Question: What does the research rate curve look like in your games?

[Buster's Uncle]

Oh yeah - drive a CP into a base and hit "b" and +1 pop, notwithstanding the cap.  Works in X, too.  Fastest way to build up a superbase.

[Nexii]

I'll keep playing and let you know.  Transcendi I found didn't give a huge jump since you need Hab Domes.  But there's probably ways to play that exploit them better (more numerous smaller bases I assume, other than SSCs).

Right now the factions are at turn 94, about 40-50 pop average, 12 bases, and 15-20 formers.  Still early game infrastructure I'd say...heading into mid game tech soon (Tree Farms & advanced terraforming just coming out).  I feel like I underbuilt formers though since most factions haven't gotten boreholes yet.  Then again, this might have been ok strategy since few went for Green.  Pacifist game means those not in FM tending to fall behind.

Playing around with probe team odds might be another way to go. It then follows that stealing costs M rather than E.  I think tech difference would be a good factor.  Absolute tech would also be considered.  I feel like there should be more protection from probe techsteal (and other actions) based on PROBE SE, than necessarily having probes in cities.  Maybe making probe elite harder would help.  I've noted that infantry probes are a good unit to tweak, for defensive only probes they can be cheaper than rover and foil probes.

Personally I'm not a huge fan of the pod shoot down mechanic.  It's like war, only with no risk (build an ODS and damage the #1).  I suppose the #1 could then war you for it, this is a game of diplomacy above all else, even strategy.  I think this can work though.  Would prefer to not redo the tech tree structure, but swapping hydro and ODP availability would be good I think.  Then Nessus and Sky Hydro are gotten at same time with ODP available earlier.  Later game, there are a lot of broken things though (Space Elev) that don't come into play right now usually.

I think the growth thing could maybe work with nutrient rollover fixed.  One thing is that you have to look at how a base will grow in practice.  The curve would have to come down to +1 nutrient row per every several population rather than every 1 population.  Right now bases mostly grow not because of many surplus nutrients.  Usually a base will only have +2-4 nutrients at best, a lot of which is from the base square.  Let's use size 6 as an example (70N to grow).  GROW+6 implies 40% of this or 28N.  Let's say it's well improved, +3N from the base square and rest is 2N (farm+solar).  That's 9 turns to go to size 7 which is awfully slow considering it was 'boom' before.  Even if +10N (all 3N production), it would be 3 turns.  And 3N/square isn't obtainable till mid-game really (TF+HF on Forest).  Mass condensor isn't an option early due to ecodmg and poor E production.  If solar was boosted it might be more viable.  Later on you'll want boreholes which produce 0 N.  More or less, I'm trying to say that you want to avoid encouraging tons of small bases in PS with this which is what will be best (because the base square produces 3N, and resource bonuses mostly).  A small base in comparison will grow with 8N, or in 1-2 turns with Tanks+Condensor.  You'd see silly things like using size 2 bases to pod into big bases, just because the absolute nutrient cost is so much less.

It might just be easier to rework the automatic worker AI, so that it 'tries' harder to get you to +2N.  Unless you feel that booming is too powerful...I'm not really convinced that it is.  In some ways yes.  Probably would have worked better if GROWTH had modified nutrient production rather than requirements (same problem with INDUSTRY).  But that would be massive rework and probably not even possible. 

[Yitzi]

I'll keep playing and let you know.  Transcendi I found didn't give a huge jump since you need Hab Domes.  But there's probably ways to play that exploit them better (more numerous smaller bases I assume, other than SSCs).

I'd think you'd get hab domes before transcendi...in any case, then wouldn't hab domes be a big boost?

Right now the factions are at turn 94, about 40-50 pop average, 12 bases, and 15-20 formers.  Still early game infrastructure I'd say...heading into mid game tech soon (Tree Farms & advanced terraforming just coming out).  I feel like I underbuilt formers though since most factions haven't gotten boreholes yet.  Then again, this might have been ok strategy since few went for Green.  Pacifist game means those not in FM tending to fall behind.

I'd be interested in a game a bit further on...

Playing around with probe team odds might be another way to go. It then follows that stealing costs M rather than E.

I don't really like the idea of having to sacrifice probe teams at the rate needed to balance that way...

I think tech difference would be a good factor.  Absolute tech would also be considered.  I feel like there should be more protection from probe techsteal (and other actions) based on PROBE SE, than necessarily having probes in cities.

