Re: Money negative while surplus
Posted: Jul 27 2018
This reply I posted before your testing post with the values.
So please understand this is NOT a reply to your latest post with the testing values in excel.
Baltagor, I'm sorry but both you and Dylan are wrong!
You don't pick a difference based on large numbers and then do the math like:
change versus total = 0.04% and thus neglectible.
Why that math is wrong is simple to understand, but basically you are using absolute treasure values to estimate the rounding error on increase/expenses and that's just simply wrong.
Any economics in the world can tell you what you are calculating is a value with no use at all, it says nothing.
If my treasure were let's say: 1,000,000,000,000,000,000,000,000,000 those 202,000,000 would be 0,0000000000000000202%
And if my treasure were let's say 1 those 202,000,000 would be 202,000,000%
So in 1 case you claim the rounding value is well, 0,00% and in the other case its 202 Million Percent!!! And this points out pretty much why it's wrong.
The correct way to estimate the rounding error is exactly the way I did it:
All expenses adding up
All income adding up
the difference is what should be added or subtracted from the treasure.
the rounding error of your math is then the actual difference between your calculations what is added or substracted in the game engine and what should be added or substracted with 0 rounding errors
and this difference divided by the value that should be added and that's a huge percentage.
You COULD, also make it relative to the total income for that turn by which the rounding error would come to 4%. and this 4% is there regardless of if I have 1 Trillion or just 1 Million.
Either way is too big.
But you NEVER EVER can figure the rounding error of 2 sums and their difference based on if I would take the integral of these differences based over a time frame of x with x larger then 1.
That's definitely wrong logic.
And let us not get into your claim that 4% and 0.04% would be a factor 10.. because I doubt you read what you just wrote by writing that. If you read your own post, I'm sure a smart man like
you will notice 4 and 0.04 is a factor 10^2 and not 10^1
So please understand this is NOT a reply to your latest post with the testing values in excel.
Baltagor, I'm sorry but both you and Dylan are wrong!
You don't pick a difference based on large numbers and then do the math like:
change versus total = 0.04% and thus neglectible.
Why that math is wrong is simple to understand, but basically you are using absolute treasure values to estimate the rounding error on increase/expenses and that's just simply wrong.
Any economics in the world can tell you what you are calculating is a value with no use at all, it says nothing.
If my treasure were let's say: 1,000,000,000,000,000,000,000,000,000 those 202,000,000 would be 0,0000000000000000202%
And if my treasure were let's say 1 those 202,000,000 would be 202,000,000%
So in 1 case you claim the rounding value is well, 0,00% and in the other case its 202 Million Percent!!! And this points out pretty much why it's wrong.
The correct way to estimate the rounding error is exactly the way I did it:
All expenses adding up
All income adding up
the difference is what should be added or subtracted from the treasure.
the rounding error of your math is then the actual difference between your calculations what is added or substracted in the game engine and what should be added or substracted with 0 rounding errors
and this difference divided by the value that should be added and that's a huge percentage.
You COULD, also make it relative to the total income for that turn by which the rounding error would come to 4%. and this 4% is there regardless of if I have 1 Trillion or just 1 Million.
Either way is too big.
But you NEVER EVER can figure the rounding error of 2 sums and their difference based on if I would take the integral of these differences based over a time frame of x with x larger then 1.
That's definitely wrong logic.
And let us not get into your claim that 4% and 0.04% would be a factor 10.. because I doubt you read what you just wrote by writing that. If you read your own post, I'm sure a smart man like
you will notice 4 and 0.04 is a factor 10^2 and not 10^1