I'm quite sure that most of you read only first post in a thread at 2K forum. (if read any)
Usually its ok, but not in this case. Thing that we discuss is complicated and I couldnt write answers to all possible questions in my first post. But I DID answered 95% of them in next posts. And because my replies were huge, I didnt think It was good choice to update first post. There were also new ideas popping as we discussed the thread, so its worth reading the whole thread. You are most likely to find some or complete explanation to all questions and possible downsides of what I proposed. THAN, if you have questions left or my reply didnt explain something completely, you may ask something.
I dont want to repeat what I already explained and didnt recieve solid counterargumets, but I will have another try in key point: so called "meritocracy based system".
The biggest disagreement that I and MadDjinn have is: whether respective GPs should give more bonuses to civs with greater respective yields, therefore improving this field even more?
It does have logic, because as it was said in this thread: "Producing GAs as a Warmonger seems hilarious to say the least" (though this is not the best example because we do not choose between GAs and GGs, they are generated separately). BUT why dont you think that "Producing GSs as a Warmonger seems hilarious to say the least"? But THAT is what happens in 90% of cases. In SP even "GAs when warmonger" can be quite effective if you are going puppet cultural victory. Gameplay-wise that is NOT hilarious at all.
Also, I'll quote part of that thread not to repeat:
MadDjinn said:
I think we'll just have to disagree on intent. The model I'm interested in is a meritocratic system. Do well in an area and you get better output when using a specialist in that area. It's a core model for many good systems. What you seem to be proposing is more about letting people get bigger bonuses when they aren't doing well in an area. Ie, 1500 culture is 'less' when you've got 300 cpt (3 turns gain when culture focused) than when you've got 100 cpt. (15 turns gained when not culture focused) These models don't work together. As is, I disagree with your intended model since it seems to be meant for covering over 'bad' play rather than encouraging 'good' play.
Rinnero said:
The point is not to make it easy for everyone, but to allow civs with moderate science not to fall too much behind. For example, one civ gets 250spt, the other one 350spt. If +50% turn during 10 turns ability is used, that gives 1250s and 1750s respectively, making strong civ even stronger. Instead I believe its better to make GSs (and therefore RAs) a bit alternative way to get science, that only partly depends on spt. (partly because current era depends on spt) So that you could at least not to fall too far behind, but still be weaker than scientifically strong civ. In the example I made, the second ability that fits that idea should give instant 1500-1750s. In that case, weaker civ will still fall behind in time, but not that much. Also, though you get more science total with first ability when you have 350spt, its still sometimes better to use instant ability not to wait 10 turns.
Exponential growth of numbers when one multiplies the effect of other can create situation with only one best choice (so that when you have both high science output and a lot of GSs you gain huge science: having 20% more spt and 20% more GSs will result in (1.2*1.2=1.44) 44% more science).
There is a reason why all multipliers in Civ V are applied to BASE value. Otherwise all these +50% +50% +50% bonuses can lead to extremely large numbers. And having another +50% when there are already some of them is much better than first +50% when there are none of them, forcing you to go deeper into one (scientific, considering that science is usually the best bet) way.
Right now GS is not only the usual/best choice, its also the most gaining choice. You are in industrial - what is better: 450g+30inf, 12-30culture/turn, 500 production (wonder rush), or 3000science? In standard development (in MP) you are 80% to choose 3000science. But what if we are to choose from 1500c, 1500s, 750g+30inf or 500p? That is less obvious choice. That system clearly counters science is always safe bet theory. If I have 300spt and 150cpt, Id rather choose 1500c, but I may also consider to grab 1500science to get key tech (mech inf, for example) 5 turns earlier, or even 750g if I really need to buy a unit in emergency.
I'll add one more thing: Actually I'm not against such deep specializing "meritocracy" model, I'm against it being implemented in CivV in current state.
It would be so nice to be able to win properly going cultural, or into money -> really different gameplay. BUT this
dont work in CivV! (if you are to change only GPs). For example, if you were to choose to double one of four yields: culture, production, gold or science. What would you choose?... That is the answer...
If you choose to have double culture you just wont survive, the same goes for money. Production is possible choice for rush (especially if Liberty production rush), but if you wont achieve anything at the beginning, the one with double science will beat your armies with stronger units.
CivV will need great changes in other fields to make cultural civ competetive with scientific. (for example twice stronger policies will make having 5 more policies almost as good as 5-8 more techs, maybe

)
Flat bonus will allow to boost areas you are bad at to medium level (in MP that is usually culture), or it still can help you to go deeper into something if you really want to do this, ignoring anything else. If you dont generate something much, that doesnt mean you play bad. That means you chose that as your strategy.
Krikkitone said:
The Settling needs to be reworked as well. (so that it is worthwhile v. the others)
I'd say
Manufactury: 4, +6 with Chemistry, +10 with Combustion
Custom House: 4, +6 with Economics, +10 with Flight
Academy: 6, +9 with Scientific Method, +15 with Plastics
Landmark: 6, +9 with Archaeology, +15 with Mass Media
Thats what I proposed at 2k thread. Your late game techs are probably more relevant to real life. And higher numbers certainly wont hurt late game choice in settling/bulbing, but I didnt calculated this in depth, so in this case I wont say that exactly this or that number is better
Rinnero said:
Late-game increase of yield of all settled GPs will make settling in late game a decent option even if there is only 50 turns before end of game. +3c, +2g at solar power plant, +3s, +2p at nuclear power plant. And the other increases in mid-game should be +2-3 (not +1-2, and including landmark)
That would make bonuses from settling as:
Early / mid / end (in brackets with freedom finisher)
- 6 - 9 (15) - 12 (18) science/culture
- 4 - 6 (10) - 8 (12) production/gold
Its better to separate instant and over time bonuses, so that the goal will be almost similar but with different flavor. And youll need to think through difference and decide whats better (make these options so that its not obvious what to choose during as long part of game as possible)
Krikkitone said:
Of course one thing that would need to be added is to increase the cost of GP with increasing numbers of cities (ie +100 per GP, +25% per city)
I noticed that and tried to attract to this everyone's attention, but everyone ignored it.
Rinnero said:
By the way, when I choose whether to annex city or not, it says that puppeting it will not increase cost of social policies AND great people. But it doesnt seem to increase anything anyway. Is this a bug? If so, a major one. Because even 15% increase in GP cost will benefit tall empires (tall empires have been nerfed because of -50% reduction in increase of policy cost for each city, and tall empires' faster GP will help them in scientific race)