Research

[Ganja]

In my game its 1645(ish) Prince level and there have been no major wars. The problem I have is 6 out of 10 of the AI are now in the Industrial era. Is this a freakish occurance or is the time line still off?

 

[Thalassicus]

I don't really know, it depends on a lot of factors. Map size, game speed, mods in use, civilizations present, even the terrain around those civs -- lots of flat terrain tends to result in superpower civs conquering everyone around them and teching faster.

 

[spfun]

Just like to make a comment about research in general in civ5 after seeing this mod, i find i get like 12+ great scientists per game easily, i get so many free techs in this game its crazy. whats the point of making future techs 150% for example if im just going to pop em for free... mods which make tech costs 150% across the board but do nothing to nerf GS & research pacts are terribly balanced imo this mod only been quite small doesn't really fit into that description per say but research pacts are still pretty damn good even with the time addition. ( i never use em tho funnily enough, maybe because its imbalanced )

honestly i think free techs are horribly imbalanced in this game & theres to many things you can do to get them & they're easy to get & you can PICK what tech you want... ive beelined into the industrial age saving free techs for key techs, using ICS trading spam strat for order policies in a redicilously short time. The AI have no counter... thats with no mods.

i'd like to see a mod that nerfs Research pacts more, & nerfs GS tech popping greatly.

 

[Seek]

Thal - what do you think about reintroducing first-to-tech or -era rewards? Imo, they're really fun and would help provide incentive for more varied beelines!
__________

I'm also putting together a small mod for personal use that takes the GS mechanic that Alpaca put into the Play With Me mod. It limits beakers by number of techs, a la civ4. I'm really enjoying it! The goal is to nerf beelining and make GSs only give partial techs at mid-Renaissance. We had some discussion about the GS mechanic a while ago, so I thought I'd drop a line and see if it's something you'd consider adding here. (I've also put in more expensive RAs by era; there have been some discussions recently in the Strategy forum showing how exploitable they are and the simplest solution is this, I think.)

I'm considering releasing it as a mod component (Alpaca has given me his blessing for this), so an alternative could just be to include it in Combined (= less work for you!) Once I finish testing it and get it balanced I'll PM you a copy.

*If anyone else is interested in seeing this as a standalone mod, please let me know as that will help me decide whether to release it!*

PS - Maybe this thread should be renamed to Balance - Research?

 

[Thalassicus]

Seems reasonable enough, would be great to include.

 

[Ahriman]

I'd *really* like to see a GS nerf along these lines of what alpaca has.
They goofed when they removed all the GS specialist slots but left the GS-bulb unchanged.

The right solution IMO is to make it 1 scientist slot for each of library, university, public school, research lab (or something like that) so that GSs are available, but nerf the lightbulb ability.

 

[Tomice]

What was wrong with the old lightbulbing nerf? Not being able to choose the most expensive tech available seems like a reasonable approach to me.

Alpaca's way gives you more freedom I admit, but it hardly helps against early beelines.

 

[Seek]

What was wrong with the old lightbulbing nerf? Not being able to choose the most expensive tech available seems like a reasonable approach to me.

I found the Harder Free Tech mod's method to be quite clumsy, affecting both GSs and actual free techs indiscriminately. It's rather frustrating to win the GL and only be able to get The Wheel or some other weak tech. In other words, I didn't find that it added to my enjoyment of the game - after a while I stopped loading it for games because of this.

I'm assuming Thal hasn't committed to including it into the mod until he's had a chance to playtest it, so if he decides not to you can always go back to Harder Free Tech if you don't like it.

Alpaca's way gives you more freedom I admit, but it hardly helps against early beelines.

I think early beelines have already been seriously nerfed with the patch. Playing with TBM, you do get two slots with the NC, but it requires a lot of tradeoff:

• You can't expand, therefore missing potentially good city spots
• You can't build many military units
• Once you've built the NC you have to invest two of your precious citizens to working the slots for almost twenty turns, hindering both production and growth

If you're going for CS, by the time you get your first GS in most cases you could have hard-teched it, and if you're going for a deeper beeline like Gunpowder, you won't get enough beakers to fully tech it anyway. Add to all this the 10 turn cooldown timer for GSs and costlier RAs, and you're looking at much less practical beelining.

I should have a decently balanced version out for testing in the next few days.

 

[Spatzimaus]

The right solution IMO is to make it 1 scientist slot for each of library, university, public school, research lab (or something like that)

I'd actually already done that in my own mod, and it helped tremendously, at least in reducing ICS.

