Career Mode Progression and Game Design Analysis

[Falkenherz]

I built my first plane and realized I don't have a ladder to EVA my Kerbal. How far am I from ladders? What? Why are they over there!?

It seems, for Kerbals, ladders are a high tech item, right next to wheels...

[Ignath]

Man, I hope SQUAD reads and pays attention to this OP. Spot on with all points.

[Moss]

Heat shields make command pods unstable and flip out when re-entering due to being "physics-less". The only way around this is to deploy the stupidly overpowered parachutes at unrealistically high points in the atmosphere, shrouded in flames no less. Either that or rely on the stupidly overpowered reaction wheels in the pod, which may not even have a pilot or the electric charge to use them.

I'm sick of relying on exploits to work around poor design.

[Daid]

The whole tech tree issue can be summed up quite easy. And it was easy to expect from the devnotes.

They tuned the tech-tree by looking at the testers team progress. As the tester team is most likely highly experienced players, the whole tree is now tuned towards experienced players.

So far the lack of bigger fuel tanks is what I found odd. The tiny first tank, that's normally a more advanced "keep your lander light" solution. Now I just need to stack a whole bunch of tanks to get a single big one. With all the wobble. In the end, I just upgraded the pad first, so I could just use solid boosters to get myself into space, instead of rocket engines. It's more part efficient, and more cost efficient. But I keep forgetting to set the thrust limit, causing quite some fire...

(Overal, I do enjoy the update so far. But I'm 100% sure the new tech tree will scare new players)

[Kerbart]

Excellent post.

For the maneuver nodes, what about having prograde/retrograde only first, and then unlocking normal and radial later on?

And once again it becomes clear that we need a (Gemini-style?) two-Kerbal capsule right from the start of the game. Indeed it is annoying that right from the start you'll need a two-seater cabin for various contracts, without having a satisfying solution for it. If it were possible to couple two Mk-I's together in tandem without it looking ridiculous I wouldn't complain. But you can't, so I will.

I agree that the upgraded tech tree is better, but doesn't guide the player towards understanding how to build better rockets.

What about unlocking certain parts through contracts instead? Or making them available? Of course it opens the room for abuse (keeping the "save Elon Kerman from orbit" contract just to keep having access to the two-seater capsule) but it would take away some frustration.

[Xavven]

Wow, thanks everyone! So many thoughtful responses I don't know where to begin!

A proper tech tree provides options that are completely reliable at the early stages, but give somewhat sub-par performance. The nodes that follow give better-performing items that require more skill to get that performance.

Yes! This philosophy would help the tech tree immensely.

Of particular annoyance is starting with the modular girder, ... There are little to no uses for those girders in the very early game...

Good point! I noticed this too.

Perhaps contracts instructing the player to visit certain biomes?

I agree! To break through the 45 science tier I flew a few parachute-assisted hops to local biomes (shores, grasslands, etc.) with mat bays and goo. This would make for a good contract type instead of the overly precise survey contracts, as the player doesn't yet have the tools to maneuver that accurately in early game.

It's all trade-offs and figuring out how to balance science vs funds, which I'm actually quite enjoying.

Yes! This is what it's about. The tech tree should have clear paths that support different objectives. One possible flaw in the progression is that early on you have to unlock many research nodes and/or buildings all at once to open up any of the gameplay paths, and they open up all at once. There is less choice involved the way it is currently laid out. Great comment! Thanks.

I built my first plane and realized I don't have a ladder to EVA my Kerbal. How far am I from ladders? What? Why are they over there!?

This is another excellent example. Ladder ought to be placed earlier in the tree.

There were more of your comments I wanted to respond to but it's time to go to work! I'll write more later.

- - - Updated - - -

And once again it becomes clear that we need a (Gemini-style?) two-Kerbal capsule right from the start of the game. Indeed it is annoying that right from the start you'll need a two-seater cabin for various contracts, without having a satisfying solution for it. If it were possible to couple two Mk-I's together in tandem without it looking ridiculous I wouldn't complain. But you can't, so I will.

Oh I just have to comment on this before I leave for work. I couldn't agree more! Tourism contacts are just so close to being the perfect early game contract, but a two-seater capsule or command pod is missing and the parts you need are just out of reach! Two Mk1 commands in tandem DO look completely wrong!

[klgraham1013]

Honestly, I don't know if I can keep playing stock career. I know I said I would play through it, but...

Ok. So here's my notes so far.