Definitely.

Maybe making probe elite harder would help.

Another good idea.

Personally I'm not a huge fan of the pod shoot down mechanic.  It's like war, only with no risk (build an ODS and damage the #1).

There is risk; if you try to shoot down a satellite, there's a 50% you fail and your ODS gets shot down.  What I'd do is just create an advantage by making ODS's cheaper than other satellites.

I think the growth thing could maybe work with nutrient rollover fixed.  One thing is that you have to look at how a base will grow in practice.  The curve would have to come down to +1 nutrient row per every several population rather than every 1 population.

Or just have really short rows.

Right now bases mostly grow not because of many surplus nutrients.  Usually a base will only have +2-4 nutrients at best, a lot of which is from the base square.  Let's use size 6 as an example (70N to grow).  GROW+6 implies 40% of this or 28N.  Let's say it's well improved, +3N from the base square and rest is 2N (farm+solar).

That's fairly early in the game; once the midgame seriously gets going, you should be able to use condensers to get rainy+farm for 3, and later on hybrid forests let you get 3N even with forests, or you can use enrichers for 4N.  (Though I do think allowing condensers to be in the same square as solars etc. would be a good move.)

That's 9 turns to go to size 7 which is awfully slow considering it was 'boom' before.  Even if +10N (all 3N production), it would be 3 turns.  And 3N/square isn't obtainable till mid-game really (TF+HF on Forest).  Mass condensor isn't an option early due to ecodmg and poor E production.

You wouldn't need condensers in that many squares unless it's a naturally arid area.  And yes, it would slow things down significantly; I'm not sure that's a bad thing.

More or less, I'm trying to say that you want to avoid encouraging tons of small bases in PS with this which is what will be best (because the base square produces 3N, and resource bonuses mostly).

  A small base in comparison will grow with 8N, or in 1-2 turns with Tanks+Condensor.  You'd see silly things like using size 2 bases to pod into big bases, just because the absolute nutrient cost is so much less.[/quote]

This really only applies earlier in the game...and at that point bases will tend to be small.

It might just be easier to rework the automatic worker AI, so that it 'tries' harder to get you to +2N.  Unless you feel that booming is too powerful...I'm not really convinced that it is.  In some ways yes.  Probably would have worked better if GROWTH had modified nutrient production rather than requirements (same problem with INDUSTRY).  But that would be massive rework and probably not even possible.

It would actually be doable, and not even that hard...the problem is more that it'd be a real pain to balance.  I think the answer is to make it slightly easier to use condensers to boost raininess, and accept that you're probably not growing to size 6 or 7 until the early midgame.  (Or if you prefer to play with size 7 bases by turn 30, you can do that...there are some very good reasons that all the non-bugfix changes I make are optional.)

[Nexii]

Condenser+Solar on the same square would change things a lot.  I think it's an okay idea but it might be hard to tell graphically if both are there.  In that case probably condensers should produce 1 N themselves (or possibly even 0, if it's a huge radius). 

Anyways I don't feel the main tech speed factor is pop boom - at least not in the early game.  I find it rare you would be able to get size 7 bases in 30T...not unless you're modding a lot of things to be very cheap/powerful.  Pre-psych modifiers, E production is just too poor for it to be a good strategy.  Further a lot of times, pop booming beyond your ability to improve the land does you no good.  For example in my game the PKs had the Monsoon Jungle, and boomed up to 9/16.  Yet they have less labs per turn than Uni, Morgan and even the Believers.  Granted this is with Forests at 12T to terraform. 

- reducing the production power of advanced terraforming
- significantly reducing the former time of all terraforming and especially advanced terraforming   
- no pop boom (maybe.  I think you'll have similar balance issues regarding what factions can hit 7/8/9.  it will be even more extreme as grow will get even more powerful the more you get.  imagine trying to balance INDUSTRY beyond +5, for example)

Perhaps a hybrid solution is better, GA as a requirement to pop boom on top of a +4 GROWTH SE.  The bigger issue is usually the non-GA boom (+4 GROWTH and Creche for +6).  I think pop booming is fine really it's just that often no PSYCH has to be dumped in lieu of nutrients in order to do it.  In Civ you needed a GA to boom - I think it was a mistake for them to just require GROWTH and no GA.  On Transcend at least, it takes a lot of Psych in the early game to get a GA - it's usually not possible before HoloT+TF (and sometimes even HF/Research Hospital).

