Recommended Posts

As has been said many times, the game needs a technology tree to make researching more satisfying. So I made a simple example of it. Each blueprint can be unlocked when all its roots have been discovered. In addition, the chance of each discovery depends on its position on the tree (the more to the right, the less chance - more valuable blueprints).

Each research event (RE) randomly select one of currently available blueprints. Let's see an example. At the begining you've got 100% chances to discover tethers (game should make one artifact accessible for the player, near the landing site - it would also work as a really basic tutorial). Once you unlocked tethers (RE-1),  five new blueprints became accessible. Each of them has a weight of 1/2, so they are equally likely to be unlocked in the next RE (RE-2). Let's assume that you get power cells and next, RE-3 takes place...

sample results:

Spoiler

 

RE-3
input: wind vane (1/2), solar (1/2), generator (1/2), filters (1/2),  battery (1/4)
output: solar

RE-4
input: wind vane (1/2), generator (1/2), filters (1/2), battery (1/4), solar panel (1/32)
output: generator

RE-5
input: wind vane (1/2), filters (1/2), battery (1/4), solar panel (1/32)
output: battery

RE-6
input: wind vane (1/2), filters (1/2), big generator (1/16), solar panel (1/32), big battery (1/64), 
output: ...

and so on.

 

 

tt.thumb.png.33823e161d7fb62a4f9eebe0c1dfb299.png

I do not know if the weights of successive levels are well balanced, but I think this concept makes sense. What do you Guys think about it?

Edited by gp.

Share this post


Link to post
Share on other sites

Forgot, cant edit. Recently, augments have been implemented and more of them are being planned for vehicles and platforms. Since they are not essential for the game, I think it would be an interesting idea to spread them in a solar system as special artifacts. Augments for different techs (on chart: yellow, red, blue, purple - and more as the game grows) on different planets...

Share this post


Link to post
Share on other sites

It could have separated the technology tree based on power generation type, as the solar panel as it a passive generator which also relies heavily on battery, it should be on the same branch, while generator in the other hand is an active generator, in which it power up the base when it is needed to.

Feedback aside, I do agree with this, researched technology should have a tree, so we won't keep getting researched that's useless to us in the current state

Share this post


Link to post
Share on other sites

I like the idea that you'd progress your understanding, using your progression from power cells to larger battery as an example. It makes sense you'd figure out the basics before scaling it up (or down).

Share this post


Link to post
Share on other sites

I am kind of mixed about this idea, tech trees are common in quite a lot of games already. Although, I do feel learning by pods from the ground is not a good thing for the game to have either. I think only time will tell what may happen. You have a good idea, don't get me wrong, but I just feel there are already enough games with tech trees.

Edited by Knouvael'Qeskre

Share this post


Link to post
Share on other sites

@Knouvael'Qeskre I don't want to put this tech tree in game's UI. It's just to show my view on how research system should work. It would not change anything about researching, except of its results.

BTW: if anyone have any suggestions or want to make his own version of this tree, you can edit this: link

 

original version:
https://www.draw.io/
file: tt.xml

Edited by gp.

Share this post


Link to post
Share on other sites

I like this idea a lot but how does it fit with the various chest types?

For example, if I land in an area with an abundance of "trees" and collect 30 chests, 2 each of the 15 types, would that be 30 RE's each with a blueprint as the output? Or is there a weighted chance to get a blueprint, altered by how many times you have researched the same chest?

Share this post


Link to post
Share on other sites
On 4.05.2017 at 5:19 PM, gp. said:

Weight in my tech tree were to keep equal progression in different branches of it. So 1/32 or 1/512 doesn't mean absolute probability, but is information that one tech is more likely to be discovered than another. So every research would give you next tech.

Let's say you've got:


solar panel 1/32 = 2/64
big battery 1/64
rover 1/8 = 8/64
habitat 1/64 
fuel condenser 1/32 = 2/64

sum: 14/64

Absolute probability of them:


solar panel 2/14
big battery 1/14
rover 8/14
habitat 1/14
fuel condenser 2/14

I guess I didn't describe it exactly.

I posted this in another topic. So..

10 hours ago, SiBrit said:

I like this idea a lot but how does it fit with the various chest types?