Strangely, for all the complaints about GS bulbing, the research problem I always found more critical was simply the massive tech acceleration in the late game. That is, once your cities have stacked Library+University+Observatory+PublicSchool+ResearchLab, you're looking at 5.25 research points per population, so a decent-sized empire would start getting everything in 1-2 turns. By the time your first Tank rolled off the assembly line you'd be ready to upgrade them to Modern Armor.
So in my own mod I drastically reduced the numbers; Libraries were +1 per 3 (and 1 culture), University 30% (and 1 culture), Observatory 20% (and 3 culture), Public School 40%, Research Lab 50%, Medical Lab 10% (long story). Instead of 5.25, you're looking at a cap of 3.33 per population. And I reduced the specialist slots like you mentioned, plus a few tweaks to the era multipliers.

Why do I bring this up here? Because it made the Academy more valuable; if you feel you need that extra research production to make up for the slower pace, you're less likely to spend that GS grabbing a single tech when you could instead use him to add ~10 research per turn (4 plus all the multipliers, or is it a base of 5? I forget.) until the end of time. Sure, there'll be times when you really need a certain tech ASAP, but that should be the LESS efficient way to use the GP. At 10 per turn, how many turns would it be before that Academy gave more than a tech's worth of beakers? (Okay, there's an opportunity cost, since the Academy would take the place of a Trading Post or something. But you're still coming out ahead over the course of a full game.)

Don't think about this purely in the context of my own mod. Just think: what if you gave all Academies +2 research at some point in, say, the early Renaissance? Wouldn't that drastically reduce the appeal of bulbing every GS you get, and increase the desire to place Academies near your core cities instead?

But this only really matters if you're using a mod designed around ensuring the game is still competitive for that long; who really cares about tech boosts in the Modern Era if the game was effectively over the moment you got Tanks? The core game just isn't very good at that, so in the absence of a good balance mod, this sort of thing won't have the impact I'm describing.

 

[Ahriman]

What was wrong with the old lightbulbing nerf?

Its ok, but its not as simple.
Giving a fixed set of beakers is much simpler than changing which techs you can or can't access with a bulb, which also leads to MM to make sure that you can get exactly the techs you want when you use the scientist.

Also, Seek's point about it messing with Great Library.

Alpaca's way gives you more freedom I admit, but it hardly helps against early beelines.

Well, its hard to get more than ~1 great scientist in the early game, and 1 beeline isn't that powerful.

the research problem I always found more critical was simply the massive tech acceleration in the late game

Thats a problem too, but a different one.

when you could instead use him to add ~10 research per turn (4 plus all the multipliers, or is it a base of 5? I forget.)

By the time your academy could be producing 10 beakers per turn, techs are expensive enough that the benefits from bulbing would be many dozens of times that.

 

[Seek]

Why do I bring this up here? Because it made the Academy more valuable; if you feel you need that extra research production to make up for the slower pace, you're less likely to spend that GS grabbing a single tech when you could instead use him to add ~10 research per turn (4 plus all the multipliers, or is it a base of 5? I forget.) until the end of time. Sure, there'll be times when you really need a certain tech ASAP, but that should be the LESS efficient way to use the GP. At 10 per turn, how many turns would it be before that Academy gave more than a tech's worth of beakers? (Okay, there's an opportunity cost, since the Academy would take the place of a Trading Post or something. But you're still coming out ahead over the course of a full game.)

Don't think about this purely in the context of my own mod. Just think: what if you gave all Academies +2 research at some point in, say, the early Renaissance? Wouldn't that drastically reduce the appeal of bulbing every GS you get, and increase the desire to place Academies near your core cities instead?

But this only really matters if you're using a mod designed around ensuring the game is still competitive for that long; who really cares about tech boosts in the Modern Era if the game was effectively over the moment you got Tanks? The core game just isn't very good at that, so in the absence of a good balance mod, this sort of thing won't have the impact I'm describing.

Although this is a standalone mod, afaik Gradual Research (aka Balance - Research) is primarily used as a part of the Balance - Combined modpack. In the Balance - Improvements mod the Academy is buffed at Scientific Theory. Also, with this mod by the Modern Era, techs cost 180% from vanilla.

So your points are perfectly valid in this context.

Thal - I'll bring it up again: shouldn't this thread be renamed?

 

[Spatzimaus]

By the time your academy could be producing 10 beakers per turn, techs are expensive enough that the benefits from bulbing would be many dozens of times that.

