Logical tech tree

[KerikBalm]

On 9/3/2018 at 10:26 PM, Wjolcz said:

Because career mode is so poorly designed there's nothing else left to do.

Meh, there's just as much as sandbox to do once you've unlocked all the tree. I very rapidly advance through the tech tree, then play it like a continuous sandbox save, just with a budget. I try to go independent of contracts and use science-> funds to finance my missions.

Also... KSP has tutorials now, thus the the tech tree shouldn't need to be one. It should be like a "tree", branching with consistency in each branch.

You can have a power branch (or electronics that splits to power generators/storage, probes, communication), an engines branch: basic SRB that splits into other SRBs, the LV-T30 liquid fuel engine-> other LF engines, the juno that leads to high speed low ISP engines (whiplash and rapier) and the opposite (goliath and wheesley). Some should require 2 branches (maybe the panther requires the wheesley and whiplash trees) (LV-N requires LF engine research and nuclear/RTG research), etc.

To some extent this is implemented (there is an electronics branch, there is a definite plane branch), but like tech trees in many RTS games, you should be able to start specializing in one path quite early... and for KSP is basically a choice between planes and rockets, so I would move the juno and basic plane wheels to the 2nd node, and allow the first unlock to be a choice between a jet engine or a LF engine.

 

[Pthigrivi]

Here's what I came up with a few months ago when I was thinking about all this. Generally the idea here is to create an earlier split so players with different play-styles could push to unlock things down the lane of their choice and still be able to branch to other strategies. So you could start out sending small probes to collect lucrative science and then use that to unlock bigger rockets, or use planes to explore around Kerbin before moving into lunar manned missions, etc. Each branch starts out with the simplest of parts for that category so even new players could pick and choose what new game elements they'd like to explore without becoming overwhelmed. 

Edited September 5, 2018 by Pthigrivi

[Tyko]

Funny. I've played this game since 1.0 and restarted on countless occasions for one reason or another. I don't think I've ever unlocked the middle of the tech tree with the plane parts. There are a few parts hidden in there that I wanted like fairings, so I just moved those to other nodes so I wouldn't have to unlock a dozen parts to get the only one I cared about.

 

[Blaarkies]

8 hours ago, Tyko said:

Funny. I've played this game since 1.0 and restarted on countless occasions for one reason or another. I don't think I've ever unlocked the middle of the tech tree with the plane parts. There are a few parts hidden in there that I wanted like fairings, so I just moved those to other nodes so I wouldn't have to unlock a dozen parts to get the only one I cared about.

 

Fairings are in nodes like " Specialized Construction" , " Advanced Construction" . I don't think they are hidden in the middle of the tech tree with the plane parts

[Fraktal]

Me too have been brainstorming about a tech tree with logical progression. Note, however, that I'm the masochist type who plays with 10% science income because I like to run my space program as a long-term project rather than a speedrun. Problem with it is that if I throttle the speed of progression too much, all I achieve is getting the player frustrated.

So, without further ado, here's what I have. Note that it's highly incomplete and primarily focused on propulsion tech and plane building. It might also be nonsensical at several places, as I wrote this on-and-off literally over years. I also don't like it when a single node unlocks a dozen parts, half of which aren't what I want from the node and end up never using whatsoever; I'd rather split it apart into several nodes that each give 1-2-3 parts closely related together. I mean, yeah, more nodes means more science to unlock the whole tree, but doesn't the science lab provide theoretically infinite science for as long as it's being fed with data and thus, you'll never actually run out of science to collect before finishing the tech tree?

 

