Task #18

Develop a Research Tree

Added by vomov almost 5 years ago. Updated over 4 years ago.

Status:Feedback% Done:


Target version:v0.9.0


The original outpost had a broad research tree, which encompassed everything from astronomy to political science, as well as the never-ending 'basic research'. Adding this to the game has two, possibly three steps:

  1. Mapping the research tree has already been done (found it in a FAQ by a guy named Joerg Stenger). This is a text file now, but transferring this to a proper format is easy.
  2. Not all effects of the different studies are known. Some might be obvious; alien pharmacology cures the alien virus, for instance. However, it is not always clear when which buildings become available. At least one playthrough is required here. Other studies yield robots (airbots?), that as far as I know, don't really do anything. A comprehensive list of effects would be good, also in-game.
  3. Next up, the potential for restructuring, condensing, expanding, or otherwise changing the research tree is there. For example: the 'alternative advanced systems' study yields geothermal power. Why not name this 'geothermal power', or geothermal studies'?

Research tree.txt Magnifier - Original research tree, (5.05 KB) vomov, 11/09/2015 03:27 PM

Research tree.svg (510 KB) vomov, 12/08/2015 03:39 PM

Research tree - new.pdf - The new research tree (558 KB) vomov, 02/24/2016 07:07 PM

Buildings.pdf - A list of building levels, based on their tiles, with the associated prerequisites (106 KB) vomov, 02/24/2016 07:07 PM

OutpostHD research tree - v0.2.pdf (602 KB) vomov, 06/19/2016 06:53 PM

Buildings - values.xlsx (20.9 KB) vomov, 06/19/2016 06:53 PM


#1 Updated by vomov almost 5 years ago

I've included an SVG - file, with a simple chart on what appears when. I'm sure I didn't get all the dependencies, but the most important stuff is there.
After two playthroughs, during which I made some notes as to when stuff becomes available, I produced the following: there's a LOT not implemented.
Even worse: most research gives new sprites, but it does not seem to have an effect; factories don't seem to produce more after reaching the result 'improved factory efficiency'.

Based on this, we can see what was implemented, and what was planned but not implemented. I would like to propose to remake the research tree, based on the original. I'd be willing to rework the current tree, and figure out what goes where, with what effect(s).

#2 Updated by leeor_net almost 5 years ago

I'm agreeing with a rework. As stated in my PM to you on OPU's forums, I have the official strategy guide which goes into more details about the research available but most of it is just superfluous fluff that doesn't really mean anything.

Taking a cue from the original game we should develop a new research tree with all stated effects, etc. Any time a building is upgraded (e.g., new sprites used), it should have an effect. This is easily implemented in the logic for a particular building.

Basic research is also never explained much in the strategy guide except that AI personalities can't be assigned to it because it involves thinking that is too abstract. It also describes the basic research as a 'think tank'. It has the effect of reducing research times for all other research which I suppose is a good effect but I would just wrap that into another applied research field (e.g., improved laboratories, improved computer assistance, etc. which would have the effect of increasing research efficiency).

#3 Updated by leeor_net over 4 years ago

  • Subject changed from Create research tree to Develop a Research Tree
  • Priority changed from Low to Normal

#4 Updated by vomov over 4 years ago

I've been reworking the research tree, while keeping the original tree as much intact as possible, but at the same time making it more understandable and realistic. Note that I've done this in small parts, so it's likely some details are wrong or don't make sense.

The starting assumption was that every field was a set of branching trees, with not too many cross-tree interactions. Notable exception; the Aerospace sub-tree might be better suited to be a tree by itself (while not being accessible until the research has been done). Also, I used the tiles of the buildings (half of which were unused in the original, I might add) to figure out where a study is needed.

I've also been thinking about certain game mechanics, such as terraforming, aerospace, and the elusive 'arts' tree. Some of my suggestions have been worked in this temporary version. Second, the organisation of the studies, their names, and their ad-libbed descriptions are also in flux. So, please comment on everything :)

#5 Updated by leeor_net over 4 years ago

  • Description updated (diff)

#6 Updated by leeor_net over 4 years ago

Kind of surprised I haven't looked at this sooner.

Anyway, some thoughts.

First and foremost, excellent work with the original research tree. It's quite thorough and may be useful as a resource for the OPU Wiki in addition to helping us make decisions with how to move forward with OutpostHD.

To address your comment about cross-tree interaction, I'm going to recommend we eliminate cross-tree interaction as much as possible. It may be an over simplification but I would see some branches (as you mentioned Aerospace) have certain requirements be met before becoming available for research. That's just my thinking on it from the stand point of programming it however I'm open to any counters to this. Programming isn't so much the hard part as is the design and logic but if I have clearly defined rules and parameters is makes developing whatever we come up with much easier... so really what I'm saying is I'm speculating and imposing artificial restrictions without needing to.

A few notes:


The 'official strategy guide' states this about basic research:

In Outpost, an active program in Basic Research will generate beneficial results for your entire research effort. About every 100 turns, progress on all current research will accelerate by 5 turns, but the minimum time required to complete any line of research will still remain at 25 turns. So, if you have 30 turns remaining before you discover your first terraforming microbe, and 100 turns of active Basic Research have been completed, then you'll only need 25 more turns to make your discovery.

Assuming this is how the game actually works, I'd call this pretty useless. If we wanted to leave it in there then we could use it as a Morale booster or have it change efficiency modifiers around the colony at random (e.g., basic research yields a 0.5% decrease in requirements for Common Metals across the board or 2.5% less food waste, etc.)

Actually everything else looks pretty good though looking through the official strategy guide a lot of the research is pretty ineffective (slight boosts to morale, etc.) and prerequisites for launching another starship. I suppose this is a reasonable end goal but I've always envisioned Outpost as being open ended so for OutpostHD we could probably strip out all of the stuff related to aerospace except for launch vehicles for putting new satellites into orbit, etc.

This is going to be an interesting balancing act the more I think about it.

#7 Updated by leeor_net over 4 years ago

Okay, so I just realized I entirely missed the proposed new research tree.

I'm looking over it now and I like it a lot, actually. It's much more defined as to what the individual things do and I like the addition of things like automatic trucking, asteroid capture for resources, etc.

This works well I think. What I'd like to see is a definition of the benefits for the end result of research... e.g., what does Spew 2 or Factory 3 do? Increased efficiency and durability, sure, but any specific amounts? Reduction in cost of construction? Etc.

Nice work! I really wish I'd commented on this sooner!

#8 Updated by vomov over 4 years ago

It's not finished yet, but I thought I'd put it up anyway; some stuff has a description now, and I've done some basic balancing calculations with a reduced number of possible values.

I've found the fancy flowchart a bit hard to manage as information gets added, so I've made a table with the 'for now' guesses.
However, as buildings are changing their input requirements and output over the course of the game, I'm not sure this table will be appropriate or sufficient.

#9 Updated by leeor_net over 4 years ago

Defining actual resource values may be a bit premature but defining the value modifiers makes a lot of sense.

Storage capacities also make sense.

Later on we can define actual value for resources broken down into specific types. I've found it very, very difficult to properly balance this even in this early stage of the game. What I initially thought were reasonable values turned out to either provide a LUDICROUS amount of resources or were completely insufficient such that operating two structures was enough to drain a players resources within a few turns.

Also available in: Atom PDF

Redmine Appliance - Powered by TurnKey Linux