Intersite Game - Turn Discussion Thread

[Darkmantle]

The first to increase score is most likely built a warrior and allowed pop to grow so they are the least likely person to attack. Anyone who had the slowest score increase probably went worker or worker worker first and is the best person to attack. Unless of course they just suck but that is also a good reason to kill them anyway.

[kjn]

Well, score increases from tech is readily separable from those from pop, at least early game. We know one civ went warrior first, and they have a three-food surplus, and will grow to 2 T8, but that will only give a score increase to 31.

Anyway, we will increase in score quicker than the other teams early game, not since our play is superior, but because we have simply started with more expensive techs (ie the techs we have to research are cheaper), and because we will have our first tile improved a turn before any other team.

If we want to sow confusion (and IMO it is not worth it) about our research, go Mining this turn so that it looks like we researched Mining on T9 instead of Hunting, without giving any indication at all that we are research-swapping. If we are to do C&D to the enemy, do it without calling attention to it.

That can make people believe we are going for BW from T0. But IMO it is not worth it.

[Ruff_Hi]

I am all for swapping techs ...

Cost Nothing
Benefit Enemy confusion, hidden strength

A big reason Sullla got dog piled in that pitboss was that he was tearing it up. Sandbag where you can and you (appear) to stay with the pack more.

[spacetyrantxenu]

If it costs us nothing and there isn't enough for us to do in the opening turns, swap to Mining. We have a surplus of analysis for a small decision like this. I find it fun to mess with people anyway. I doubt we'll be the only team trying to hide research in this game and isn't it always better to deny your opponents info whenever possible? They'll figure it out before too long if they're tracking techs, but why make it easy for them?

#sociopath

[T-hawk]

To the "no downside" crowd, there are in fact a couple, although incredibly minor:

- It commits us to our next tech sooner, rather than waiting until Hunting actually comes in.

- We invest beakers in the next tech without a known-tech discount, which we might pick up later if we contact a number of civs that know Mining.

But I agree with the chorus that the obfuscation is more valuable than these very tiny concerns. The known-tech discount is 30%/9 = 3.3% per civ, so it would take contact with 3 civs knowing Mining to get a single beaker of discount, which is probably unlikely.

[sunrise089]

I agree with the swap techs crowd, though I also very much appreciate T-Hawk thoroughness in showing us this is a 'very low cost' maneuver not a no cost one.

I think I understand where Sullla is coming from, but I agree more with spacetyrantxenu - there isn't a lot to do, so if we get some small satisfaction and confuse anyone just a bit at no cost to our micro plan, then why not? As a bonus this sort of fun might make us all pay a litter closer attention towards finding future opportunities.

[kjn]

I had an hour while traveling, and wrote down the following thoughts about counter-C&D:

Counter-C&D efforts

How to appear less threatening or dominant than we are, or how to give misdirection on our purposes, strengths, force mix, or similar.

Research misdirection

Post-Alphabet there is little to do about this - everything is readily shown to all parties. Before Alphabet, it is sometimes possible to infuse confusion as to which techs are researched.

Researched techs are apparent through two mechanics: soldier count increase and score increase. By timing score increases with research costs and GNP estimates, it is often possible to deduce the tech researched.

There are two schools of thought for this research-swapping. The first is to switch relatively often and for long times. This usually makes it clear for other parties that a counter-C&D effort is in progress. This is especially true if GNP varies for no reason, or if research times become unreasonably long. The other, which I favour, is to only switch when it can realistically confused for another tech by an outside observer.

Eg, in our case right now we can delay research of Hunting by 1 turn by spending a turn researching Mining instead. The soldier count increase is 2k for both techs, so the only differentiator is the research time, 8t for one vs 9t for the other. Since both techs are valid techs to research early in the game, and GNP stays the same, this is undetectable from the outside, and it might make them think we are going Mining -> BW (a very strong military and production opening) instead of Hunting -> Pottery (a strong growth and economy opening) once we switch to the discounted techs.

Ie, in this case the research swapping can serve as a deterrent. The subterfuge will probably end on T21-22, when we finish Pottery and start Mining, and definitely around T30, when we finish Mining for real, but it might be enough to dissuade an AGG warrior or Quechua rush.

Do note that this latter purpose can only be fulfilled by making it appear like we researched Mining first. Further research-swapping would actually be detrimental to this subterfuge.

GNP hiding

The graphs and demos will probably be closely followed by most teams, and can't be hidden easily or at all. But it is still possible to use masking efforts here.

Great People Points is one effective way to mask research. Two priests that are run for 17 turns to generate a Great Prophet that bulbs Theology generate the equivalent of 53 beakers per turn, and all that shows up in the demos are 2 in GNP and 2 in production. Later on the efficiency of Great People is lowered, but I think we should look into bulbing as one possible way to get ahead without being noticed as getting ahead.

This means that wonders that help us generate more GPP serve a good purpose here. These are:

• The Great Library
  
• National Epic
  
• The Parthenon
  
• Statue of Liberty (but game is probably over by this time)
  
• Temple of Artemis

I'd actually go so far as to add the Parthenon to our list of possible wonders to target, on par with the Oracle (go for it if we see the opportunity).

Another thing to keep in mind is to not run binary research. Binary research is slightly more efficient and flexible, but a 100 GNP spike is very noticable, far more noticable than a steady 80 or so GNP. If most other teams run some sort of binary research, then we can be sure that our steady rate is compared to their spikes. It won't fool people who follow techs researched via score increases, soldier points, and the tech trading screen, but detecting a GNP lead that way is more work. A GNP spike is a simple, easily understandable number.

[SevenSpirits]

- We invest beakers in the next tech without a known-tech discount, which we might pick up later if we contact a number of civs that know Mining.

The main effect of the known tech bonus at this stage of the game is that if we've met someone who knows a certain first-row tech and don't finish that tech exactly, we will lose a beaker to rounding error.

I don't remember tech costs, but if this is a possibility I suggest we just work on hunting until it's done.

[spacetyrantxenu]

Good points kjn, but we can't ignore the old standby measure of progress, score. If we go full Parkin on wonders (no offense intended! ) that will definitely show up in score, even before we start powering up our economy through GPPs. I'm not saying to avoid building wonders any more than anyone would suggest we should slow play our start to look less scary. If we're good enough or fortunate enough to race out to an advantage, well it will be up to our diplomacy team and our war department to see us through.

One question about binary research, I haven't done research into how and where beakers get floored and wasted, but I know we're efficient at 0% and 100% slider, and in my SP games it usually seems efficient at 50% slider as well. For the more observant among us, is that a correct observation? Is do we lose beakers to rounding when running a 50% science/tax slider? If not maybe we could artificially cap our research around 50% if we are worried about it becoming too visible a sign of our position to our opponents (and, importantly, only it doesn't slow us from meeting our tech targets on time). Obviously we're not at that point of the game yet so maybe this is a conversation to have later, but I'm still curious to know about 50% slider efficiency and what opinions you guys have about appearing weaker through not running regular binary research.

My personal vote would be to run regular binary research and damn the diplomatic consequences.

[novice]

You never lose more than 1 bpt to rounding but binary research can be useful for a lot of other reasons.