Spoiler

• Starting node gives only the Mk1 pod, the Flea, and the thermometer instead of the Mystery Goo. I originally wanted to put the Stayputnik here instead of the Mk1 and have the Mk1 only become unlockable once the player has the nodes for its components (ie. Mk1 cockpit and the first reaction wheel), but I realized the player won't be able to steer the rocket whatsoever.
• Tier 1
  • Solid Fuel Rocketry - "We are reasonably sure we can cram enough explosives into our rockets to make them fly higher, preferably without blowing them up too soon in the process."
    • Unlocks the Hammer.
    • Propulsion research branches here: the SRB tree is cheaper to research than the liquid fuel tree, but doesn't go very far and is overkill for smaller payloads to begin with. Not quite a "schmuck bait", more like a stepping stone: gives a head start, but hits its limits quickly.
  • Automation - "It should be possible to rip enough instruments out of a cockpit to make the rocket practically fly itself. Of course, there's no guarantee it will actually stay in the air."
    • Unlocks the Stayputnik, the small reaction wheel and the radial parachute (since the Stayputnik doesn't have a node for the regular one).
    • As said above, I originally wanted to invert the description of this node and develop manned flight a bit later, but realized it won't work because if the player doesn't invest his starting science in reaction wheels right away, he gets stuck with uncontrollable rockets unable to reach other biomes to gather any more science. Instead, the player gets early access to the Stayputnik to play around with unmanned flights, although there isn't much point to it due to the lack of SAS.
  • Staging - "If our rockets are going to fall apart anyway, they might as well do it in a semi-controlled fashion."
    • Unlocks the size 1 decoupler.
    • Believe me, engine-assisted decoupling (ie. attaching the lower stage with a low-heat-tolerance part and firing the upper stage into it to decouple) is a pain to get right (because if you set the throttle too low in the VAB, it won't heat the part enough to "detach"), so giving a bit earlier access to this is a major headache relief for masochists who like to experiment with suboptimal configurations, without being OP.
  • Stabilization - "The art and science of keeping our rockets pointing in the right general direction."
    • Unlocks the Basic Fin.
• Tier 2
  • Medium SRB - "If there is such a thing as a too large solid-fuel booster, we're not there yet."
    • Unlocks the Thumper
  • Liquid Fuel - "We're not sure if this stuff is explosive enough for our purposes, but our chemists insist it burns really well for as long as it gets oxygen. And it can even be extinguished for later!"
    • Unlocks the Juno and accompanying tank and intake.
    • Propulsion research branches again: the player gets early access to plane engines to more easily gather enough science from Kerbin to build their first space rocket because parachuting down from a low-flying plane is much more easier to hit the target biome with than dropping down from orbit at the exact right angle and speed.
  • Atmospheric Flight - "Flying this low might not be as fun, but at least the crew won't hurt themselves as badly when they come back down. We think."
    • Unlocks the Mk1 cockpit, the LY-01, the structural fuselage, the small delta wing, the A and D wings and the smallest elevon.
• Tier 3
  • Liquid Fuel Rocketry - "As one of our engineers put it, 'In your FACE, oxygen!'"
    • Unlocks the Reliant and the two bigger FL-T tanks.
    • Logic here is that the first stuff should be big and inefficient, with more research required to downscale it for optimal performance. That kerbals would go for power over controllability is just a bonus.
  • Turbofan Engine - "Now the fun starts."
    • Unlocks the Wheesley, the Mk1 LF tank and the Engine Nacelle (to give it a reason for existence).
  • Aircraft - "Our engineers insist there must be a quantifiable difference between flying and hurtling through the air. They just need to figure out what."
    • Unlocks the Mk1 crew cabin, the LY-05, the second and third elevons, the B and C wings and the A swept wing.
    • I feel too many parts are getting unlocked here, but I don't know what to split off.
• Tier 4
  • Upper Stage Engines - "Making our engines weaker might sound counter-intuitive, but being able to steer the rocket might be useful."
    • Unlocks the Swivel, the Terrier and the two smaller FL-T tanks.
  • Heavy Turbofan Engine - "Because bigger is always better."
    • Unlocks the Goliath. No other research nodes branch off of this one, so researching this is completely optional.
  • Advanced Jet Engines - "Some of our engineers scoff at jet engines as not true engines because there are no flames coming out of them. This should quell their fears."
    • Unlocks the Panther and the matching circular intake.
  • Heavy Aircraft
    • Unlocks the large delta wing, the FAT wing and control surface, the LY-10, the inline Mk1 cockpit, the NCS Adapter and the fourth elevon.
  • Integrated Control Surfaces
    • Unlocks the deluxe delta winglet and the FAT tailfin.
• Tier 5
  • Heavy Rocketry
    • Unlocks the Mainsail and the two larger Rockomax tanks. Again, rationale is the same as with the FL-T tanks' unlock order.
  • Small-Scale Rocketry
    • Unlocks the Spark and the Oscar-B tank.
  • Medium Fuselage
    • Unlocks the LY-50, the Mk2 cockpit, crew cabin, short bicoupler, smallest cargo bay and smallest LF tank.
  • Intake Flow Manipulation
    • Unlocks the pre-cooler, supersonic intake and radial intake.
  • Advanced Aircraft Control
    • Unlocks the AV-R8 and the standard canard.
  • Modular Wings
    • Unlocks all five wing connector parts, so the player can now make his wings as big as he wants.
• Tier 6
  • Heavy-Duty Upper Stage
    • Unlocks the Skipper, the Poodle and the two smaller Rockomax tanks.
  • Monopropellant Propulsion
    • Unlocks the Puff and the two small MP tanks.
  • Radial Engine Mounting
    • Unlocks the Twitch, the Thud and the R-11 radial tank.
  • Compact Rocketry
    • Unlocks the Ant and the R-12 tank.
  • Spaceplane
    • Unlocks the Mk2-to-1.25 adapter, Mk2 Clamp-o-Tron, inline cockpit, short rocket fuel fuselage and monoprop tank. Basically, what you need to get a Mk2 into space without making it a frankenplane of parts of mismatched sizes.
  • Reinforced Fuselage
    • Unlocks the LY-60, the long Mk2-to-1.25 adapter, larger cargo bay, larger LF and rocket fuel tanks.
    • Again, feels like too many parts are here.
  • Advanced Intakes
    • Unlocks both ramp intakes.
  • Ramjet Engine
    • Unlocks the Whiplash.
  • Advanced Control Surfaces
    • Unlocks the advanced canard and the F-22-lookalike Tail Fin.
• Tier 7
  • Superheavy Rocketry
    • Unlocks the Rhino and two Kerbodyne tanks.
  • Aerospike Nozzle
    • Unlocks the Dart.
  • Ultracompact Rocketry
    • Unlocks the Spider and the R-4 tank.
  • Active Attitude Control
    • I originally wanted to make RCS available here, but realized that the docking rings are basically useless without it. So, no.
  •  Heavy Fuselage
    • Unlocks the Mk3 cockpit, small cargo bay, cargo ramp, passenger module, short and medium LF tanks.
    • Too many parts here.
  • SSTO Construction
    • Unlocks the LY-99, all Mk3 adapter parts, medium and large cargo bays, engine mount, monoprop tank, all three rocket fuel tanks.
    • DEFINITELY too many parts here.
  • High-Speed Intakes
    • Unlocks the shock cone intake.
  • Combined Cycle Engine
    • Unlocks the Rapier.
• Tier 8
  • Advanced Engine Construction
    • Unlocks the Twin Boar, the Vector and the Mammoth.
  • Nuclear Energy
    • Unlocks the RTG and the LV-N.
  • Particle Physics
    • Unlocks the Dawn and the three xenon tanks.

