Page 1 of 1

Balance Suggestion for next turn

Posted: Tue Jan 19, 2021 17:37 UTC
by BBB
Hey All,

I wanted to follow-up on my previous balance thread with my suggestion for how I think the balance of the game could be improved to make it more fun. I'm certainly not pro at this game so happy for any feedback.

What makes games fun:
- Feeling like you have lots of different credible options
- Not feeling like you're being overly punished for bad decisions
- Feeling rewarded for taking actions

Problems with current game balance:
- Compound interest is clearly the best way to accumulate wealth/score
- The game is likely solvable in the sense that there is an optimal way of playing which is maximizing trading/loans early followed by banking. If you don't follow this exact path, you will lose.
- Once you follow the optimal path, the accumulation of score is driven by the ticks of the clock rather than specific actions, meaning subsequent actions have less value and that it is virtually impossible to catch up to a player who's ahead of you unless they themselves make a big mistake.

Right now the choice between building planets and banking are fundamentally broken. Referring back to my previous forum post, colonists are barely worth more than the credits needed to obtain them, and probably convey a negative value add once you account for sector defenses and lost opportunity cost of equivalent trading. If you engage in planet building in this game, you will fall behind, and you will never catch up to a player who simply trades and keeps all of their credits at a single bank planet system. While an argument could be made that planets are needed to secure trade routes, there are currently too few players to make securing routes particularly critical, and also there's only a limited window where it is both feasible to construct many planets, while it is also profitable to keep trading. By the time you can secure enough trade routes for it to convey a strategic advantage it's no longer worth trading, and your investment will largely be a waste of time as you've now fallen behind the player who just banks their money.

Furthermore, trading is also broken as it is not a viable long-term strategy. Once you hit level 18, you will actually accrue more wealth trading at that level and letting the level 19 upgrade cost accrue interest as compared to continuing to upgrade. One could argue that hull upgrades would count towards planet building, however I would refer you back to the previous paragraph. Trading is only an annoying stepping stone rather than a viable strategy.

So, how do we make the game more fun:
- More credible options:
The important factor here is making sure that there is an equivalence between various strategies. These strategies, generally speaking are:
Trading
Banking
Planet Building

In order to "balance" these things the rate at which they generate value should be ordered and be proportional to the effort needed to engage in that activity and risk of doing so. This is the opposite of what's described above, since the safest option (i.e. only having a single well defended bank system) is also the most profitable. The best way to change this is to rebalance the production rates accordingly. Noted from previous post that there was concern that too many credits were being obtained too quickly, therefore I would err on the side of reducing production rates rather than increasing.

I would suggest the following:

Planetary interest rate should be reduced to 0.005 and IGB should be 0.002.

This would have the following impacts:

- Hull upgrades and trading are now economically feasible up to Level 30 rather than current Level 18. It continues to make sense to upgrade your ship hull purely from a trading standpoint, which opens up the possibility of planet building.
- By allowing trading to be feasible it creates an incentive to build planets because people will still be interested in trading later in the game, and holding territory and being part of clans will be more valuable since port degradation will start to play an effect.
- A fully loaded planet is now worth a maximum of 4.5B, or approximately 45 credits per colonist. This means that the investment in building planets actually pays off as well.
- By making planets valuable, and incentivizing many players to have planets, this will require that people invest resources in keeping planets and incentivize players in building/securing planets. This will make the game more dynamic and strategically deep and open up avenues for minor victories and defeats which can swing scores.
- All of the above shifts focus of the game on actions which take turns.

Those are my 2 cents. Eager to hear your thoughts.

Re: Balance Suggestion for next turn

Posted: Wed Jan 20, 2021 20:40 UTC
by ProCra
Indeed, you have valid points. One single correction: the Colonist production of credits should be comparable, if not bigger, than the interests on both IGB or planets accounts. This way the game turns into having more planets and to secure them. And also the upgrades are more dependent on the planets than on the clicking on a link for a trade route...

Re: Balance Suggestion for next turn

Posted: Thu Jan 21, 2021 13:14 UTC
by thekabal
BBB wrote: Tue Jan 19, 2021 17:37 UTC - Hull upgrades and trading are now economically feasible up to Level 30 rather than current Level 18. It continues to make sense to upgrade your ship hull purely from a trading standpoint, which opens up the possibility of planet building.
...
- A fully loaded planet is now worth a maximum of 4.5B, or approximately 45 credits per colonist. This means that the investment in building planets actually pays off as well.
Interesting discussion and I have more to say, but first, could you share how you came to those two conclusions? Specifically which equations you used to determine both?

