Research points & research speed?

[The Boz]

This has been bothering me for a while now. Why are there research *points* AND research *speed*? What unique feature does speed give that couldn't be fulfilled in a much more straight-forward way as +% research point bonus, which is much more obvious to see (because speed is not shown when looking at the tech generation resource)?

 

[Athos173]

Let's suppose empire 1 has 100 society and a research speed of 110%. Meaning that he has an effective research rate of 110.
If he receaved an additonal 10%society generation his effective research rate would be121.
If instead he receaved a 10% speed increase his effective research rate would be 120.
Meaning that they intervene at different stages give different values and they also add realism. Moreover they are affected by different things. While society research generation rate is affected mainly by investment research speed is affected by scientist skill level.

 

[kabill]

Not 100% sure but I think speed is a multiplier on everything *after all other modifiers*.

E.g. if you have +25% research points from science jobs and +25% research speed, that means your scientists are producing 4 * 1.25 * 1.25 = 6.25. But if you have two +25% research point bonuses, that means your scientists are only producing 4 * 1.5 = 6 science per job.

Further, I think research speed affects all sourced, whereas research points modifiers only affect specific sources (i.e. jobs, research stations, etc.).

Whether it's worth the additional complexity is obviously open to question. But assuming I've not misunderstood how it works, it is meaningfully different.

EDIT: Ninja'd

 

[The Boz]

For the scientist bonus: wouldn't -X% cheaper techs do the same job, but in a more clear manner?

 

[Athos173]

Same arguments apply.
Tech 1 (physics) costs 5000.
Empire produces 100at 110% speed.
If speed increases by 10% the tech is finished in 42 months.
If cost decreases by10% the tech is finished in 40 months.
They intervene at different stages and makes it easier for the player. For example if he has 200 resource he will invest it in a better scientist rather than more research stations.

 

[kabill]

For the scientist bonus: wouldn't -X% cheaper techs do the same job, but in a more clear manner?

Kind of, but the stacking maths are different. I.e. with +speed, each bonus of the same magnitude reduces in effect, while with -time each bonus of the same magnitude increases in effect.

E.g. If you have +50% research speed already, adding another +50% research speed results in +100%, which is a 1/3 increase on what you already had rather than 1/2. But if you have say -25% research time, adding another -25% research time results in -50%, which is 2/3 the time you already had rather than 1/4.

There's nothing inherently wrong with that but, in general, I think cost reductions are much harder to balance where there's a lot of potential sources stacking together, as you either need to make the modifiers small (and therefore of negligible importance individually) or risk breaking the game (e.g. -75% research costs would be absurd, it's like a +300% modifier on your research)

 

[Zarine]

For the scientist bonus: wouldn't -X% cheaper techs do the same job, but in a more clear manner?

Cheaper cost is bad !
Basically, reducing the cost means that each additional reduction give you additional boost.

Let say you take 100 days at 10 points per day to make a research = 1000 points of research.

I get a first -10% bonus for cheaper tech, it is now 900 points and take 90 days.
The bonus is indeed -10% time : (900 - 1000) / 1000

Now, let get a new -10% bonus. It will now be 800 points (because that a total of -20% bonus)
However, the real benefit from this is now -11.11% time : (800 - 900) / 900
In other word stacking those bonus grant each time more benefits and you should always look for more.


Whereas if you increase the point generation, each additional bonus get reduced benefit:

+10% points => 11 points per day => 90.909 days to research ( about -9.09% time)

then we add +10% more => 12 points per day => 83.333 days to research
And this translate to only a bonus of -8.33% time.

Thus at some point, stacking more won't bring enough benefits and you should look for something else.

 

[The Boz]

But leaving the scientist at, like, -3% per level, while converting all other +speed bonuses to +point bonuses would work easily, and it'd be *much* easier to convey the precise technology advancement rate.

 

[Flame13223]

Don't fix what's not broken. The current system makes perfect sense to me.

 

[Gilbert95]

The other issue with reducing costs is that if the source of that bonus goes away, for example if it is attached to the scientist and they die or get reassigned, then you get no benefit on the current tech being researched. Its price suddenly jumps back up. Whereas with increased speed, that extra accumulated progress does not go away if the source leaves or is otherwise turned off.