D
deleted
Guest
deleted
Last edited by a moderator:
SirJethro
Paterfamilias
Nope. There are some threads on the board that go into detail. It has been tested pretty thoroughly and you lose the excess. You need to manage the last turn wisely.
D
deleted
Guest
deleted
Last edited by a moderator:
Grey Knight
Old hacker
No, it does not. You can test this easily by taking a game where you can get a fairly low turns per tech, and saving the game the turn before you get the technology. Run a turn, and see what the rate is at various levels. Reload, and increase your research rate (thus over-researching the current tech), you'll see the tech rates will be the same the next turn.
So you can make a few $$ by micro-managing the research rate down while not increasing the time. Early in the game, when you are not researching very fast, you can start micromanaging 1-2 turns before learning the tech. Late in the game, when you have short times between techs, you have to start three (not just one) turns ahead of time in order to get the maximum cash.
Edit: I'm on painkillers right now, and I don't think I was very clear. 1-3 turns before learning a tech, try turning down the tech rate. Keep going until the "turns to research" goes up, then go back so you keep the same "turns to research". It doesn't always change. It would be really, really, nice if there were hotkeys to do it, if not from the main screen, at least from the Domestic Advisor screen!
Cheers,
Shawn
So you can make a few $$ by micro-managing the research rate down while not increasing the time. Early in the game, when you are not researching very fast, you can start micromanaging 1-2 turns before learning the tech. Late in the game, when you have short times between techs, you have to start three (not just one) turns ahead of time in order to get the maximum cash.
Edit: I'm on painkillers right now, and I don't think I was very clear. 1-3 turns before learning a tech, try turning down the tech rate. Keep going until the "turns to research" goes up, then go back so you keep the same "turns to research". It doesn't always change. It would be really, really, nice if there were hotkeys to do it, if not from the main screen, at least from the Domestic Advisor screen!
Cheers,
Shawn
SirJethro
Paterfamilias
Does anyone know whether the AI manages its finances well in this regard? It would be interesting to know if the AI does or doesn't recognize the waste impact.
Grey Knight
Old hacker
I think Soren has commented on this. I don't recall exactly, but I think he said that he didn't program the AI to do this.
Cheers,
Shawn
Cheers,
Shawn
TheNiceOne
Emperor
Grey Knight, you are mostly right, but I don't agree that there is any reason to reduce the research until the last turn. You will save the same amount of cash - on average. A simple example shows this:
Assume a tech cost 1300 research points and you have an income of 400 gold and want to research the tech in 4 turns. By micromanaging as early as possible, you'll set the reasearch rate to: 100% - 80% - 80% - 70%. By checking the last turn only, you'll have a research rate of: 100% - 100% - 100% - 30%.
The total tax will in both cases be exactly the same: 280 gold.
If you choose examples with less round numbers, you'll find that the net result may differ with a gold or 5, but each will be best 50% of the time.
The big exception is if you only need to set the research to 20% the first turn. In this case it may be that 20 - 10 - 10 - 10 is enough, and then you'll waste reasearch by keeping it at 20 - 20 - 20 - 10.
So the conclusion is that unless you have a research rate of 20% the first turn, then there's no reason to check reaserch rate again until the last turn.
Assume a tech cost 1300 research points and you have an income of 400 gold and want to research the tech in 4 turns. By micromanaging as early as possible, you'll set the reasearch rate to: 100% - 80% - 80% - 70%. By checking the last turn only, you'll have a research rate of: 100% - 100% - 100% - 30%.
The total tax will in both cases be exactly the same: 280 gold.
If you choose examples with less round numbers, you'll find that the net result may differ with a gold or 5, but each will be best 50% of the time.
The big exception is if you only need to set the research to 20% the first turn. In this case it may be that 20 - 10 - 10 - 10 is enough, and then you'll waste reasearch by keeping it at 20 - 20 - 20 - 10.
So the conclusion is that unless you have a research rate of 20% the first turn, then there's no reason to check reaserch rate again until the last turn.
TheNiceOne
Emperor
Hm, I must reply to myself to tell I was a bit wrong the last time. My above post is mostly correct, but the exception should be a bit more general:
It may be smart to check the research 2 turns before finish, if the original research rate (in 10%) is less than the total number of turns you research the tech.
I.e., if your original research rate is 60%, and it will take you 6 turns or less, then you don't need to bother checking the reserach rate again until the last turn. But if the research time is more than 6 turns, it will be smart to check the 2 turns before the tech is dicoverd as well.
It may be smart to check the research 2 turns before finish, if the original research rate (in 10%) is less than the total number of turns you research the tech.
I.e., if your original research rate is 60%, and it will take you 6 turns or less, then you don't need to bother checking the reserach rate again until the last turn. But if the research time is more than 6 turns, it will be smart to check the 2 turns before the tech is dicoverd as well.
Grey Knight
Old hacker
I'm not sure about the math, but I do know that if I care enough about my gpt to fiddle with the research rate, I'd rather have the gold THIS turn rather than the NEXT turn.
Good point, though. I suspect you are correct on the math.
Cheers,
Shawn
(Who will now worry about it less before the last turn
)
Good point, though. I suspect you are correct on the math.
Cheers,
Shawn
(Who will now worry about it less before the last turn
)
Evincar
Iron works is a legend!!
On shimshan's observation, this usually happens if another civ just discovered the tech: The price drops so you may actually have already paid more than the new cost, thus you will get it with just a single scientist.