Sorry, but no. Let's say the game lasts 400 turns, which is the assumed starting point for a Future Era start. Let's say you got your first GS on turn 100. So that's 300 turns during which an Academy produces beakers.
+4 for 300 turns is 1200 beakers. That's the cost of a late Renaissance/early Industrial tech, right there, and would buy a dozen Ancient/Classical techs. Okay, we might subtract half off that as Opportunity Cost for not placing a Trading Post on the worked tile, so the practical gain might be more like 600. It's still far more than the cost of the techs available at the time you gained the GS (talking about your FIRST GS here); late-game GSs would have less time to build up that research surplus and would gain more from bulbing, so there's obviously a break-even point somewhere in between.
Obviously by the time you're getting +10 a bulb can save a huge amount, but remember that the Academy that's now getting +10 was getting +8 before that, and +6 before that, for hundreds of turns. It all adds up.

To see what I mean, let's pretend your city got a University on turn 150, an Observatory on turn 200, a Public School on turn 300, and a Research Lab on turn 350 before the game ends on turn 400 with a spaceship launch. You'd only place an academy near a core city, so it's reasonable to assume it'd get the full set ASAP. That means, using the core game's numbers:
> 50 turns of +4 = 200
> 50 turns of +6 = 300
> 100 turns of +8 = 800
> 50 turns of +10 = 500
> 50 turns of +14 = 700
Total: 2500 beakers. T14 techs (first tier of the Modern Age) cost 2600. Again we might chop off a fraction to represent the lost gold/production/whatever on that tile, but the fraction is less than before, because research buildings in the core game scale faster than ANYTHING. (Markets and Banks boost 25-33%, while research buildings are a uniform +50% until you hit the +100% lab?) Okay, Observatories can't be built in every city either, which'd possibly knock 400 off the total, but still over 2000 gained. (Of course, if you had a city that DID get an Observatory, then obviously you'd build your Academy there, so it's reasonable to leave that amount in.)

So, let's say you get a GS on turn 100. You can:
1> Build an Academy, and over time get a number of bulbs equal to the cost of an endgame tech. (And it's not just about the raw number; you'd be getting all of those intermediate techs a little bit faster as well.)
2> Use him right away, to gain a tech that might only cost a couple hundred bulbs. (Sure, that might take you dozens of turns, but it's still only a couple hundred. On the other hand, it'd give you the intermediate techs even faster than #1, some of which would improve your growth/research, so it has its merits.)
3> Stockpile him, paying maintenance along the way, to bulb an expensive late-game tech and/or one you REALLY need at a certain point, but get nothing until then.

Obviously, in my own mod (which is designed to have the game last 700-800 turns instead of 300-400) the break-even point is very different. But the core game's math still works in this way; option #1 is often the smart choice already, even if it's not the attractive game-changing choice we often want.
And again, we're talking about mods here, so you can make that option even better with minimal effort. Like I said, if the Academy gained an additional +2 research at some mid-era tech, wouldn't that shift the balance of whether the Academy is worth building? So if you don't like bulbing, my point was that, through mods, you can make bulbing the WRONG choice relative to making an Academy, without actually having to alter the bulb mechanism itself.

 

[Ahriman]

So that's 300 turns during which an Academy produces beakers.
+4 for 300 turns is 1200 beakers.

Aggregating yields without discounting over a 300 turn period is an insane way to try to measure benefits.

I would *far* rather have 2000 beakers on turn 100 than 4000 beakers on turn 400.

But the core game's math still works in this way; option #1 is often the smart choice already

No its not. The ability to upgrade to rifles *now* rather than in 20 turns is far larger than arriving at Advanced Ballistics a few turns earlier.

 

[Spatzimaus]

I would *far* rather have 2000 beakers on turn 100 than 4000 beakers on turn 400.

The two aren't mutually exclusive, which is the whole point here. Gaining a steady stream of beakers means you don't just get one Modern era tech a few turns earlier. It means you get DOZENS of techs a turn earlier than you would otherwise; while those few research points, as a fraction of your empire output, will drop as time goes on, you'll continue to get a benefit indefinitely. Bulbing a tech, on the other hand, saves you a dozen or more turns on one tech, and obviously makes the ones after it arrive sooner than they would otherwise, but the benefit drops quickly because moving up the tree without increasing your ability to generate research points means that it'll actually take more turns to research each tech than if you'd waited. So once you reach the point where the number of bulbs generated since building the Academy equals the cost of the tech you would have bulbed at the time you got the GS, you'll be at exactly the same tech as you would if you bulbed, AND be in better position beyond that.