Early career actually has a very nice reward system. Constantly breaking records really helps with a sense of achievement.

Having to constantly test and revert to find appropriate TWR is annoying. Delta v and TWR readouts are really needed. The trial and error was only interesting the first few times.

No patch conics and maneuver nodes actually makes it harder on new players, with only experienced players being able to accomplish more difficult goals.

Running into artificial part limits inside a giant VAB breaks immersion. Not having larger 1.25m tanks raises the part count artificially.

I have now orbited Kerbin and have unlocked one 45 science node.

I have 3 tourist contracts available that all have 2 tourists. I only have 1 man pods. I have a test contract for the 2.5m decoupler when I don't have any 2.5m parts.

I've hit a road block. I've needed to upgrade the astronaut complex, tracking station, VAB, and launch pad to continue which has limited my funds. I need 7 science for the next node. I have a plane but no ladder for my Kerbal to get down on the ground.

So that's where I'm at right now. Not sure if I even want to continue. I'm glad some people find this fun, but I find it no different than when they introduced career. That being, no fun at all.

[giltirn]

As a 300 hour vet I am also finding the early game very frustrating in 1.0. The big difficulty I'm having is the lack of structural reinforcement until you unlock struts. You need to build tall in the new aero, but without long tanks or struts you get rockets that flop around like wet noodles. You can also try building outwards as well to compromise between noodliness and aerodynamic profile, but then you are stuck with the crappy radial decouplers that always result in the attached stages bending inwards while under thrust and that are loose enough to allow the external stages to wobble around (especially when the engines are gimballed), making it *very* hard to control the rocket. I would say that struts should be unlocked from the beginning.

[Kerano]

I've been playing through normal difficulty career to get a feel for what a new player will experience. Pretty much encountered the same things:

* First few flights are great, the money rolls in quickly from the default contracts. Progression feels quick and rewarding.

* After those first flights the difficulty suddenly jumps up. There's not enough money to get all the basic parts along with all the necessary upgrades for a basic Mun shot - so if you're not experienced with strapping together SRB monsters and navigating blind, you're forced to grind out a lot of contracts for cash before you go any further. This is the point where if I was new to the game, I would be getting frustrated.

Other thoughts:

* Lack of science from contracts is a real loss to the game IMHO. It was perhaps a little easy to get contract science in previous versions of the game, but I really enjoyed that it was an alternative to the biome-hopping grind. It needed tweaking down, but 1.0 pretty much obliterated it altogether.

* Tech tree is OK for a veteran player but very unforgiving to a newbie. Number of nodes should probably be reduced a little to allow for more meaningful choices. Parts that usually work together should be on the same node, not split into 3-4 different nodes in different areas of the tech tree. I have enough difficulty getting science as a veteran, how is a new player supposed to cope with this tech tree?

* Weird stuff going on with the new aero, particularly reentry. Pods desperately fight to turn the wrong way and burn everything up regardless of how you build your ship it seems. Deploying mysteriously invincible parachutes in the upper atmosphere seems to be the only solution for now.

Edited April 28, 2015 by Kerano

[SpaceChipmunk]

The tech tree has always been a problem, so much it has prompted me to actually join and comment after all these years!

The idea of an ideals based tech tree is a good one. The first few techs in the tree can unlock parts that are universal like lander legs, basic batteries, ladders, et al...

Then it would follow the ideal that you choose. Kind of like a more detailed version of what you get in Company of heroes where you choose to follow either Infantry/Airforce/Armour ideals. In this you could follow Manned Rocketry/Remote Tech/Aeronautics.

Within these you will get technologies relating only to their fields. Then contracts should change relating to whatever path you have decided to take. Tourism contracts? Go for Manned Rocketry or Aeronautics. Site Survey contracts go for Remote Tech or Aeronautics.

As for maneuver nodes not being unlocked at the beginning. Well I can understand that. The Gemini and Apollo astronauts had to work everything out on a piece of paper. It is only in the age of the microchip that computer systems can give a good representation of when they should be making what maneuvers.

[TothAval]

You can accomplish the tourist contracts regardless. Put them one by one in orbit, it will work, use a probecore to bring them back. More grinding though, but it will do the trick.

I have unlocked several 45 science nots yet and one 90 not. Did a mun flyby with a small craft. I really like the parttests, they force for some creative vessels if you intend to

accomplish as many as possible in one flight

Anyway, career is not what i imagined either, never was. I have to constantly force myself to feel the immersion, i am not immersed by the game, which is really sad. But i haven't

put much thought into it yet so i don't have a viable suggestion on how to improve this part of the game. I really like the idea and the feeling of unlocking stuff, after a succesful mission.