Now at 98 turns of 400 I'm still only about 1/4 through the tech tree, so the tech rate must speed up a lot at some point - and not just in vs AI games that are 'won'.  Unfortunately came into a game ruining bug.  That or I'm doing something wrong (as far as teching, not the bug) - borehole/Green might be superior to planting all these expensive forests in FM.

[Yitzi]

Condenser+Solar on the same square would change things a lot.  I think it's an okay idea but it might be hard to tell graphically if both are there.  In that case probably condensers should produce 1 N themselves (or possibly even 0, if it's a huge radius). 

I think in that case they better produce 0 N.

- reducing the production power of advanced terraforming

I don't really see much need for this.

- significantly reducing the former time of all terraforming and especially advanced terraforming

Or this.
   

no pop boom (maybe.  I think you'll have similar balance issues regarding what factions can hit 7/8/9.  it will be even more extreme as grow will get even more powerful the more you get.  imagine trying to balance INDUSTRY beyond +5, for example)

Pre-Eudaimonic, nobody can hit 9, the same factions that can get non-GA pop booms unmodded can hit 8, and everyone except Morgan can hit 7.  So that would suggest that it might be best to put pop booms at 9 (so if you want to grow fast, you have to terraform for nutrients), and then put Eudaimonic at +3 GROWTH so that Sven doesn't have to run Planned (with its efficiency penalty) to get decent growth late-game.

Perhaps a hybrid solution is better, GA as a requirement to pop boom on top of a +4 GROWTH SE.  The bigger issue is usually the non-GA boom (+4 GROWTH and Creche for +6).  I think pop booming is fine really it's just that often no PSYCH has to be dumped in lieu of nutrients in order to do it.  In Civ you needed a GA to boom - I think it was a mistake for them to just require GROWTH and no GA.  On Transcend at least, it takes a lot of Psych in the early game to get a GA - it's usually not possible before HoloT+TF (and sometimes even HF/Research Hospital).

I considered that, but you still run into the problem that once you get a GA, there's no reason to get more than +2 nutrients.  The only reason I'm willing to consider allowing pop booms in the very late game is that otherwise it'd take forever to grow even with a nutrient focus, due to only a small fraction of your population being workers.

Now at 98 turns of 400 I'm still only about 1/4 through the tech tree, so the tech rate must speed up a lot at some point - and not just in vs AI games that are 'won'.  Unfortunately came into a game ruining bug.  That or I'm doing something wrong (as far as teching, not the bug) - borehole/Green might be superior to planting all these expensive forests in FM.

Could be...FM is really designed for a heavy energy focus, meaning solar rather than forest.

[Nexii]

I have a few reasons I am going to play around with reduced former times.  Only part of it is due to pop booming outpacing terraforming.

The main reason is that it requires so many formers per base.  When terraforming gets to ~40FT/square for a buffed raised solar/farm setup, that necessitates many, many formers to make it happen.  Even if a base is growing every 4 turns, that's 10 FT to keep pace (10 formers per base, 5 per base with Super).  With 10-20 bases being a standard strategy, that's 50-200 formers.  Or in the most conservative terms, to fully improve a base would be around 850 FT.  That would take 2-3 formers / base to fully improve before the end of the game - and many games end before 400 turns.  Realistically you're still looking at around 100 formers minimum which is really excessive.

I'd rather see the game get away from massive armies of formers and military as it really bogs down the speed of the game after 100 turns or so.  Early game turns take only a small fraction of late game turns, and that's a shame really because I think a lot of the late-game options are very interesting.  It just takes so long to get there - I haven't yet gotten to Future SEs is any of my test games for example.  The tech rate (turns/tech) speeds up a lot in the mid to lategame which kind of masks the problem.  The turns/tech rate should be more linear throughout the game I'd say, along with the real life time per turn.  The latter may be more difficult but ideally it would be linear increases rather than exponential.  One thing to keep in mind is that the incremental fixes (ecodmg, drones, fm exploits) have cumulatively slowed down the game a lot already.