(Obviously, since tech costs go up, the number of turns it'll take to break even will also increase, eventually reaching untenable values. So a late-game GS, obviously, should be bulbed.)

You're also grossly skewing the numbers. By turn 100 you're lucky to be researching techs costing 100 beakers (the entire Classical Era, basically). Even waiting an era or two won't get you anywhere close to 2000. According to the Era file, for a Standard speed the game assumes that the Medieval Era starts around turn 125 (250 AD), the Renaissance in turn 185 (1250 AD), the Industrial in turn 250 (1700 AD), the Modern in turn 325 (1905 AD), and the Future in turn 400 (1980 AD). So you wouldn't be hitting 2000-beaker techs until turn 300ish at a normal pace; even at turn 200 your techs would only cost 600-700.

So using a GS to slingshot to, say, Civil Service saves you a whopping 400 beakers. Relative to the general scientific output of an empire at that point it's still a decent number of turns, so it sounds good in the short term, but that's only 100 turns before an Academy would be more useful, less with the building multipliers. With a University (assuming no jungles) and an Observatory, that's 50 turns. (Add in a National College and it's even better, down to 40 turns. Yes, it's a National Wonder, one per empire, but since you can choose which tile gets the Academy, this isn't a drawback.) So for the first 40-50 turns after getting the GS, someone who bulbed CS would have an advantage, but after those turns, both sides would be at exactly the same point: 400 beakers beyond where they would be if they'd not had a GS, no more, no less.

The difference is that the person who bulbed had 40-50 turns with better units/buildings/whatever, while the person who built the Academy will continue to get more research points from then on. Which of these is more valuable will depend on how badly you needed those particular units and buildings. If you absolutely needed those Riflemen NOW, then sure, go nuts and bulb it. But if you think you'll be safe long enough to get Infantry (or are just waiting for Tanks), then the long-term investment is a better path. And we're talking about decisions often made long before that point; when you get a GS in the Renaissance, building an Academy might make it take a bit longer to get a nearby unit like Musketmen, but you'll get Tanks sooner than if you'd bulbed (since they'll come well after the break-even point).

While the costs of techs obviously raise quickly, which'd increase the number of turns to break even, your research multipliers also increase significantly, at least in the vanilla game. So the break-even point doesn't raise THAT quickly.

(I'm not going to get into the minor second-order effects, like how bulbing a tech that gets you into a later era tweaks the research agreement costs or city-state bonuses. Those just aren't important enough to skew the balance.)

 

[skallben]

No its not. The ability to upgrade to rifles *now* rather than in 20 turns is far larger than arriving at Advanced Ballistics a few turns earlier.

I completely agree. That among lots of things like earlier Fertilizer = faster growth = more science faster. Or the possibility to bulb Scientific Method and buy public schools right of the bat, short term gold heavy but gives a hell of a boost to science and nets you a tech closer to industrial era.

Oh and future era is far more likely to happen before turn 300, given you play normal settings ofc.

 

[skallben]

Posted just before my last one so I'll reply to this aswell.

The two aren't mutually exclusive, which is the whole point here. Gaining a steady stream of beakers means you don't just get one Modern era tech a few turns earlier. It means you get DOZENS of techs a turn earlier than you would otherwise; while those few research points, as a fraction of your empire output, will drop as time goes on, you'll continue to get a benefit indefinitely. Bulbing a tech, on the other hand, saves you a dozen or more turns on one tech, and obviously makes the ones after it arrive sooner than they would otherwise, but the benefit drops quickly because moving up the tree without increasing your ability to generate research points means that it'll actually take more turns to research each tech than if you'd waited. So once you reach the point where the number of bulbs generated since building the Academy equals the cost of the tech you would have bulbed at the time you got the GS, you'll be at exactly the same tech as you would if you bulbed, AND be in better position beyond that.

The problem with your reasoning as I see it is that you see this from strictly a long term science progress perspective. There are advantages that your numbers won't tell, a short term investment on it's own should not be compared to a short term investment made to make possible greater goals for long term benefits. The benefit your whole game gets by bulbing rifles can give you the possibility to not be destroyed or make a successful war. New cities = new science. And power, gold, possible resources and potentially more regional safety for yourself.

According to the Era file, for a Standard speed the game assumes that the Medieval Era starts around turn 125 (250 AD), the Renaissance in turn 185 (1250 AD), the Industrial in turn 250 (1700 AD), the Modern in turn 325 (1905 AD), and the Future in turn 400 (1980 AD).

Problem is this is nowhere close to reality, game progresses way faster. Maybe it's due to difficulty level.