This always makes me feel as if the new tech is a result of that mission, which is exactly what i like most about the idea of a career mode, well, maybe this is the reason why i like

parttesting so much

Good discussion so far.

[WanderingKid]

As for maneuver nodes not being unlocked at the beginning. Well I can understand that. The Gemini and Apollo astronauts had to work everything out on a piece of paper. It is only in the age of the microchip that computer systems can give a good representation of when they should be making what maneuvers.

If you want reality, yes, I agree that it was a lot harder for our own rocket programs at first. Noone here is trying to take away from their accomplishment. The problem is that most new players, and a lot of experienced players like myself, are simply not rocket scientists. I learned how to play through stubbornness and tutorials about how spaceflight worked. If I wasn't so fricking stubborn (I play DF, that should everything) I would probably have tossed the game in the graveyard.

It is unreasonable to expect new players to HAVE to read pages upon pages of tutorials and a few hours of videos to be able to make basic maneuvers. The game should be teaching them that (I've done the training, it doesn't).

To add to that, the guys in the ships weren't the only one doing the calculations. They had teams of mathmaticians and physicists on the ground relaying up to them what to do, and when, even for Gemini and Apollo.

[Bedwyr]

As for maneuver nodes not being unlocked at the beginning. Well I can understand that. The Gemini and Apollo astronauts had to work everything out on a piece of paper. It is only in the age of the microchip that computer systems can give a good representation of when they should be making what maneuvers.

You're absolutely right, it is something that makes intuitive sense that way. But the case against that (I'm not taking sides right now, just elucidating) is that there's a heavy gameplay cost. A new player will have to do all that paperwork and manually figure out how the thing works and only afterward have the tool that would have helped him learn the maneuvers. So from a pure gameplay perspective, the inverse is true. It's better to have the assistance as the player is learning and then remove the assistance as the player gains skill at doing orbital maneuvers. But then you're violating the intuitive technical progression. The two are in conflict and you (the game designer) need to choose what weight to give to which approach.

[MacroNova]

Good thread.

I said this elsewhere, but I really think that some of the early Kerbin Survery contracts need to be at lower altitudes so it's clear to and possible for the player to use a plane. Instead of "above 19,000m" make it "between 1000m and 8000m."

As for the landing legs, I assumed I was supposed to use them to land my Heatshield + Science Jr + Capsule, not to land the whole spacecraft engine and all. But I stopped messing with this when I ran into the same re-entry orientation issues as everyone else.

IMO, good career mode balance means I have the tools to do most contracts when I get them or shortly thereafter, and it means I only need to do a certain type of contract a couple of times.

[Riisos]

I do wish some of the techs were organized a bit differently, although I'm grateful that I can see "down the tree". But I'd like to launch a probe into orbit before sending my brave Kerbonauts up, which requires quite a bit of science investment to get to a computer with SAS. I found myself having to basically farm out the science from a few zones, which is fine but I would have rather saved that for the fun of exploration. Also I wish the first plane tech node had a pure liquid fuel tank instead of the next one up. For career mode I like the fun of simulating a space agency, but orbiting a Kerbal before a probe or plane is up kind of ruins the immersion for me so I went back to sandbox. Don't get me wrong I am absolutely thrilled and love this game, I just wish there were maybe a few more branches in the tree rather than having such deep lines.

[sherkaner]

Just as a resource for this particular discussion, I might refer people again to this thread from version 0.90, which I find still applies to 1.0:

http://forum.kerbalspaceprogram.com/threads/99521-A-more-intuitive-tech-tree

Lots of good discussion there, which mostly ends with the conclusion that a more granular tree (more control over unlocking specific parts rather than arbitrary collections of them) would be more fun and cause fewer frustrations with game progression, without making the game particularly easier.

It sounds like mods to the tech tree in 1.0 are much easier now... but I have to admit that I was really worn out by creating my 0.90 tree based on that thread and disappointed that it didn't seem to provoke any rethinking of the tech tree philosophy in the stock game. Maybe somebody else will pick up the torch and create something for 1.0, but I'm still hoping that Squad will reconsider the design of game progression in the near future. I'm really enjoying playing with the new features and aero in 1.0, but I find that career mode itself still very much feels unfinished.