Re: Balance Suggestion for next turn

Posted: Thu Jan 21, 2021 19:18 UTC
by BBB
thekabal wrote: Thu Jan 21, 2021 13:14 UTC Interesting discussion and I have more to say, but first, could you share how you came to those two conclusions? Specifically which equations you used to determine both?
Ack! I did my hull upgrade math wrong in original post (think I was using 0.0025 instead of 0.00025)...but here goes with corrections!

For hull upgrades:
Calculations basically compares the incremental increase in revenue generation by trading vs the interest gained from the hull upgrade itself, normalized by ticks (as interest and turns both accumulate based on ticks). This answer sounds more correct based on how games typically go.

Cost of Level 25 Hull upgrade: 16,777,216,000

Interest accumulates every 2 ticks, therefore...
Interest / 4 ticks = 16,777,216,000 * (1.00025^2 - 1) = 8,389,656.58 credits

Incremental hull capacity increase @ Lv 25 = 1,081,900 - 746,100 = 231,600
You accumulate a trade route worth of turns in 4 ticks, approximate profit (including route degradation) = 23 credits / hull unit
Incremental increase in trade profit per 4 ticks = 231,600 units * 23 profit / trade route = 7,723,400 credits

Therefore after Level 24 you accumulate interest faster than you theoretically can accumulate additional trading revenue. An additional factor to consider is that this assumes 100% of your turns are spent trading, a factor should also be applied based on what percentage of your turns you spend trading which further helps the calc.

Value of a planet:
This calc basically comes down to seeing how many credits a colonist generates per turn vs. how many credits you'd need in your bank to generate an equivalent number of credits:

A fully populated planet maxed out for generating credits produces:
100,000,000 colonists * (90% factor for organics production) / 400 credits per 2 ticks = 250,000 credits
Equivalent banked credits to generate 0.25m interest = 0.25M / 0.00025 = 1.0B or 10 credits/colonist

New calc: Is creating a planet a profitable enterprise?

Assume that we're populating and defending planets, and that we're providing approximately half as many fighters as colonists (approximating 10M colonists & 5M fighters for safe planet)

Value generated = 10 credits / colonist (doesn't include colonist reproduction for simplicity)
Value (reproduction) = 0.000025 colonists / colonist * 10 credits / colonist (above) / 0.00025 interest rate = 1
Total value = 11 credits/colonist

Costs:
Cost colonists = 5 credits / colonist
Cost fighters = 0.5 fighters/colonist * 50 credits / fighter = 25 credits / colonist

Total costs = 5 + 25 = 30 credits/colonist

Profit per trade route = 11 - 53 = -19 credits / hull unit (defended)

Therefore populating planets is actively not profitable. You incur more costs than you get colonist production value.

Re: Balance Suggestion for next turn

Posted: Thu Jan 21, 2021 20:32 UTC
by BBB
I did a whole post, but then I automatically logged out and lost it all :'(

Here's the main points:
1. Game balance defined by interest rate, and the ratio between interest rate and colonist credit production rate.
2. Lowering interest rates extends time where trading makes sense.
3. Increasing ratio between colonist credit production rate and interest rate makes planets more "profitable"

Based on above I believe ideal numbers (or at least starting point) should be:

Interest Rate: 0.01%
Colonist production rate: 1 credit / 150 colonists

This results in the following:
- It is advisable to keep upgrading hull until around Level 27
- Based on previous defense assumptions planets result in a "profit" of approximately 45 credits per colonist per run, which is approximately twice as profitable as just trading.
- This will go down significantly as you start to count investment in things like upgrades / planets become population capped / needing to add more sector defenses. However, there is now a period where it actually makes sense to build planets.
- Because there is a point in the game where people need to generate fixed assets, that means people now have assets in a form which can be taken! There is now strategic thinking that needs to happen about how you defend your planets, should you take risks to pull ahead on credits, or overdefend? The more you invest the more you need to worry about protecting that investment!

Personally I think this would be better and more fun. Happy to hear mroe thoughts though

Re: Balance Suggestion for next turn

Posted: Fri Jan 22, 2021 02:28 UTC
by thekabal
Truly excellent work - #YouDidTheMath! I'm stealing your examples to help with future rewrites on the TKI/BNTA branch (see below).

Starting out with the background that the 'economy' in BNT wasn't so much designed as stumbled upon, it's been known for a long time that it's not ideal.

A long time ago, I ran a fork of Blacknova Traders (The Kabal Invasion/TKI), and in other times that was reverse-forked into BNT-Advanced/BNTA (it's still in the tree on Sourceforge if you care to go deep-diving). There were lots of differences, but the biggest was that Planets were truly separate from Ships. What do I mean?

