Mechanic Unlock Flow is Off

[Eled the Worm Tamer]

well from a practical standard the ideal would be to try to have fewer things that were worth skipping in the first place I think.

 

[Ma®tin]

It would be nice to have it more configurable. Or make it adapt to what is going on in the game. A subsurface heater might not be needed too much if you play somewhere without cold waves. Getting bunch of rocket techs, when you have 0 rockets is not too useful either.

Perhaps a rework of techs in the far future to make it a tree instead of a line could be interesting.

 

[MarQan]

That is the opposite goal. The point I am making is that by skipping so many techs because the key anomalies open up so many I skipped several huge phases of colony development and went straight to the mohole. I went from "I need to keep expanding to the various deposits to stay solvent" to "mohole trivializes everything, I win." All this with half the techs, including big time ones like deep resources unresearched.

I'd go the opposite of what you are suggesting here. Every tech you research reduces the cost of further techs but those techs have a higher baseline before the reduction for doing the lower level.

As I said in probably every comment I made about research: it's too easy. You're implying my change would make it easier, while I clearly specified that it should not.
If everything on average would cost double, then you'd get the Mohole much later. While early techs are relatively cheaper. That's what a more exponential curve would do. It would also make early techs comparatively better, while still allowing you to skip useless ones.
Limitations could be added, like the wonders only being available after a certain % of techs have been researched.

The bottom line of your suggestion is in fact very close to mine.
The main difference comes when you encounter useless or very useful techs. (I mean useless and useful in a specific situation)
You're incentivising the research of useless techs, and I allow people to beeline for the really good ones.
In a balanced tech-tree neither are good.

Of course having an adaptive reserach tree like Martin suggested would be awesome.

 

[Ma®tin]

Also having a tree instead of just a straight line of research would make more sense in certain cases... why should it be possible to research improvements to shuttles, fusion or deep mining, if you have none of those researched?

Similarly getting some option to steer your breakthroughs at least a little bit could be nice. For example choosing between AI and biological paths etc.