SGLs and Me

[Sarxis]

Originally posted by skywalker
Don't kill me, either. In a recent game as the Babylonians, I got an SGL from Writing ...

What difficulty level?

[lockstep]

I did 5 test runs using Catt's scenario, popped every available tech and never got a SGL. I'm a believer.

BTW, I was wrong regarding Republic and Monarchy - it is possible to pop them both in the same game (and therefore to pop 19 techs) because there are a lot of ancient techs required for age advance (e.g. Map Making) that aren't a prereq for government techs.

Originally posted by Catt
Strange! Not popping a currently-being-researched tech is a tried-and-true tactic and also an event I never saw despite popping dozens and dozens of huts while researching available techs.

After my test runs I suppose the answer is: You can't pop a currently-being-researched tech unless it is the only tech available for research at the moment. (Note that this is quite possible for Map Making under stock rules.)

[Kuciwalker]

Originally posted by Anun Ik Oba
What difficulty level?

Regent

[UnityScoutChopper]

Dang, I wanted to be the first to post a "don't kill me" -- I had a game with 1 SGL early enough to nab the Pyramids, and another game (Byzantine), where I could SGL-rush both the 'mids and the ToA.

Pretty easy game when that happens. ;-)

USC

[Arrian]

I got an absurd string of good luck last night. Playing Greece, Standard, 8 civs, Monarch, unpatched.

3 ancient age SGLs. Pyramids, Great Library, Sun Tzu, all free (IIRC, I got an SGL either from my last ancient age tech, or the free tech I got - feudalism. Hence Sun Tzu).

I fired up Conquests to mess around with settler pumps and the timing and manner in which I "shut them off" as per the other thread, and this was the result. Actually, it was my second main try (several others with terrible terrain were tossed). The first didn't go well for multiple reasons, but primarily because I lost focus and didn't follow my own plan (I didn't build nearly enough workers). I think, however, that I saw the AI use an ultra-early SGL for the first time ever, though. In between turns, the game did that telltale little slowdown/hiccup it does when a leader is created. The very next turn, in 14somethingorother BC, the Iroquois built the Pyramids. Too bad for me, since I got a SGL from the very next tech I researched. But anyway, like I said, I lost focus in that game (why was I sitting around with ~1500 gold and access to iron, and NOT doing a mass warrior -> sword upgrade to go on a killing spree? I've no idea what I was thinking). Then the Zulu invaded and triggered my GA in despotism, which I despise.

Anyway, in the second game (with the 3 SGLs) I did something I cannot recall ever doing while at peace: staying in despotism long after I had the republic. I researched Code of Laws first, and then Philo second, so Republic was the free tech. I had it very early. But I then went ahead and researched literature, math, currency, construction and monotheism (50 turn pace on that one) before finally making the switch. The reason was simple: I had so many units (hordes of workers... at one point getting up to ~45 of them) that I thought the support costs would be too high, especially when I only had 2 luxuries. I needed time to build marketplaces & temples at the very least. I now wish I'd saved the game back when I got republic, so that one could test out how brutal the support costs would have been. I kept my pumps going longer than I have in the past, building worker after worker, and using these to develop terrain in a hurry, with the intention of then using them to boost the pop of various cities from size 7->12. The pump cities themselves grew to 12 on their own, pretty much. I think I may have added 1 worker each to Athens and Sparta.

Curraghs and Galleys have brought me contact with the rest of the world. That gave me an extra influx of cash, because the overseas civs were at first isolated and at least 2 of them had built up decent gold reserves (one had 200g, the other ~150). The tech lead is, well, what one would expect when one is granted three of the best wonders in the game for free (the very best AI civ is still 1 or 2 techs short of getting out of the ancient age). I upgraded a bunch of warriors straight to med inf (90g a pop, ouch), built some horsemen, and smacked down the weakest civ in the game (Russia). This gave me my 3rd luxury... Ivory. The Statue of Zeus completed 12 turns later in Athens. Not that I need it, but it's nice to have, right?

I intend to have fun with overseas imperialism in this one. I've no need to crush the other civs (France, Carthage) on my continent, especially since this is unpatched so the FP is useless. It's all about luxuries now, and seeing just how far I can push that tech lead. Hmm, I need more galleys. Lots more.

Challenge? Clearly not. The game was pretty much over when I rushed the Pyramids. But I'm having a blast anyway.

-Arrian

[Theseus]

Sounds like fun.

I think we are going to see a lot more late departures from Despotism.

[Catt]

Originally posted by lockstep
After my test runs I suppose the answer is: You can't pop a currently-being-researched tech unless it is the only tech available for research at the moment. (Note that this is quite possible for Map Making under stock rules.)

