Not so fast. I have a ton of other fun graphs to show you first. Oh … not bothered with my silly graphs? Ok, here are the results ...
Hmm - highslide ad got in the way of the most critical piece. It read 86907g with expenses at -49,583 per turn. Two more turns (so, T398) and I am toast. Here is what happened to my bank balance. That will teach me to give someone else my credit card ...
My detailed projection was estimating T394 as the critical turn. I was 4 turns off - it seems my expense projection must have been high. Let’s see where I went 'wrong' ...
Not too flash ... 10k different 100 turns down the track. I wonder where the difference is.
Hmm - not that good. My initial forecast was based on 90 units per turn (mainly Longbows). As it turned out, I checked some turns and building Longbows was generating(!) coins via hammer overflow. I had mis-calculated just how many hammers Mining Inc would generate per city. As such, I turned most cities onto building MachineGunners (120h). This meant that I only produced about 75 units per turn - thus my projection is off.
But what about that flattening of the green line at the end … what happened there? I think I hit some sort of internal limit on the unit array as the game crashed if I kept on building units. I had to swap all production to beakers.
You can see that I currently have 8,176 units. That doesn't sound like it is close to any sort of nice round number limit (ie 10,000) - maybe the limit includes units that are currently dead? While we are looking at unit counts ... I did manage to generate a few GPs over the last 100 turns ...
Wow - that is a surprise. Maintenance expenses actually decreased. And I am not sure exactly what is going on here. I don't think that I had that many cities 'Loving the Ruff'.
That is a pretty good match for the first 20-30 turns. I reviewed and adjusted my projection formulas and I had to add a second decay factor from T310 onwards to get the blue line down to the green line.
No surprise there - it was a straight polynomial and it stayed that way.
As I said above, I updated my projections based on the additional 'actual' data (I ignored the unit limit issue which explains the last few turns where the green line drifts from the projected blue line) and you can see the result above. I am projecting expenses over 90k by T500. Based on that projection, I would have burnt through 12,000,000 gold coins by T500.