[giltirn]

I do wish some of the techs were organized a bit differently, although I'm grateful that I can see "down the tree". But I'd like to launch a probe into orbit before sending my brave Kerbonauts up, which requires quite a bit of science investment to get to a computer with SAS. I found myself having to basically farm out the science from a few zones, which is fine but I would have rather saved that for the fun of exploration. Also I wish the first plane tech node had a pure liquid fuel tank instead of the next one up. For career mode I like the fun of simulating a space agency, but orbiting a Kerbal before a probe or plane is up kind of ruins the immersion for me so I went back to sandbox. Don't get me wrong I am absolutely thrilled and love this game, I just wish there were maybe a few more branches in the tree rather than having such deep lines.

Just a quick comment; you could use a regular fuel tank and remove the oxidizer. I expect the plane fuel tanks are lighter, but basic tanks should do the job.

[tater]

As for maneuver nodes not being unlocked at the beginning. Well I can understand that. The Gemini and Apollo astronauts had to work everything out on a piece of paper. It is only in the age of the microchip that computer systems can give a good representation of when they should be making what maneuvers.

Mission control usually worked them out on paper and sent them to the spacecraft (except Buzz Aldrin, who did his PhD in orbital maneuvering/rendezvous). The current "no nodes" method is the sort of "seat of the pants" stuff that was simply not done (distant visual range rendezvous "by hand" was tried with Gemini, and they used FAR more propellant than they did when they planned the maneuvers). The current node UI represents a combination of pilot and ground control working these things out, and it doesn't make sense to me to have it off, well, ever.

[Riisos]

Just a quick comment; you could use a regular fuel tank and remove the oxidizer. I expect the plane fuel tanks are lighter, but basic tanks should do the job.

You are right, that would work, but the OCD designer in me still wants that dedicated tank, hehe. I could make it fly with a rocket engine too but it's just not meant for it ya know? Totally a good thought though, thanks.

By the way does anyone know what happened with the aerodynamics to affect what I'm going to call "cruising altitude"? Even with the basic jet engine ("Wheesley") I used to climb to a higher altitude to get a better cruising speed for longer flights, but now my designs seem to run slower at higher altitudes. Is this the effect of thinner atmosphere starving my engines or something?

[Scorpians]

I agree, i noticed i unlocked the 2.5m converter part but i was still along way from 2.5m parts.

[Robotengineer]

I would like to add that survey contracts should be limited to an altitude below 13km before the turbojet is uncovered, I have tried to do a contract requiring a survey above 18km with a basic jet engine and stalled out around 14km. On the patched conics side, I was able to do a rescue mission, Mun landing and Minmus landing all by eyeballing the orbits, I like the challenge of having to rendezvous without patched conics, so I would rather it be a difficulty option than just starting out with patched conics for the Kerbin system. Other than those things, I completely agree.

[MrCynical]

I definitely agree with the point the LV-909 is now too late in the tech tree. Yes, you can get into orbit without it, but it requires clumsy designs with upper stage engines as big as the launch stage which isn't a great arrangement to expect a new player to come up with.

For similar reasons I'd suggest moving the steerable fins one tier earlier. With only the capsule torque and a gimballed engine for steering rockets big enough to get to orbit are very vulnerable to flipping over in the lower atmosphere. The fixed fins don't help all that much - even a few degrees off the vertical and it becomes impossible to stop the rocket pitching over towards the horizon, usually on a much too shallow trajectory to reach orbit. Now yes, it is possible with practice to fly those designs, but you again have this back to front difficulty progression where the first rockets a new player will fly are horrible wobbly things which are a pain to steer even for an experienced player. They have to get out to a tier 4 node to have the parts needed to make a rocket that's reasonably easy to fly.

[nadreck]

Hmm, I had read the comments about the heat shield before I had a chance to play KSP 1.0 at all. So rather than jumping right in with career I played with Mk1 re-entry in sandbox until I was comfortable with it (3 suborbitals and 2 orbitals) then sat down with a career set to 50% for each of the reward sliders but 100% for the penalty ones (so quite close to vanilla Hard).

1st 'flight' was capsule on the pad with goo and Jeb getting out for eva report "flying above" and on the pad (also counted as first flight). That gave me enough for first level of the tree.

2nd flight was Val orbiting using the Reliant in two stages. Had to go back to space centre while in suborbital just above 70km to accept the contract to orbit Kerbin.

