes cost 1000 science to unlock? You could unlock one right away, and then would need to earn a lot of science to unlock another. There would then also be some other branches that had things like structural parts, science parts, electronics, etc on a more normal progression.

So one playthrough you could go for probes first, another planes first, and a third manned first, but in all 3 you'd be able to unlock the thermometer and girders with the same amount of science points invested.

[CaptainKipard]

I don't like that. It seems contrived for no purpose other than restricting what you can do.