Some notes:

• As seen above, I'd give the 2HOT Themometer as starting science instrument instead of the Mystery Goo Container and scatter the rest across the tech tree. Tier 2 is the barometer, Tier 3 is the accelerometer, Tier 4 is the atmosphere analyzer, Tier 5 is the goo, Tier 6 is the materials bay, Tier 7 is the gravioli detector, Tier 8 is the lab.
• Drogue chutes could be mid-game, parallel to the development of the bigger planes.
• To continue the relevant discussion in the thread, I think the most basic ladder could be made available early (Tier 3?) and since it cannot retract, thus is in danger of burning off during reentry, it still gives the player an incentive to invest into the retractable ladders before risking manned missions to Duna or Eve.
• Antennas, I think, can be safely pushed back to Tier 4 because you don't need an antenna in LKO and players don't generally go for the Mun until they have the Terrier (not to say it can't be done, I've done a no-Terrier Mun landing before).
• For solar panels, I'd move the basic OX-STAT down to Tier 3 as an anti-frustration feature. The bigger, fixed one, can be put at either Tier 4 or 5, doesn't matter; deployable ones can be put at Tier 6 but the retractable ones should be moved higher up so that the non-retractable ones aren't instantly obsolete. Tier 8 is available for the Gigantor. Fuel cells and the RTG are fine where they are.
• Radiator panels can be moved quite a ways up because you simply don't need them until Tier 7 unless you start launching solar probes. So... Tier 5 for the small radiator, Tier 6 for the small TCS and the edge radiator, Tier 7 for the large radiator and medium TCS, Tier 8 for the large TCS?

 

 

 

[Kernel Kraken]

Probably because building planes is more complicated than building a rocket, at least at the beginning. I would like a tech tree that makes sense, though.

[Jacke]

On 8/14/2018 at 9:38 PM, Kerbart said:

Famed cartoonist Gary Larson has a great story. It involves a cartoon which displays a 1950s household. Fedora-wearing husband comes home, putting his trenchcoat and briefcase down. Wife is cooking dinner. Oh, and they are mosquitoes. The caption is along the lines of “how was work, honey?” “It sucked.”