So to get the amount of formers down, I'd suggest not over 10FT to fully improve any tile.  The ultimate raised farm/solar could be slightly over this amount.  At this rate one super former or two normal formers can more or less keep pace with growth. Faster growing bases might need a little more but I think that's okay.  To balance against this, the cost of a former would have to increase.  This is the trickier part.  I'm thinking around 40M and Clean not being an option.  If this makes the game too fast at advanced terraforming then there's two options: reduce the power of it (arguable, since there isn't much to reduce aside from condensor/borehole), or tweak the tech curve itself. Condensor to 0N with a bigger radius, and Borehole perhaps to 0/5/5 would be okay.  That or more ecodamage from advanced terraforming - make it painful before TF/HF.  Although these facilities are already very powerful, so I more favor the reduction in FOP.  They are already very critical early game - usually to the point of excluding other tech paths.

As far as booming tweaking ideally I suppose max growth, growth to boom, and GA required would all be variables.  Then one could play around with custom SE sets (such as modding Eud to +3 grow, or my own modified SE set). 

[Yitzi]

The main reason is that it requires so many formers per base.  When terraforming gets to ~40FT/square for a buffed raised solar/farm setup, that necessitates many, many formers to make it happen.  Even if a base is growing every 4 turns, that's 10 FT to keep pace (10 formers per base, 5 per base with Super).  With 10-20 bases being a standard strategy, that's 50-200 formers.

Well, keep in mind that some people will decide to go the forest route for fewer formers but lower resources...but yes, maximum terraforming is expensive.  Which, to tell the truth, it probably should be, in order to balance it against forests.  The only concern is the speed of the game...

I'd rather see the game get away from massive armies of formers and military as it really bogs down the speed of the game after 100 turns or so.  Early game turns take only a small fraction of late game turns, and that's a shame really because I think a lot of the late-game options are very interesting.  It just takes so long to get there - I haven't yet gotten to Future SEs is any of my test games for example.  The tech rate (turns/tech) speeds up a lot in the mid to lategame which kind of masks the problem.  The turns/tech rate should be more linear throughout the game I'd say, along with the real life time per turn.  The latter may be more difficult but ideally it would be linear increases rather than exponential.  One thing to keep in mind is that the incremental fixes (ecodmg, drones, fm exploits) have cumulatively slowed down the game a lot already.

Which is sort of the idea.

So to get the amount of formers down, I'd suggest not over 10FT to fully improve any tile.  The ultimate raised farm/solar could be slightly over this amount.  At this rate one super former or two normal formers can more or less keep pace with growth. Faster growing bases might need a little more but I think that's okay.  To balance against this, the cost of a former would have to increase.  This is the trickier part.  I'm thinking around 40M and Clean not being an option.

Keep in mind, though, that expensive formers are punishing in the early game.  I'll have to think about how to deal with it...

As far as booming tweaking ideally I suppose max growth, growth to boom, and GA required would all be variables.  Then one could play around with custom SE sets (such as modding Eud to +3 grow, or my own modified SE set).

I don't think it makes sense to have them be independent variables.  The maximum should go all the way up to booming, and rather than requiring a GA it would just be set high enough that the +2 GROWTH from a GA is vital.

[Nexii]

Yea, early game it could be somewhat punishing.  But this is balanced out some by SUP costs which are more relevant early, especially for formers since they don't change in cost all that much.  Each less former you have to make is +1 mineral/turn.  Which early on can be like +1 IND since you won't get more than 10M/base before higher terraforming.

For FOP as it is, forest+borehole isn't really weaker than raised farm/solar other than in pacifism scenario (many pacts).  Theoretically this could change (depending on growth boom and other ways condensors work).  But until then I would say forest should cost about equal in former turns.  You could go the route of trying to make them equal in power & former turns, or making farm/solar stronger at enricher since it comes later.  If not then farm/solar pre-enricher could be a bit faster than forest.  You could vary it between 5-10FT/tile but I feel much more than 10FT/tile is a major contributing factor to the game slowing down.  One thing to keep in mind also is that if you've planted forests already, to convert to Farm+Solar raising is more FT (incremental cost) on top of that already spent.  Of course, if only half the base is improved you aren't going to replace forests before doing the unworked tiles.

Overall I think the tech speed is a bit too slow early.  There's a few approaches.  One is to decrease early facility M costs.  Another is to reduce the L per tech.  It's about getting the right feel since lower L/tech can mean outpacing infrastructure.  And low M costs means that L ends up being the bottleneck.  There's some give of course due to the E/L slider.

Perhaps rather than a GA toggle to require booming, the amount of +GROWTH from GA could be modded.  That might be more intuitive.  Then I could give GA +5 GROWTH for example if I wanted early booms to hit +9, with your proposed GROWTH changes.

