SGOTM 22 - Lurker thread

[TalladegaKnight]

A good SG map should try to reduce the random factors in game, not the opposite. Unfortunately, this SG enlarges the most effective or even deadly random factor, AI's choice of DOW target, since human player is limited by the conditions to fend off early invasion. All those limitation on expansions, the location of copper/Iron, no tech trade together greatly weaken human player's potential of defense. I'm not saying that this setting is too difficult, but a DOW from your neighbor could make huge difference on the result for 2 teams that are of the close skills, it's far from popping a silver in the capital. In SG15, human player were set to face the worse diplomatic situation and some teams (PD and Kaku IIRC) got early DOW, not a deadly event for them since human player got enough resources to defend and they got paid back by a early GG, and other benefits in PD's play.

The odd of producing GSpy, is another factor which is totally uncontrollable except the 1st one.

As I said before http://forums.civfanatics.com/showpost.php?p=13276144&postcount=2291

SG12 was the best setup map I had played, this map also eliminated the early DOW factor since human player was set to be in war with everyone from beginning.[emoji38]

Sounds like you just nominated yourself as designer for the next one. [emoji6]

 

[BSPollux]

The random factor thing allways comes up and each time "the worst random factor" is exactly whatever happened that game. This isn't chess, there is randomness in CIV. And my response is always: I think you shouldn't try to have as few dice rolls but try to have as many as possible so that it evens out. If you roll a dice once you can have extreme results (six, one...) but if you roll it 100 times the average will come close to the mathematical average of 3.5

Lets say I wanted to make a map fair that starts the player alone on an island and the game has goody huts enabled. If I placed only one hut on the island it would be the most random. The more huts you'd have the more likely it would be to have a reasonably fair outcome for everyone.

 

[Duckweed]

Sounds like you just nominated yourself as designer for the next one. [emoji6]

Nah, I just dropped by and couldn't help to say something. Most of the SGs I had played so far were excellent to outstanding designs. I had not realized how great of SG12 until I had played a few more SGs. The designer of SG12 was both a senior SGTOM player and map maker, so that he knew how important of those random factors could affect the game results.

@BSPollux

I have no intention to criticize your design and to sell my points to you. I respect and appreciated your work on designing the recent SGOTMs. However, while you have spent lots of time on making the maps, there are more players spending more time on playing the game. Those SGOTM maps are intended to provide as fair as possible platforms for competition.

I know very clearly how those build-in random factors works, they are unavoidable, but some are much more significant than others. A tough DOW from your neighbor could make 10+ turns of difference of teams of close skill, or even end the game of a team of lesser skills, while a battle lost at 99% odd is nothing in a game, you can only blame yourselves that you did not bring the extra units.

One of the basic rule for SGOTMs design is that a map should not let a team lose in the early or middle stage of the game. This game was the only game that ignored and even enlarged the most significant random factor in the game. There are ways to solve the problem even keeping your limitations, such as putting human player adjacent to a peaceful AI, or separated in another island, or even go for another extreme, bordering with an aggressive AI and human player is the only target so that all teams must prepare for the invasion.

 

[Frederiksberg]

I think the game design is very creative and following the game has been entertaining for the spectators. That said, I believe the game has been very frustrating for the teams that faced an early DOW. Xteam lost the game outright and the former phoenixes and Fifth were set back quite significantly while Anti and Suspects have had the luxury of picking their own targets in the beginning of the game. There is no doubt in my mind that this is the most random SGOTM so far.

I think you shouldn't try to have as few dice rolls but try to have as many as possible so that it evens out. If you roll a dice once you can have extreme results (six, one...) but if you roll it 100 times the average will come close to the mathematical average of 3.5

Lets say I wanted to make a map fair that starts the player alone on an island and the game has goody huts enabled. If I placed only one hut on the island it would be the most random. The more huts you'd have the more likely it would be to have a reasonably fair outcome for everyone.