He said that he got tons of mail over the cartoon, pointing out that the female mosquitoes are the ones who suck blood, not the male ones.

Hmmm.  So, no one thought that because male mosquitoes survive on nectar, they probably suck it too?

Anyhoo, I started playing KSP back in version 0.23.  Sandbox was way confusing, lots of parts.  Career less so, but I still didn't quite get it.  Looked up some player-made tutorials and videos.  First suggestion was to launch a Mk1 pod by itself onto the pod and have Jeb EVA onto the pad and get science.  That was just wrong.  That it is still the same now, in KSP 1.4.5 (and likely 1.5), is worse.

I was born before Yuri Gagarin flew.  I grew up watching the Space Race.  I bought Liftoff! when it came out.  And BARIS.  And BASPM.       All flawed but all enjoyable to a degree.  KSP is and always will be a toy space program.  But it should be better than it is now.

Squad and all of us want KSP to be successful, because that's the only path that can lead to a better KSP.  That means it needs to both attract and hold new players and old.  It needs to be like P.T. Barnum's Three-Ring Circus.  Each part needs to be approachable and entertaining.  And there has to be more things to do.  In stock  And that also means in a better career than we have now.

 

On 8/10/2018 at 4:22 PM, Tyko said:

The game is called Kerbal Space Program, not Kerbal Airplane Program or Kerbal Rover Program or Robot Space Program. Since the game bills itself as being about launching kerbals into space, it's no surprise that the tech tree starts you with the ability to launch kerbals.

Right now I'm following @Triop, who just finished a stock Pole2Pole mission.  He's now working on driving from KSC to the Temple with the help of a ferry across the ocean gap.  Triop plays sandbox.  But it and other mission reports and challenges show players are interested in a lot of the things possible in KSP.  That some of those aren't even possible in any way until halfway to later in career is wrong.

Like @5thHorseman, @klgraham1013, @Wjolcz, and others have noted, there's a lot of issues with the parts as they are in the stock tech tree.  There's also issues with the various costs and parameters (fins that cost more than rocket engines, etc.).  Some of that is being addressed.   But there's definitely a problem with sequence of play that will happen in any stock career.  Just reading the various comments in this thread on what comes where is damning.

Change is always going to have some for and some against, some extremely so on both ends.  Anyone who's followed the comments in KSP Weeky on the new model redesigns coming in KSP 1.5 will see that.  But Squad has decided they need to do some change there for KSP's benefit.  I think the same can be said for stock career and its progression.

Again, like 5thHorseman, I've looked at changing the tech tree as part of making a new career mod.  It's possible now without code, just scripting.  But as he mentioned, it's bloody complex.  I've already come to the conclusion that for a new career mod, it shouldn't change the layout of the tech tree.  All the nodes should stay the same, as changing the layout of the stock tech tree is a complexity that doesn't add anything.  I think there's more than enough nodes.

What needs to be done is to move parts.  And some should be moved into the Start node.  That career can't start as real-world rocketry did, with uncrewed probes, is just wrong.

The other major change that I think should be done is to move away from tech tree progression based upon Science.  Contracts have been in the game since KSP 0.25.  They are a superior method for giving direction and unlocking parts and tech nodes.  Doing particular science should be a goal of contracts and the Science accumulated and act like another sort of Reputation.

Stock career and its tech tree and progression isn't the only long-standing issue that really needs some Squad love.  But it's a major one.

Edited September 18, 2018 by Jacke

[Tyko]

5 hours ago, Jacke said:

On 8/10/2018 at 3:22 PM, Tyko said:

The game is called Kerbal Space Program, not Kerbal Airplane Program or Kerbal Rover Program or Robot Space Program. Since the game bills itself as being about launching kerbals into space, it's no surprise that the tech tree starts you with the ability to launch kerbals.

Right now I'm following @Triop, who just finished a stock Pole2Pole mission.  He's now working on driving from KSC to the Temple with the help of a ferry across the ocean gap.  Triop plays sandbox.  But it and other mission reports and challenges show players are interested in a lot of the things possible in KSP.  That some of those aren't even possible in any way until halfway to later in career is wrong.

Yes, the game is a very flexible open sandbox and some ingenious players have come up with great ways to do things the were probably never considered when the game was designed. This is really cool and I support it. What I don't support is making the game unnecessarily more complex for new players or those who spend most of their time launching kerbals into space. 

