Well, that's why they're fallback tiles. You don't actually WANT to work them, it's just that for one reason or another you can't work really good tiles right now. And while it's not something you optimize for, it is an advantage to have better fallback tiles, however minor.
Yes, if the water refinery actually was free and pre-built in every city (or if water tiles just had these yields by default), then that's of course the case - but these fallback tiles come at a price, a rather high one I might add.
I think I agree with what you mean but not with what you say. Because what you say can be refuted by giving one single example of a land-focused map where it's actually sensible to research the tech.
Now I had this one game a couple of days ago where, on a protean map, my capital got placed on a three files wide strip of land surrounded by ocean. So great! On top end there was land but also the AIs, two cities where all I could get in and it was close enough to the AI to expect a DoWed for forward settling. On the bottom end there was a peninsula. Long story short it was very sensible to settle 7 of 8 cities on the coast although I generally optimize for land tiles. This game I actually got the water refinery tech.
Yes, right. I should have probably worded that differently - what I meant was something like: "If there aren't already a bunch of water resources that you'd want to work anyway..." instead of just excluding that specific map type.
I'm not 100% sure it was worth it but actually optimistic. Which is why I wouldn't call the tech horrible even on land maps. Certainly it's not good when the majority of your cities isn't costal. But unless your playing a map entirely without water there can still be games and situations where the tech is actually worth considering. Probably even good on water maps.
It's horrible as a tech that you "always take" is what I meant. It's actually one of the few techs that can be taken situational, although I'm really skeptic a scenario where a single city would get one - haven't done the math though, so my opinion on that case may just be completely wrong.
I have been playing mostly Quick/Soyez and winning in the 225-250 range. I am just working my way though the Achievements. I did play two Apollo games, quit the first one and won the second, but don't remember what turn. I have not really been in a 'hurry' to win, just practicing my wide play. Once I get most of the Achievements I plan to play only Apollo and at that point I may be forced into Academy spamming, but so far I just don't 'feel' it. I don't have an optimized path through the tech web, just picking the tech that solves my most important need at that moment. Rapid expansion with health no worse than -10 or so is my goal. Maxing population growth is my defacto mode of maxing tech pace. Getting nearby land before the AI does is more important than the tech web victory race.
See, I think that's the difference. I've spend most of the time in the game with trying to maximize my strategy and throwing out as much as possible to get to the end quickly - my fastest Transcendence Victory (probably the most reliable and stable victory condition) was turn 206 on Standard Speed - admittedly an insane start and very far away from any of the AIs (I literally almost didn't build any military throughout the whole game except for some scouts that roamed the lands around me to not be surprised), but in general I manage to win on turn 210-220 (sometimes ~230, if the game starts bad - or I just lose because my neighbor attacks me before I'm ready
), and there's still a lot of room for further optimization.
So what I'm basically saying is: The thoughts you put into the posts above probably make a lot of sense for the way you play - but the more you go towards just trying to win fast, the more of this stuff just doesn't matter anymore, because a fast victory is mostly won by just throwing out all the additional stuff (like affinity resources - they're basically only some free gpt from trades - affinity buildings and units will never pay for themselves) and instead pushing right through the middle. (Which is, I agree, a bit sad - but that's the way things are right now)