I don't agree with this line of reasoning. The current state of a game of CIV is NOT the average of all events, hammers, commerce and food but rather the sum. Following your example we now add the dice throws together instead. This sum is a so called random walk and one characteristic of a random walk is that the standard deviation grows with the square root of the number of steps. Thus the difference between two outcomes of a series of throws grows with the number of throws. Similarly the difference between teams in amount of gold, techs etc would be significantly larger if the lone islands are populated with a large number of huts. Obviously some teams will get lots of maps while other teams will get expensive techs. It doesn't even out. You can try this if you don't believe it.

 

[BSPollux]

@Duckweed:

As I said I underestimated the snowball effect of my later changes and that was a mistake. But that's unrelated to your point about randomness.

You'll get my reply in a PM.

 

[TalladegaKnight]

@Duckweed

If you build it, they will come.

 

[Sam_Yeager]

IMHO as it stands I think only the A-Team is likely to have finished by the planned end date. It's possible that one or two further teams might lose/abandon the game by that date.

 

[Noble Zarkon]

Thanks for the game BSP

That was incredibly tough for Monarch though, what exactly did you do to those AIs??

 

[Sam_Yeager]

Thanks for the game BSP

That was incredibly tough for Monarch though, what exactly did you do to those AIs??

This post might help.

 

[Noble Zarkon]

Thanks - this game is the poster child for the mantra of the snowball effect, a tiny advantage at the beginning of a game (e.g. settling on a plains hill for an extra hammer) has significant effects way beyond what you would expect.

 

[Frederiksberg]

Missing the GW was not a tiny disadvantage because the GSpy was essential to keep up in tech.

Several factors made this game difficult:

1. Game rules prevented REX.
2. The AIs were given above average land.
3. AIs were aggressive and the teams started in the middle as a potential land target for 4 AIs.
4. The no tech trading options means that the AIs are given a multiplier on their research to compensate the loss of trade. The human player gets no such compensation (As far as I remember from some old discussion thread).

Looks like Anti and former Phoenixes will manage to win. I doubt that Suspects will make it. They had some good luck in the beginning with Brennus attacking Gilga and Saladin and Darius going at each other while they beat down Hannibal but now they are punished for not setting up a spy game and they lack a sufficiently strong offensive unit to go with their cannons. The game may still be winnable but time is running out. I guess they will still get bronze if a retired game can win a laurel

 

[Noble Zarkon]

Missing the GW was not a tiny disadvantage because the GSpy was essential to keep up in tech.

Yes no doubt this was a major issue for us - we did not put a high enough premium on the Great Spy / tech stealing.

My point re tiny advantages was the changes BSP made to the AIs which led to them racing away in Tech on Monarch despite no tech trading. Interesting to hear there may be a coding contribution to that as well.

 

[zbgayumn]

IMHO as it stands I think only the A-Team is likely to have finished by the planned end date. It's possible that one or two further teams might lose/abandon the game by that date.

I feel better knowing we weren't the only ones who tanked. I think we had a roughly 1500 year war with Brennus. In retrospect, I wish I had razed the one city we had captured. Maybe it would have set him back just a bit.

Not that that would've helped us win the game. At best, it would've kept us from losing by conquest. I'm sure we would've lost the Space Race instead.

I had never heard of the bonus the AI gets with No Tech Trading, but I knew something was going on. In our game Lib went in 1090 (IIRC)! That doesn't happen in Monarch games where trading is allowed.

 

[LowtherCastle]

What turn did teams get DoWed, anyone know? We monitored the fists and got Brennus to Pleased to prevent him from DoWing us, but I don't recall how early on that was.

 

[LowtherCastle]

Incidentally, BSP, Anti-TSR has no issues with extending the deadline for teams that wish so. Let them finish...

 

[kcd_swede]

Good gosh, NO EXTENSION...make them end their pain as soon as possible.!!!

