I am also astonished to find that people only just discovered the empire modifiers on the later singularities; I've been stacking those for aeons. It's clear then that with additional slots the individual effects have to be reduced.

I would chalk this up to the fact that the "on empire" effects are not visible in the tech tree. Atleast when I did a testing game, they were not visible to me. Only when I activated the singularity and moved my mouse over the effect in the singularity UI on the top left, did I see the effects. I'm also not a Riftborn player so it seems like I missed the changes in the patchnotes. 

But then again, we didn't exactly have a big discussion about balance like this in a long time (or ever?). And maybe those who like to play Riftborn rather kept their mouth shut and enjoy huge bonuses in multiplayer for their personal gain. 

For singleplayer the AI isn't a challenge anyway, so a change like this is rather inconsequential. 

I just hope that since we have discovered a huge issue like these Riftborn empire bonuses, or the often-mentioned and completely unbalanced juggernaut modules for factions like Riftborn or Cravers finally get adressed. 

Even if most of us play singleplayer and enjoy their favourite faction being changed for the better, in the end when we want to try something else, or play multiplayer, these huge imbalances are be a big issue. I already provided proof. Just do yourself a favour and watch the end of Koxsos youtube recording again, where he had no chance against the Riftborn and looked astonished at the empire graphs. 

Getting some improvements for your faction which are well thought out like the recent changes to Lumeris, Horatio, UE and so on are appreciated. But if issues get not adressed, players will start having less fun because the game is no longer challenging vs the AI. Its simply not fun to win every fleet battle with those juggernaut modules, or be ahead in science in every game because of singularities, even ahead of science specialized factions. (This situation arises whenever you get access to developement grants by the way, which should have been nerfed long ago.)

In the past we have had issues like these before. Think of "infallible authority" or whatever the Vodyani hero skill was called. Or think of stacking infinity shields on Craver carriers. These issues are comparable to what we havenow. They make a faction so good that it simply becomes opressive. Regardless of how much you like your faction and want it to be competitive and win over others, you have to realize that some things are simply bad for the game and must be changed. 

More inflation feedback from reddit.

Not entirely sure what's going on, except buy low, sell high taken to extremes is clearly a problem.

Image 1.

Image 2.

Wooh, so I experimented with this "high inflation one turn, low inflation the other" and wow... So on turn 150 I was able to buy heroes for 200 dust and such... I literally bought everything on the market one turn, sold everything the other. Also, upgrading trade companies was super cheap too during the right turns... I had 90 trade ships in all 3 of my trade companies in a matter of a few turns after I realized how to exploit this mechanic...

All in all, I have no idea how to tweak this to work right, just wanted to share an ignorant player's experience with you.

Inflation is based on the total dust per turn of all empires last turn (any single transactions of dust e.g. market have no effect)

There are two things creating this issue with inflation oscillation:

Primarily it’s that only the previous turn’s dust/turn is used, if the game was to use an average of the last 10 turns (possibly with a higher weighting on the more recent) then the oscillation would likely be marginal or happen slowly over many turns but unlike before the oscillations magnitude would decrease each turn instead of increase.

The other issue is the minimum price of market resources is 40% of their base price * inflation (This is personally my (& others) biggest single issue with the games economic system by far)

This is what allows for a ton of dust to be made and for the whole purpose of these inflation changes (to make pacifism hurt militarists) to be completely sidestepped & thus be ineffective.

Everything should just have a minimum value of 1.

Quickly threw together excel sheet for inflation and the results speak for themselves. (Top two are using G2G Mod's inflation model)

Feel free to play around with the spreadsheet with better values: Inflation.xlsx

(y axis is logarithmic scale)

This is inflation of 300 turns under very simplified condition (Revenue Growth = static growth of 5%/turn, Upkeep/Revenue Ratio = 0.2). Although I'm yet to look into your sheet thoroughly (it's monday morning here :/ ) , I think we're doing quite same thing. Can't understand why things gone so different :(

Well, I used a bunch of mostly random numbers to create the dust income/upkeep costs which certainly would be one major difference. (notice how mine go below x1.00 inflation, that's poor numbers fueling the graph)