Unlocked next level, upgraded the astronaut complex and orbited Jeb while trying to build to a Mun rocket. No fatalities but had a couple mishaps with design until I came up with one using the first gimballed engine on a pair of side boosters, with something like my 1st orbital rocket (with the 1 ton tanks now though to keep the parts count down to 30).

By the time I had finished a bottle of Dead Reckoning that I opened as I started my career mode Jeb was onboard Salsa Shine 1 flying on an orbit that will probably take him through the Munar SOI. The Kerbal calendar was just under 3 hours into the first day.

Plans for my next session are to upgrade mission control and have Valentina fly enough contract missions (combining as many contracts as possible) to build up the funds to upgrade the tracking station so Jeb can plan his Munar Flyby with a tracking node as he only has about 300m/s of DV left according to my calculations and I don't think that by dead reckoning (even if I drink two bottles of it) I can bring him back to an atmospheric entry course on that slim a budget. If I can accomplish that all in a 2 - 3 hour session of game play I will have everything I need to do a first Minmus mission with landing.

The minmus landing and hopefully enough lucrative contracts to afford upgrading the VAB and R&D center in the 3rd session, and enough science to bring me up against the wall of what I can do before upgrading the R&D centre and VAB, but then that step gives me surface samples and more complex vehicles.

To me that seems reasonable progression for my level, obviously someone new to the game would not be getting there that fast the first time, but all the necessary steps to get there would have that new space capsule smell. So I don't think game progression is broken, nor, to me is the tech tree.

A newb, encouraged to play around in sandbox and maybe with tutorials, should be able to, on vanilla 'normal' career, make progress without having to create new careers over and over or get frustrated.

[Flenser]

I think that the OP hit this dead on. I think that the issues could be broken down into two components: Techtree balance, and Facility upgrade balance. Looking at the facilities, each facility has three levels, with the updates ranging from moderately expensive (Administration) to prohibitively expensive (R&D). Furthermore in some cases the upgraded building is overpowered vs. the time when it's first needed. For example the 30 part limit on the VAB tends to come into play around the time that the player is about to attempt a mun landing, upgrading the VAB increases the part count to 255, and adds the basic action groups. At the point that the upgrade is needed simply increasing the part count to 100 would be more than sufficient. I have two thoughts about how the facility upgrade could be addressed:

1. Increase the number of facility levels from three to five. This way the level two upgrade become more affordable and not so over-powered.

2. Breakup the building upgrades based on features. Returning to the example to the VAB this would mean that the upgrades to the part count could be purchased separately from the upgrades to the action groups.

Option 2, gets interesting when it's applied to R&D, instead of having the upgrade raise the maximum science level from 100 to 500 for everything, an upgrade could raise the maximum science level for a subset of the nodes in the tech tree. So a player which is focusing on space-planes could choose the level2-aircraft upgrade which would allow the nodes related to aircraft (basically the Mk II) components to researched but leaving the rest of the 160, and 300 science nodes locked. Purchasing the level2-aircraft upgrade would make the level3-aircraft upgrade available, purchasing the level3-aircraft upgrade would allow the player to have the MKIII components which potentially leaving the rest of the techtree locked at level 1. It could make for some interesting choices.

[ArgenTum]

I agree with everything in this thread. Career mode needs fixing. I wonder if Squad ever really sat down and though hard and long on what they want with career mode. Should it be about managing money and/or science? Or is it a kind of "tutorial", to make it easier for new players to get into the game? Or something else? It's hard to properly balance a game, it you don't have a clear vision for what you want with that game mode.

Now, 2 suggestions:

1. Why not put contracts in the tech tree, so if you go for "rocketry" you also (apart from the actual parts) unlock contracts about mun and minmus. Or if you go for "flight" you unlock contracts about surveys. That way it would be clear to a player what you are expected to do with that research node and at the same time keep the list of contracts clean from contracts you can't really complete. (the unlocking contracts should be shown as a thumbnail before you unlock the node, just as parts are today. Maybe the node would actually be called "moon explorations".) this might also make it easier balancing the tech tree. "When you unlock this node, all the parts should be there for you to make it to the mun."

2. Maybe science and money should be more exponential, where early on you count you science in tens, but soon the nodes costs thousands and perhaps even millions. Then Kerbin experiments could give access to early tech node, but you could never grind to level 2 or 3. For that you would have to complete higher tier contracts (unlocked in tech tree) or go to the mun or Minmus. And then for level 4 or 5 (or whatever) you would have to collect science from Eve or Duna, or from even higher tier contracts.