I mean that today, and for most of the history of BNT, Planets actually 'share' or more accurately are given the same levels in some stats as the player's ship. These: computers, sensors, beams, torp_launchers, shields, armor, cloak. It's hard to imagine a sensor that can span the Universe, or shields that reach that far, or especially armor that is so large on a ship that it stretches to every planet that player owns, ha.

That creates a rather odd game balance where in short, you have to take on the player's ship to cripple his planets, and once you do - ALL of his planets are weak to those attacks. But if, as you said, the big players are hard to catch up to... then all of his planets become nearly untouchable too.

TKI/BNTA made each planet maintain it's own defenses. That increased the costs for players, making each planet truly independent, and as a result, you could actually have a "bank" planet with crazy strong defenses, and newbie planets that were just developing and very vulnerable. That increased the value to strategic play on the defensive side, and also decreased the impossible-to-catch big players since you could only feasibly keep up with so many planets. There were even more systems contributing to that, like our stars were variable-size, supporting more planets in some systems and less in others.

One other major difference is that BNT uses POW() for most levels, where BNTA used essentially *100. That helped again with preventing the runaway crazy top players, and also kept the numbers from getting so large that PHP & MySQL can't keep up without entirely punishing the player progression. This was incredibly controversial, but another fork by a different programmer, Alien Assault Traders did similarly with even better success.

An interesting aside is that we also proposed changing both of the variable-groups you pointed to - bank interest, and trading values. Trading values turned out to be devilishly difficult to tweak, despite clear values for each commodity. The logic is rather brittle there. But the bank interest was easier! We proposed changing that and the inventory factor. Let me be blunt and say the community lost its effing mind at those suggestions. Like, full-scale revolt and demands for servers that did NOT change those settings. For most of the TKI games I ran, we actually disabled the bank entirely as a result. People seemed to prefer that, as silly as it sounds today.

I bring these up because I think your points, examples, and conclusions are valid. Other programmers (including me) came up with different solutions to different problems that overlapped in some places, and I thought it might be interesting for you to know how those went.

Now, all of that "history" out of the way, back to your points. The economy in BNT needs a tremendous overhaul. Even approaching a rewrite in parts. Sadly, development on BNT is almost non-existent, with TMD focused on maintaining the current code while spending more time on his new BNT project. The last public/published code changes were over 7 years ago and were from (checks Sourceforge) me.

Similarly, my TKI work is bogged down in architectural fixes. Making the codebase correct, using modern development methods, and trying over time to fix it enough to allow for the kind of large changes needed in the present/future.

Yet, I support your suggestions. I think they would improve the game. Plus they are only a change in configuration, not code. No heavy lift. Also, with only 20 players even on the main game here, I suspect the outcry would be much smaller now than back in the day with over 1,000 active players in some of the main games.

It's not my call, as TMD runs the game here, but again, I think you've made a persuasive case with solid math backing it up and it's an easy change that can be reverted in a future round if it doesn't go well.

** Special nitpick note: TKI/BNTA was coded to run in either "Classic" mode which played identically to BNT (it was the same logic/code) OR "Advanced" or TKI mode which had the different behavior listed above. It's confusing because we're talking about a decade-long project moving away from BNT and then back. The current TKI is identical in playstyle to BNT, since we reverse-forked years ago, but in the future, if I can ever make it out of the pit of fixes, will be back on the plate. There were lots and lots of crazy changes in BNTA that I'd love to revisit. Spies, Diplomats, new devices like the Last Ship Seen Device, different ship TYPES like the original Tradewars 2002 had.. the list goes on and on, and was honestly so much fun.

Re: Balance Suggestion for next turn

Posted: Fri Jan 22, 2021 16:05 UTC
by BBB
Thanks for the feedback!

My gut feeling is that somebody may have done this math before aha. I didn't realize it at the time but my suggestion is basically just a 2.5x factor of the "vanilla" economy settings that you find in the FAQ (i.e. (1/67) / 0.00025 ~ (1/150) / 0.0001) ~ 60. Since my main conclusion was that planet building balance is defined by that ratio, I'm really just reverting back to the original settings, with a slightly slower ramp up.

It was mentioned in the previous thread that the colonist production rate was scaled back to reduce growth, I think the issue was that the interest rate was not scaled back at a corresponding amount to protect that important ratio.

Thanks for the support! I was hoping this would be an easy-enough suggestion to implement, I'm hopeful to see it get implemented and see how it plays out.

BTW - I'm not sure if it was your version, but I do remember playing a game mode where it was possible to invest in different base upgrades / etc. I do like that idea a lot, however I see the point that these kind of fundamental changes are much more difficult to balance.

Cheers,