[Mart]

For the research, I think, appropriate would be to "flatten" research speed. In the beginning it may be varying 10 or more turns, depending on faction, map size, SE settings, etc. Later, when more techs, terraforming, facilities come, it may be a tech per turn, sometimes even two techs per turn.
So a good thing would be to make research even more expensive in late game.

[gwillybj]

I start my games with the research rate in alphax.txt set at 50%, then start the game with tech stagnation on. I try to adjust the Economy and Psych allowances up to push Research to 20-30 turns, but the last 10-15 techs still come too fast. At 100% Economy, dealing with drones because of the 0% Psych, and 0% Research, and they still come less than 15 or even 10 turns. I've even played around with an even slower tech rate (as low as 30%). Those last few techs just won't slow down.

[Yitzi]

Yea, early game it could be somewhat punishing.  But this is balanced out some by SUP costs which are more relevant early, especially for formers since they don't change in cost all that much.  Each less former you have to make is +1 mineral/turn.  Which early on can be like +1 IND since you won't get more than 10M/base before higher terraforming.

Even so, not being able to get any terraforming without spending 40 minerals on it seems too much.

Instead, perhaps it would make more sense to increase the cost of super former to 8 (i.e. a 200% increase) but make it available at environmental economics and have it triple terraforming speed, then at advanced ecological engineering that increases to quadruple.

Overall I think the tech speed is a bit too slow early.  There's a few approaches.  One is to decrease early facility M costs.  Another is to reduce the L per tech.  It's about getting the right feel since lower L/tech can mean outpacing infrastructure.  And low M costs means that L ends up being the bottleneck.  There's some give of course due to the E/L slider.

As it stands, M tends to be the bottleneck in the early game, so I think a small decrease to M wouldn't be amiss.

Perhaps rather than a GA toggle to require booming, the amount of +GROWTH from GA could be modded.  That might be more intuitive.  Then I could give GA +5 GROWTH for example if I wanted early booms to hit +9, with your proposed GROWTH changes.

If you want to make GA required for a boom, it'd probably be simpler just to nerf Planned to +1 GROWTH and keep it at 6; that way, you need Dem (or planned if you can't run Dem, or both if you have a GROWTH penalty)+creche+GA to boom.  Although that might make Planned too weak, though...

I start my games with the research rate in alphax.txt set at 50%, then start the game with tech stagnation on. I try to adjust the Economy and Psych allowances up to push Research to 20-30 turns, but the last 10-15 techs still come too fast. At 100% Economy, dealing with drones because of the 0% Psych, and 0% Research, and they still come less than 15 or even 10 turns. I've even played around with an even slower tech rate (as low as 30%). Those last few techs just won't slow down.

Thanks...what difficulty level is this at?

[Nexii]

Each faction could also start with 1 former.   Takes some of the RNG out of early game, for example monolith discovery is a huge boon vs not getting early monolith.  This might be a more balanced starting setup (1 pod, 1 former).  Gaians could get a Rover or 2 formers.  Also, formers at 40 is no more expensive than a recycling tanks which gives 1/1/1.  A former can do as much improving in just a few more turns.

The other issue with Super is that it can't be combined with Clean till later (unless you take out Clean for Formers, which can be done by flags).  I think the benefit would have to outweigh the cost, as the only savings would be maintenance in that case.  Also consider 3x former speed isn't always 3x the improvements, there's movement time wasted in practice for a former to get around.  And with bigger divisors (3/4) more FT get wasted on things like roads that don't take much time.  Cost 4 would probably be more appropriate (double).  A lot hinges on the chassis and cost mode of course...all customizable so not a big deal.  On the downside, boosting Super makes already very strong techs (EnvEco/EcoEng2) even more mandatory to beeline.  I might place it elsewhere but that would be a tough thing for me to comment on where else might be appropriate.

Thanks for the feedback Mart...I would have guessed a factor of 4:1 but 10-20:1 on the rate is not much of an exaggeration at all.  Since I was curious on this, and I couldn't find a graph or any previous discussion on how the tech cost curve looked, I decided to plot it out.  The attached graph is for a Normal sized map on Transcend (the curve is not as steep on lower difficulties it would seem).  Really simple graph but X-axis is # of techs accumulated and the Y-axis is tech cost (in labs points).  I have the data points but more or less the tech cost curve is linear, with some curvilinear increases in the early game.  This would explain why tech gets so fast late.  L rate as the game goes on is non-linear, whereas L costs are not. 

