Discount Curve: Valuation Poll 1

[Stalker0]

I have long wanted to determine some kind of discount calculation for yields in the game. We all know that yields now is better than yields later. I would rather have 10 science now than 10 science in 20 turns. But how MUCH more is it...that's the real question.

So I'm going to try out a few informal polls to see if there are at least some solid general agreement in the community on a few ranges. That at least gets the party started.

I'll try out this poll to start, and then see how it goes.


Exercise
Assume a game on Standard Settings that will go 400 turns (game will end on Turn 400 using a Science Victory). You are given something that provides you X amount of science every turn starting from the specific turn for the rest of the game. This bonus is not multiplied by any % modifiers, it does not impact the strength of your Great Scientists, and it does not impact happiness, so its literally just a bonus slapped on at the end of each turn. Example: Option 2 in the poll is 10 SPT on Turn 100. so starting on turn 100, you would receive 10 extra science every turn.

In the poll, determine which bonus you think would give you the best benefit for the overall game.

 

[Arthur Jama]

Assuming an average size empire I would go for option 3. In one city challenge maybe option 2.

 

[Stalker0]

Assuming an average size empire I would go for option 3. In one city challenge maybe option 2.

lets assume a normal play, nothing fancy like a OCC

 

[Rosete]

There is no tomorrow in a civ game. Only victory! Picking the short term thing that gives you instant benefits will always be better than delaying things for future yields. Civ is also a zero sum game. I can win by beating down everyone else, a loss for one is a win for me, always. That extra science may allow me to field spearman sooner --> more barb camp clears --> more money --> more military --> beat down neighbors. On deity, snowballing is the win condition, you can't just bunker up and tough it out.

By math the earlier option is also best. 1 when you only have 5 /turn is a massive 25% buff. 100 when you're getting ~1000/turn on turn 200 is a pittance.

 

[N.Core]

By math the earlier option is also best. 1 when you only have 5 /turn is a massive 25% buff. 100 when you're getting ~1000/turn on turn 200 is a pittance.

For me, the latter is actually the better one. Since earlier in the game, instant yield bonus is far more impactful than per-turn yield. But this switched once the per-turn yield is outweighing the instant yield bonus.

If you have less per-turn yield earlier in the game, you can recoup the loss by chasing the instant yield rewards.

SPT also does not really matter, the only thing that matters is if the yield you gain actually reduces the amount of turn time required to finish the research.
Reducing 3-5 turns from mid-game research is far more impactful than reducing only 1 turn from early research.

Early snowball is easier to stop since there's only a small difference rather than mid-game snowball where the difference can be 5-8 techs away.

 

[Amask]

Early techs also don't matter as much 'cause you have so much stuff to build already anyway.
Also you aren't getting anywhere close to 1000/turn by turn 200.

 

[CppMaster]

Picking the short term thing that gives you instant benefits will always be better than delaying things for future yields.

Not always, though. That's the point here. That depends on numbers. For sure, I'd prefer to get 1000 science in 2 turns instead of 1 science in 1 turn even though it's delayed. That's a very extreme example, of, course, but it proves that sometimes delaying a reward might be better.

 

[BliTTzZ]

Option 3 is the way to go. There is still plenty of turns left to play (most of the time) and 100 SPT is a good chunk of science increase over your base science (I doubt it will be higher than 500 SPT on turn 200).