How did you model the 10avg? I ended up doing it by calculating the inflation as normal for that turn but then averaging the result with the inflation with the previous 9 turns (if there were not 9 previous turns then It just 1 for each missing turn) since the average was done with equal weighting would not expect such a result as yours as averaging should continually dampen oscillations.

e.g. avg of {0.25, 7, 0.25, 7.5, 0.25, 8, 0.25, 8.5, 0.25, 9} = 4.125

next turn: avg of {7, 0.25, 7.5, 0.25, 8, 0.25, 8.5, 0.25, 9, 4.125} = 4.512

Not the best example as it does not add the new one for that turn via the original way but I think it somewhat shows that averaging should make inflation tend towards the point where it oscillates least.

When I first looked at your graph it did not make sense as the 10avg is reaching the same peaks as the vanilla which given how averages work should not happen, for it to reach the same peak, especially after a few turns at the minimum the value for that turn would need to be far beyond that peak the pull the average up for every 1 it moves up from the last turn the sum of the averages parts would need to increase by 10.

Also, I noticed that the numbers on your table for 10avg are the raw inflation, not the avg(10).

You have also calculated it different from me. I did not have two different inflation columns, there was a single inflation column which both worked out the standard inflation and then averaged it out with the past 9 turns.

Inflation(N turn)= AVERAGE(MAX([NetGlobalDust(Turn N-1)]*(0.002/12),0.25), [Last 9 inflation points])

I have managed to get minor oscillations with 10avg but they have been about 2 magnitudes smaller than the g2g version produces, and also require crazy high pre-inflation upkeep costs (levels I’ve not seen in a game ever) I can't even put them on the same charts due to the massive difference between results

To make there graphs Dust increased by 100 a turn with the increase increasing by 5 a turn.

Upkeep increased by 50 a turn with the increase increasing by 2.5 a turn.

Which lead to a dust/turn of 5,331,300 & an upkeep of 2,649,750 (pre-inflation) both of which don't happen in-game.

While I don’t think average10 is perfect it’s magnitudes better than the current system and if it breaks then it’s most likely a different bug that allowed crazy rates of upkeep or income.

You went all out with the mathematical formulae, it all looks far more complicated than it is once you read it. I do like the Expected Inflation formula, it’s simple and works better than the others.

The main issue (as with all of them) is that it requires hard code changes to implement, but in the case of this formula the necessary hard code changes do not have to disrupt vanilla. It could be done by having the hard code be changed so the DustGlobalProduction is the global gross dust and then also having a new variable of DustGlobalUpkeep. In the vanilla XMLs all that needs to be done is subtract DustGlobalUpkeep from DustGlobalProduction a line before it’s used and then it will act as it did before. Then in the G2G mod the XMLs can be changed to implement the Expected Inflation formula.

Forgive me if I'm misunderstanding something, but the issue is caused by inflation being determined by by net dust production, which means that at high fleet upkeeps, it starts oscillating because you make terrible dust income at high inflation due to the upkeep being affected by by dust inflation, which then plummets inflation, so then you make tons of dust relative to that previous turn, and back and forth forever, yeah?

So why not just base the inflation on Dust income before upkeep? Wouldn't that solve the issue?

Your right on the cause of the oscillation but removing upkeep's effect on inflation (which requires hard code edits like all the other suggestions made & affects vanilla) could lead to scenarios where everyone goes bankrupt for having a fleet as upkeep can make them have negative dust a turn but with upkeep not being factored in the game will act as if everyone is producing a positive amount of dust and it will likely increase inflation. 

While the upkeep being affected by inflation change was meant to give power to pacifists over militarists this I expect would be a bit too much.

Would that not be dependent on how inflation scales with income, and be a gradual process that militarists could see happening over the turns as pacifists skyrocket their income, and try to attack before that can reach the point where it ruins them?

It just seems like thats mostly an issue of balancing the exact numbers. And, as you said, its meant to enable pacifist players to attack militarists in an economic fashion; would driving them into negative dust income for having a lot of fleets not be the eventual goal for them?