[Buster's Uncle]

I'm been keeping out of this because I play wall-to-wall exploits, and my data is therefore maybe worthless.  But I try to max out whatever science will not leave me losing EC per turn and make up the difference killing worms as I explore.  3-4 turns/tech for the first four or so then it gets a lot longer fast. All too soon, I have to up my psych to 40% and it takes another dive.  Gets up in the teens late early game before it starts getting shorter again.  Like the others, late game gets around 2techs/turn towards the end of the tree (I've been letting at least one opponent live for a long time so I can build/terraform like Sim City by that point) and goes up as high as four briefly with the future techs before the rising price eventually drives it back down to two for the rest of the time I'm playing out the clock.  The game doesn't go long enough for the tech rate to quite ever make it back down to 1/turn.

Gaians (or Mindworms w. Minds lately, who have a tech nerf but make up with growing crazy fast and working a lot of energy so it makes little difference) at Transcend, but again, I cheat like a toddler learning chess.

[Yitzi]

Each faction could also start with 1 former.   Takes some of the RNG out of early game, for example monolith discovery is a huge boon vs not getting early monolith.  This might be a more balanced starting setup (1 pod, 1 former).  Gaians could get a Rover or 2 formers.  Also, formers at 40 is no more expensive than a recycling tanks which gives 1/1/1.  A former can do as much improving in just a few more turns.

Maybe...I still don't love the idea, but I suppose each modder can make their own decisions.

The other issue with Super is that it can't be combined with Clean till later (unless you take out Clean for Formers, which can be done by flags).

That seems ok to me; that "later" will probably be before it becomes a serious issue.

I think the benefit would have to outweigh the cost, as the only savings would be maintenance in that case.

That's something...and at the point where super former would otherwise become available you get the same percentage benefit as unmodded.

Also consider 3x former speed isn't always 3x the improvements, there's movement time wasted in practice for a former to get around.

True, but if you're spending 30 former-turns on a square that will probably be fairly minimal.

And with bigger divisors (3/4) more FT get wasted on things like roads that don't take much time.

That's an even bigger problem with your idea; with my idea, you can get regular formers for that (and because they don't take much time you don't need as many of them).

Cost 4 would probably be more appropriate (double).  A lot hinges on the chassis and cost mode of course...all customizable so not a big deal.  On the downside, boosting Super makes already very strong techs (EnvEco/EcoEng2) even more mandatory to beeline.  I might place it elsewhere but that would be a tough thing for me to comment on where else might be appropriate.

I can't think of anywhere else that would be appropriate; keep in mind also that making terraforming cheaper makes it mandatory to use advanced terraforming of some sort (either raise/mirror/condenser or boreholes with your forest), whereas if advanced terraforming represents a significant investment you can have people just going plain forest in order to save on formers.

Thanks for the feedback Mart...I would have guessed a factor of 4:1 but 10-20:1 on the rate is not much of an exaggeration at all.  Since I was curious on this, and I couldn't find a graph or any previous discussion on how the tech cost curve looked, I decided to plot it out.  The attached graph is for a Normal sized map on Transcend (the curve is not as steep on lower difficulties it would seem).  Really simple graph but X-axis is # of techs accumulated and the Y-axis is tech cost (in labs points).  I have the data points but more or less the tech cost curve is linear, with some curvilinear increases in the early game.  This would explain why tech gets so fast late.  L rate as the game goes on is non-linear, whereas L costs are not.

Any idea how non-linear L rate is?

And, by the way, the full formula for cost has already been posted.

[Nexii]

Yea, either way with FTs it's something to consider - try to make most improvements divisible by 2x/3x/4x (or all, if multiple former power levels).  The main goal is to just get the ridiculous number of formers down, modders can do what they like for the forest vs farm/solar debate

I think that formula may slightly off what the game gives in practice - seems to overestimate.  But if anything it seems to verify that for the most part the tech cost curve is linear.  There's definitely more to it than just TECHS.

By what others are saying and my own experiences, I would say L production ends up being somewhere around (X^2.5)/16 per turn, where X is the # of techs you have.  That's on a normal map on Transcend.  The divisor varies some depending on infrastructure, peace/war, SE choices, and all that - but I'd estimate the growth exponent somewhere around 2.5.  Granted if you fit the L cost curve to that, the game might slow down some late....the early game tech would be faster (not a bad thing either IMO).  Too much of the game is spent in early tech level.