For example, if I land in an area with an abundance of "trees" and collect 30 chests, 2 each of the 15 types, would that be 30 RE's each with a blueprint as the output? Or is there a weighted chance to get a blueprint, altered by how many times you have researched the same chest?

In my design, chest type doesn't matter, every one gives you a blueprint. But, now I think that's too easy. I guess there would be two solutions for this problem:

  1. Either, there are two large categories of chests: with (maybe tech-chests from wrecks) and without blueprints.
  2. Every type of chest has assigned counter value which increments after chest being researched. This counter is also a probability factor for resource instead of blueprint.  
    In a given example:
    solar panel 1/32 = 2/64
    big battery 1/64
    rover 1/8 = 8/64
    habitat 1/64 
    fuel condenser 1/32 = 2/64
    
    chest-specific counter: 4 (2^2 - chest was opened twice; if more: 8, 16, 32 and so on) / 64
    sum: 18/64

    absolute probability:

    solar panel 2/18
    big battery 1/18
    rover 8/18
    habitat 1/18
    fuel condenser 2/18
    
    resource drop: 4/18

This way development of technology is gradual and logical. Also, exploring for different artifacts is beign rewarded.

Edited by gp.

Share this post


Link to post
Share on other sites
On 5/1/2017 at 6:58 PM, Knouvael'Qeskre said:

I am kind of mixed about this idea, tech trees are common in quite a lot of games already. Although, I do feel learning by pods from the ground is not a good thing for the game to have either. I think only time will tell what may happen. You have a good idea, don't get me wrong, but I just feel there are already enough games with tech trees.

Tech trees are used so much because they are logical.  They make sense.  Doesn't matter how many games use them.  There's no point in being unique if the base of the game progresses in a completely chaotic and random path which can lead to important tech not being available through the exponential reduction of chances to get what a player needs.

 

@gp.

I'm sure your layout of probability makes sense to most people, but all those numbers just leave me scratching my head so I don't know how to respond to it.

Share this post


Link to post
Share on other sites

Long story short: you need to 1) bring fractions to the same denominator and 2) add their new counters (and omit the denominators). 3) A new integer (let's call it E) is created which is the denominator for 4) new fractions. They determine the absolute probability of researching specific blueprints.

In the initial stage we had for example wind vane, solar panel, generator, power cells and filters, each of which weighed 1/2. (1/2 + 1/2 + 1/2 + 1/2 + 1/2 wouldn't make sense)
They already have the same denominators, so we add only counters: E = 1 + 1 + 1 + 1 + 1 = 5. Then we create a fraction of probability for every blueprint: using E = 5 as a denominator, and their respective counters (in this example, all equal 1). We get 1/5, which now is a floating point number (0.2) and means absolute probability of specific blueprint being researched.

Change I suggested a moment ago:
After step 2) we also add up E and a value of a chest-specific counter of reopenings. It equals 0 if chest has never been researched. And it grows exponentially (1, 2, 4, 8, 16, 32, 64...). Steps 3 and 4 are the same.

Spoiler

 

So, in this example: [reopenings][counter of reopenings]    :    [probabilities of getting ANY of listed blueprints]  [probability of getting resources] 

[0][0] :  [5/5]  [0]
[1][1] :  [5/6]  [1/6]     (E = 5 + 1)
[2][4] :  [5/9]  [4/9]     (E = 5 + 4)

 

As English is not my native language, it's hard for me to explain it clearly. But that's just one  idea to make progress in game satisfying. There sure are many more and better of them. Hope devs know what they're doing :)

Edited by gp.

Share this post


Link to post
Share on other sites

I appreciate your attempt to help me understand.  Thank you.  You have my respect for knowing as much English as you do.

Share this post


Link to post
Share on other sites

So... today I got really bored and wrote a simple c# application that simulates the operation of my algorithm.

Spoiler

app.png.0a872b1e124533c3fa86f08b1beb1728.png

If anyone is curious about the results, there they are:

tests.thumb.png.4a21d23cfebf4aeeb46505f927379dc3.png

based on: https://tinyurl.com/techtreeastroneer

Edited by gp.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now