I think that's exactly correct. In my test if you block Writing (by researching it at 0%), you'll get every tech except writing and those requiring writing -- so you'll get the very expensive Construction, Currency, and Monarchy, but once all available techs are gone, you just might get Writing from a hut even while you're researching it.

Catt

[Aqualung71]

AT LAST!!! FINALLY!!!

After playing many more Conquest games than is healthy over the last 2 and a half months, I have now just generated my first SGL! I guess that proves my copy of the game is not broken (at least, no more broken than everybody else's....ouch!). Another distinct possibility is it shows how bad a player I am!

A question - can you keep more than 1 SGL at a time? I know you can't with MGL's.

On the science question, I used my SGL for the Science GA since it was late Industrial and no wonders in sight. I recall it reduced the turns to research Radio from 8 to 7 (not 25%, but rounding I guess), so nothing spectacular.

To my simple brain, a 25% research bonus over 20 turns effectively saves you an aggregate of 5 turns regardless of what you're researching. Depending on your game circumstances, this could be incredibly valuable if it gives you earlier techs that you can sell back to the AI for other techs, cash and gpt....I've always assumed that selling techs for gpt is a way for the human player to slow down the AI research rate. Any comments?

However, comparing the benefit of a "5 research turn reduction" to the value of an individual improvementswonder in one city only is not that easy.

So what's the consensus of opinion out there? I don't get the impression that many players use an SGL for the science bonus.

[vmxa1]

I don't knwo for sure, but I beleive you can.

[vulture]

My first conquests game, and I stayed in despotism for a while after I had republic. I found it pretty straight forwards to add up the number of cities (3 support) and towns (1 support) to see what my support costs were going to be, and then added up the number of pop points I had, plus one per town/city for the extra commerce generated by republic (ignoring the effect of markets and libraries - I didn't have markets yet, and was mostly building libraries at the time). It was pretty easy doing this to estimate the point when switching to republic would give the same post-unit-support income as despotism, but I didn't reach that point until quite some time after republic became available. Switching straight away would have left me about 2 gpt income with 100% tax - and I needed to have the luxury slider on 10 or 20% to keep my cities happy.

[Mad Bomber]

My personal best is 3 SGL's in one game. It happened when I was playing the Phoenicians in the Mesopotamian conquest. In the game only had about 5-7 cities that I had eeked out (being sanwiched by most of the other civ's rather quickly) Just after I had completed my first wonder, I got 2 SGL's on successive completions of tech. With 3 Wonders completed, I was on my way to a rather easy (and completely bloodless) victory. To top it off my third SGL sealed the victory in 460BC.

[vulture]

Originally posted by Thriller
AT LAST!!! FINALLY!!!

After playing many more Conquest games than is healthy over the last 2 and a half months, I have now just generated my first SGL! I guess that proves my copy of the game is not broken (at least, no more broken than everybody else's....ouch!). Another distinct possibility is it shows how bad a player I am!

A question - can you keep more than 1 SGL at a time? I know you can't with MGL's.

You can have more than 1 at the same time. IIRC someone did a test on the strategy forum by setting every tech to give a bonus tech, so they got the entire tech tree in one turn, and got multiple SGLs out of it. Quite a cunning test, I thought.

On the science question, I used my SGL for the Science GA since it was late Industrial and no wonders in sight. I recall it reduced the turns to research Radio from 8 to 7 (not 25%, but rounding I guess), so nothing spectacular.

Imagine, for the sake of argument, that radio costs 800 beakers. If you can get it in 8 turns, you are producing from 100 to 114 beakers per turn. With an SGL boost, that goes up 25% to 125 to 142 beakers per turn. To get to 7 turn research you need 115, so you're guaranteed to get that. To get to 6 turn research, you need 134, which isn't guaranteed. However it means you can either generate more cash along with 7 turn research (you'll be able to lower the slider for the last turn more than you would with no boost) or you'll be able to alternate between your current science setting and 10% higher, to get radio in 6 turns while effectively breaking even. So you can get the full effect, but you have to work at it.

Another thing to bear in mind - research times drop by 20%, not 25%. You get a 25% boost to your science output. If O is your old science production, and N is the new, then:

N = 5/4 O

the old time taken is To = B / O (B is no. of beakers needed). New time taken is:

Tn = B / N = 4/5 B / O = 4/5 To = 0.8 To

So you don't knock 25% off the research time, but 20% (so radio should take 6.4 turns, which means 7, in a simple estimate).