What if my hobby is re-purposing my Volkswagen GTI into an offroad vehicle that can pull a 3 ton boat? I buy aftermarket parts to raise and strengthen the suspension, I rebuilt the frame to handle the tow load, etc. People might think that's really cool and I could become a popular youtuber showing off how I did it and giving tips to others. That doesn't mean it's Volkswagen's job to modify the base standard GTI with tougher suspension and frame. It's perfectly correct for Volkswagen to say "most GTI customers buy our car as a quick runabout and we're not going to add complexity/weight/cost that doesn't benefit a majority of our customers". And other customers who are driving it around town would rightly complain that adding all that stuff is going to make their car slower and less fuel efficient.

I love the idea of a more branched tech tree with research paths specifically for probes, manned rockets, planes, spaceplanes, rovers, landers, etc....let people pick their own path. This is a lot different than just overwhelming the low tiers with a bunch of stuff just to accommodate players who aren't launching rockets

 

Edited September 18, 2018 by Tyko

[Jacke]

41 minutes ago, Tyko said:

Yes, the game is a very flexible open sandbox and some ingenious players have come up with great ways to do things the were probably never considered when the game was designed. This is really cool and I support it. What I don't support is making the game unnecessarily more complex for new players or those who spend most of their time launching kerbals into space. 

Consider [long overstatement].

Oh, come off it!  Just look at the rover wheels.   The first teeny ones at tech tier 5, then finally a decent wheel at tier 7, and the last 2 at tier 8.  Same tier 7 node also contains the External Command Seat and the Rovemate probe.  Why not even one wheel lower down, better at Start?  Where the External Command Seat should be too.

Edited September 18, 2018 by Jacke

[Adstri]

To be honest I prefer a realistic tech tree. However, the old one is good for beginners.

Maybe make them switchable?

[Tyko]

4 hours ago, Jacke said:

Oh, come off it!  Just look at the rover wheels.   The first teeny ones at tech tier 5, then finally a decent wheel at tier 7, and the last 2 at tier 8.  Same tier 7 node also contains the External Command Seat and the Rovemate probe.  Why not even one wheel lower down, better at Start?  Where the External Command Seat should be too.

The first RL rover attempted was 1969 - same year of the first Apollo landing. First rover that required an external command seat was the lunar rover in 1971. Doing Apollo style moon landings requires a similar tech tier.

May not be ideal for people who want to cruise around the KSC on a go cart, but it's consistent with other real tech progression.

Edited September 18, 2018 by Tyko

[Triop]

11 hours ago, Jacke said:

Triop plays sandbox

I do play a career too. Allmost got the tech tree full without ever leaving orbit.

I get my science points by doing survey contracts on Kerbin, lots of fun.

I don't like space, So I'm glad I can play the game without ever launching a rocket.

[Jacke]

5 hours ago, Tyko said:

The first RL rover attempted was 1969 - same year of the first Apollo landing. First rover that required an external command seat was the lunar rover in 1971. Doing Apollo style moon landings requires a similar tech tier.

I know.  I was watching the Space Race at the time.

So that means some rover gear should be in Tier 6, where there is the equivalent of the Apollo Command Module and Lunar Module.  Perhaps in Advanced Exploration.

Of course, it would help if you could train Engineers to fix rover wheels on Kerbin, similar to history.  And not have to send them outside of Kerbin SOI as well as get experience with respect to Kerbin, Mun, and Minmus, as it is now.

Edited September 19, 2018 by Jacke

[klgraham1013]

6 hours ago, Tyko said:

The first RL rover attempted was 1969 - same year of the first Apollo landing. First rover that required an external command seat was the lunar rover in 1971. Doing Apollo style moon landings requires a similar tech tier.

May not be ideal for people who want to cruise around the KSC on a go cart, but it's consistent with other real tech progression.

If we're going to go with history, than cars and planes should come before rockets.  Obviously, the tree needs to be some combination of historical and gameplay considerations.  I personally think rover wheels should be early in the tree to accommodate the massive science rewards around the KSC in career.  Whether that mass of science around the KSC is good for gameplay is another question.  Also, there's an expectation players have from a tech tree.  I believe having all the wheels so deep in goes against an average players expectation.

Edited September 19, 2018 by klgraham1013

[Superfluous J]

17 minutes ago, klgraham1013 said:

If we're going to go with history, than cars and planes should come before rockets.

 

If we're going to go with history then 90+% of the parts would not even be in the "start" node. They'd be in the "100 years ago" node. The "Start" node would then have all of the fuel tanks and most of the engines in the game.