Hard to believe that our delay of alphabet led us so wrong. We wasted a GSpy that could have got us into near parity in tech if we had known what the AI were doing. And then when we saw what they were doing after we eventualy got alphabet, we did not panic enough.

We set the game for conquest with Praets/catapults, and there was no catching up after that, for us.

I think it was fun (I enjoy the struggles more than the spreadsheet micro-managed fast victories). But it was much too contrived of a map (the scenario was OK, but the map was slanted so much...) to play like a normal game of civ BtS. Not a great way to inspire more participation in the next one, but a brutal game now and then can keep it interesting for the vets, I suppose.

Good game, everyone. Thank you BSPollux and AlanH for providing us with ulcer-inducing joy.

 

[LowtherCastle]

Bummer. I didn't look at any of your saves (if I even can yet) but I thought with 3 AIs DEAD you folks had a chance. Anti-TSR saw the runaway trains early on and our strategy was to stone-age the most runaway, each in turn. Ironically, the map was laid out perfectly for that, using galleys across the ocean: Saladin-Zara-Gilgamesh-Brennus-Hannibal-Darius. I don't know how the runaway sequence in other teams' games went but evidently targeting the closest AIs, Hannibal and Darius, first was a fatal strategy.

 

[WastinTime]

We put Darius in last place with a faraway city gift although his land was pretty poor and he may have ended up last anyway

Sent from my SM-G900P using Tapatalk

 

[kcd_swede]

Bummer. I didn't look at any of your saves (if I even can yet) but I thought with 3 AIs DEAD you folks had a chance. Anti-TSR saw the runaway trains early on and our strategy was to stone-age the most runaway, each in turn. Ironically, the map was laid out perfectly for that, using galleys across the ocean: Saladin-Zara-Gilgamesh-Brennus-Hannibal-Darius. I don't know how the runaway sequence in other teams' games went but evidently targeting the closest AIs, Hannibal and Darius, first was a fatal strategy.

Just getting a victory with only 2 of the 3 restrictions would have been reletively easy. Units in Rome didn't matter, but Pop limited the early expansion which usually defeats Monarcg AI... and not being able to connect taken cities was a pain. With the restriction on settling cities, it was going to be a tough game for us.

But some time about 200 turns before the end, we could see that 7 World Wonders was going to be the botleneck for us. At no point did we have enough of our inferior units to take on Ethiopia, who was teching like mad down the Wonder-tech path. We didn't even know who was running away in tech until it was too late to do anything about it. Some bad decisions, and poor assumptions on our part. Can't blame luck, as most of ours was good - with AI DOW's helping us, and some good rng help in the early game.

Early on, we were thinking chariots and then HA, as we knew a conquest game requires early attacking to kep down the big guys. We got scared by the spears that the AI had before we got archery, though, so we went for the crappy ice iron city instead.

Lack of coal, oil, uranium, (and probably aluminum?) was disappointing, but hardly mattered unless you decided (incorectly imo) that space race could beat a conquest. But before even 1000AD, we already knew that space era techs would be needed to have any chance on the 7 WW.

Honestly, I don't think I'd win this game if I started it up again for fun - even with map knowledge and knowing who needs beating down and when.

I have nothing against an extension for other teams that have a chance to win laurels. However, I shoud point out that we did do a very sub-optimal final flurry of turns in order to make the deadline... but otoh, we did so with knowledge that we would almost certainly fail no matter what, due to decisions made early that seemed good at the time but were not appropriate for the actual game situation (which we were blind to for far too long-our own fault). Anyhow, we had fun, so it was another success. Congrats to any teams that won.

 

[LowtherCastle]

We put Darius in last place with a faraway city gift although his land was pretty poor and he may have ended up last anyway

Sent from my SM-G900P using Tapatalk

Right, and one more irony is that we did that because he was the early AI tech leader. Then his growth stalled while others improved. Brennus started very slowly but his empire got so large he became a tech giant. Not sure BSP foresaw and planned any of that but it's quite interesting in hindsight.