To my simple brain, a 25% research bonus over 20 turns effectively saves you an aggregate of 5 turns regardless of what you're researching. Depending on your game circumstances, this could be incredibly valuable if it gives you earlier techs that you can sell back to the AI for other techs, cash and gpt....I've always assumed that selling techs for gpt is a way for the human player to slow down the AI research rate. Any comments?

Saving 5 turns is true, despite the appearance of what I wrote above, because you are saving 20% of the research time on 25 turns of research, doing it in 20 turns instead. But you may have to play around with the slider to save all 5 turns, rather than converting it to extra cash.

Selling techs to the AI doesn't slow them down - if anything it speeds them up. It just speeds you up more by allowing you to run at 100% science, so yuo can pull away from them (overlooking the fact that the trailers can research different techs and trade with each other, while you have to research everything yourself).

Assuming the AIs all have banks and universities built in equal abundance, it doesn't make much difference whether their trade goes into research or buying techs off you at full price - they end up with the same number of techs per 20 turns either way. The exception to this is the tech leader, who will pay up to double the value of the tech, so he can be slowed down if you get the full amount. In practice though, I think most people will sell techs for way below their asking price, because if they don't, other AIs will as soon as you've sold the tech to one of them. Knowing that Scientific method is worth 100 gpt to a civ, would you accept 30 gpt from them if that is all they can afford if you know they'll buy it from your main rival for 30 gpt next turn anyway? Net result, all the trailing civs get to keep up, buying techs from you for a fraction of their value. The reason they have so little gold free is either they are doing their own research (not that common) or because they are already funneling all their own gold into the coffers of your main competitor.

I'm gradually coming around to the idea that selling off techs lets the AI catch up with you sooner (especially if you already have enough income to stick at 100% science), whilst it also speeds you up, but not as much. Whether this is worth it, or whether you'd prefer to slow yourself down, slow them down more, and have more money disappear from circulation into research is probably dependent on the game situation and how you want to win.

Best way to slow the AI down - start a war with the main researcher amongst them, and get everyone to join in.

However, comparing the benefit of a "5 research turn reduction" to the value of an individual improvementswonder in one city only is not that easy.

I'd guess that in the ancient and medieval eras, the wonders are almost always a better bet. There may be exceptions (if the only wonder available is the great wall, although that's been beefed up now), but wonders will do much more for you over the course of the game. Early industrial, you definitely want ToE and Hoover. If you know you have prebuilds timed nicely for them then it might be worth using it for the science boost. Maybe you'd prefer to keep it for the Suffrage. Post Suffrage, I don't think I'd keep one for the early modern wonders unless the tech race was very close., so then I'd use it for the mini-GA. But I don't think I've ever had a game yet where the Ai was in a position to compete for any of the modern wonders even without my using prebuilds.


So what's the consensus of opinion out there? I don't get the impression that many players use an SGL for the science bonus. [/QUOTE]

[Mountain Sage]

Originally posted by Theseus
I think we are going to see a lot more late departures from Despotism.

Absolutely, or a revival of the old Despotism-Monarchy-Republic (or even Demo) road, especially if you have many units.
Feudalism seems to be something intermediate between Monarchy and Republic. I stayed in Feudalism during my whole game on my Viking's domination game.
Any experience with Feudalism?

[Aqualung71]

Originally posted by vulture
Selling techs to the AI doesn't slow them down - if anything it speeds them up. It just speeds you up more by allowing you to run at 100% science, so yuo can pull away from them (overlooking the fact that the trailers can research different techs and trade with each other, while you have to research everything yourself).

Yes, of course selling techs to the AI will move them forward in the tech race immediately. Where I was coming from was the idea that if you sell scientific method to the leading AI for 150gpt, for example, then that gives it a lot less available gpt to increase its research rate or to generate gold to buy techs from other AI civs, so this will slow them down overall. At the same time, it gives you more gpt avaiable to increase your own research rate. Of course, you really need to sell techs to all AI's on the same turn if you can, soaking up whatever gpt they have available (assuming your reputation still allows you to). Then you watch your current deals and whenever a gpt is up for expiry, you try to repeat.

I'm not sure however, whether the combination of these factors would help you pull away from the AI's in tech, or help bring them closer to you. I've found that a tech lead can be quickly eroded when AI's are researching different techs and in a couple of turns you can move form being 5 in front to being almost equal. This is when you kick yourself for not selling them the techs first!

I guess there's no hard and fast rule on this issue. As you've said, it depends largely on your game position and your objectives.

[Guynemer]

Perhaps this has already been addressed, but are some techs much more likely (as in guaranteed) to produce a SGL for the first to discover them?


I've played a pair of games since C3C, one as the French and one as the Yanks, and in both I received a SGL with philosophy and democracy.