So using a GS to slingshot to, say, Civil Service saves you a whopping 400 beakers.

Hmm, how many civs can do this?
Pre-education GSs are largely unattainable in most games (or am I missing something?)

The difference is that the person who bulbed had 40-50 turns with better units/buildings/whatever, while the person who built the Academy will continue to get more research points from then on. Which of these is more valuable will depend on how badly you needed those particular units and buildings.

Like I said before, I think this has everything to do with what you do with your new wonder/unit/building/whatever. If you bulb for a reason with a clear goal there are benefits to be reaped.

If you absolutely needed those Riflemen NOW, then sure, go nuts and bulb it. But if you think you'll be safe long enough to get Infantry (or are just waiting for Tanks), then the long-term investment is a better path. And we're talking about decisions often made long before that point; when you get a GS in the Renaissance, building an Academy might make it take a bit longer to get a nearby unit like Musketmen, but you'll get Tanks sooner than if you'd bulbed (since they'll come well after the break-even point).

Gut feeling tells me you war less in your games than me. The time between musketmen and tanks could let you conquer half the world. Really. I know what you are saying, there are times I wonder-monger and let loose my industrialised powerhouse later on aswell. Renaissance pretty much encourages you to pick a route, builders vs. military. Say for instance that I play a builder game and get my GS at Education. To get as many wonders as possible, it will be a tech race and I cannot wait until Industrial era for that.

---------
Just to make one thing clear, if Libraries had still had specialist slot I would generally agree with you. Bt then again, a steel beeline

I think to a point this is all circumstancial and playstyle comes into it. Bulbing with no goal is silly in my book. Anything without a goal is silly in my games but then I am also perfectionist with my Civ games.

 

[Tomice]

Its ok, but its not as simple.
Giving a fixed set of beakers is much simpler than changing which techs you can or can't access with a bulb, which also leads to MM to make sure that you can get exactly the techs you want when you use the scientist.

Also, Seek's point about it messing with Great Library.

Well, its hard to get more than ~1 great scientist in the early game, and 1 beeline isn't that powerful.

I see your point (and Seek's), so I'm fine with fixed beaker method. Might really be the better choice.

 

[Thalassicus]

My primary concern with the fixed beakers method was early slingshots. Since they basically removed the option for us to use specialists in early game (though that reduction of choice does bother me), the early slingshot problem itself is no longer an issue. Therefore, it seems reasonable to use the fixed beakers method. It fits with the fixed hammers Engineers give.

I think most of the discussion above is just about the concept of time value of money:
http://en.wikipedia.org/wiki/Time_value_of_money

It's an enlightening read if anyone wants to learn more about the concept. I ended up going into this topic a lot in detail in an accounting class. In particular what you want to look at is rate of value inflation. Since I haven't seen the inflation rate of CiV researched anywhere yet, I tend to avoid numerical analysis in this realm and base decisions off playtesting experience.

Basically I do what accountants do -- make &%$# up!




Seriously, how can we say a new machine is worth $50,000 for each of the next 10 years with 20% resell value deprecation per year? It's pure BS!

 

[Seek]

Calling testers!

I've gotten the GS/RA mod I described a few posts back to where I think it's pretty well balanced.

If anyone would like to test it, they can find it here; I would love the feedback!

Thal: There is some extraneous code in one or two of the lua files, they are harmless, but it would be nice to eliminate non-essentials. If you have a moment and are willing to look them over and let me know what I can safely delete, I would very much appreciate it. I'm totally in the dark with lua, so I'm hesitant to do so myself. Thanks in advance.

 

[Thalassicus]

Sure, I'll take a look at the code.

On that topic, something I'd like to mention again is research agreements. The ideal solution I see is marekb's suggestion (discussed on page 2 of this thread) combined with increasing cost per agreement (like golden ages) instead of per era. This of course isn't possible yet, but eventually it's what I want to do. The per-era route is one I've seen many mods take so I guess I'll finally give in to pressure. The main problem is the nerf to isolated starts, which are already sorta weak (average expense per agreement is higher for them with the per-era shift). I don't really like replacing one problem with another... but since the problem of late-game research pacts is probably bigger than the issue of isolated starts, it seems like an acceptable temporary tradeoff.

It'd be great to find a way to counterbalance the nerf to isolated starts with a buff in another way, and am open to suggestions. I've started exploring the start location placement code and that might be a route to take... though the code is so insanely complicated (over 10,000 lines) it'd be very difficult. I'm wondering if there might be a better option somewhere out there.