Duplicate Test Map

bobbyboy29

I was saying boo-urns...
Joined
Oct 14, 2008
Messages
1,401
Location
Melbourne, Australia
Ok so I had an idea. Inevitably what's going to happen if this game ever gets up and running is that we're going to have a whole bunch of people arguing things like "workboat first"... "no worker first". And "Work the plains hill"... "No work the floodplain!"... And a whole bunch of analysis backing up each persons case.

My idea is to load an SP noble map, go into worldbuilder and turn all the tiles into ocean. Then, manually recreate the terrain (or what of the map we have seen so far) and put in all the units/cities/buildings etc... This way people can load the map and literally test out different paths and see where they take them. You can run as many different paths as you like and that way we can know which is best.

Later on if we're at war you could even put in all the troop positions and test what the most effective strategies are (kind of like having infinite reloads).

I can see a few obvious drawbacks.
1) We won't know exactly where we start so if we get it wrong you could run out of space.

2) If we don't know the enemy leaders/civs there might not be an efficient way to edit them in (unless we don't have unrestricted leaders in which case we could jjust add them all in...)

3) People would actually have to be bothered to go through the painstaking process of creating the map...

Let me know what you think of the idea!
 
Brilliant idea, I would never have thought of it myself. The beginning of the game is my favourite part and I'd certainly be willing to run such simulations. With one turn per 24 hours you could easily play 50 turns in a night in the simulation and make decisions accordingly.
 
I would like to point out that I have in the past used spreadsheets to good effect for early game simulations. I prefer them to actual simulation, as I find you get a much better overview of how your choices affect the longer term outcome, and a much better feel for how changing a specific tile assignment on an early turn will affect the outcome 20 or 50 turns ahead.

That said, actual simulations are useful as well, in particular to get a feel for things that would never be evident from a spreadsheet, such as barbarian threat.
 

I don't think that this would be against the rules, since it is not downloaded save, it is map created based on our current knowledge. And I'm in for this idea, the only thing that IDK how we would resolve, is where to put starting tile if the map isn't centralized.
 
I was just assuming we would keep a test map going with this many players available.
I can probably help with WB. I find it strangely calming and relaxing.
Given the slow pace of the game, it should be easy enough to keep the test map up-to-date for the extended opening.
 
I used this strategy extensively (at least for the first 100 turns) in the last SGOTM and I have to say without it we don't finish nearly as well as we did. Depending a lot on how much time I have, I could put it together, but that's a big if. To hit on some of the drawbacks here:

1) We won't know exactly where we start so if we get it wrong you could run out of space.

While true to an extent, you can use the forest/terrain type to at least get in the ballpark of north and south. The forests are a consistent line that can be used to gauge north and south and specifically if we are north or south of the equator. East west isn't too critical as the world wraps around. Your centering will be off (though if you have a coastal start you can make an educated guess as to where you are), but that's not nearly as important as the north/south location is.

2) If we don't know the enemy leaders/civs there might not be an efficient way to edit them in (unless we don't have unrestricted leaders in which case we could jjust add them all in...)

XML in the actual leaders in place of the 'wrong' leaders. You have to dig a bit to get the right 'settings' for each leader so it appears correctly in-game, but it can be done rather easily if one knows what they are doing.

3) People would actually have to be bothered to go through the painstaking process of creating the map...

Yea, it's painstakingly long, but I've mentioned I can do it if time permits and there's a few others here it seems that are open to it. The map itself isn't too difficult, especially if we're playing this one turn at a time. The really hard part is playing the turns identical to those of the server game to keep the food and hammer bins identical to that of the real game - to my knowledge there's no easy way to adjust that.
 
A sandbox sounds good - personally, I feel they are at their most useful for the opening turns, where what you have to model is simpler and theres less likely to be interferance from barbs/other players.

I like spreadsheets too, nice to turn my work to evil gaming use!

Have a simple one that tracks hammers invested in enemy cities via espionage screen, although we need city visibility for it to work.

Been having a fiddle with something to back engineer city sizes from pop numbers/demogs, but it's not quite there yet.

Would it be worth doing some 'Cloak&Dagger' on the stats/demogs and score information?
 
Would it be worth doing some 'Cloak&Dagger' on the stats/demogs and score information?
Oh yes. Oooooooh yes. You wouldn't believe the amount of info we gathered just by reading demo screens last game.
 
Oh yes. Oooooooh yes. You wouldn't believe the amount of info we gathered just by reading demo screens last game.

Hehe - I've seen it being done and it does seem there's a lot to be learned! :)

Think it's well worth it myself... I'd be willing to put some work into this... it's the sort of thing I'm sad enough to find most interesting. :p
 
Ah OK. Cheers classical_hero. If I'm reading your post right, it's against the rules to make a separate sandbox simulation of a part of the map?
 
Well you could make a simulation if you wanted based on the starting position, but it will not be like the map due to the fact that you do not know the map. The rule means that you cannot save the game so you can have a look at the map via the world builder. We will be strict on anyone found to have violated that rule, since it basically cheating, since you will get to see what everyone else has done.

I have been on SGOTM teams where we have simulated the map and I have found out that it is generally a waste of time. To be honest doing what Niklas suggested is the best idea on what you plan on doing.
 
Thanks for the clarification there CH. Save and WB would be all kinds of wrong, totally with you there.

I generally agree with you on the simming the map.... although I do think they can be useful for the opening ~50 moves, but I find the modeling too complex for a sandbox after that... and never quite know what the other players are going to do, etc.

I like spreadsheet tracking too - excel is your friend. Apart from that horrible paperclip thing. :p
 
well, in the demo games i participated so far (german forum:)), we ALWAYS had a "sankasten", meaning a sp map, we rebuild from screens from the actual game. no worldbuilder, no original saves, nothing.

i can even provide a small program (if the guys in the german forum who wrote it, allow off course), so you can adapt in your "sankasten"-game research, gold, food per city, GPPs basically everything for all civs, to turn it into a perfect copy of the live game...

...its a lot of work, cause you have to keep the sandkasten up to date, and you have to do it manually (same rules of not using wb and original save as her aplies in the german forum as well, off course)

BUT IT PAYS OFF BIGTIME, YOU CAN SIMULATE EVERYTHING, cottage growth GPPs, research... ...EVERYTHING:king::goodjob::D... ...if we dont do a "sankasten" here, i´m not too inclined to play (most likely wont), cause for me planning is the cornerstone of a DG... ...I like playing around with the "sandkasten", simulating different path and options up to 20-30 turns ahead, to see possible impacts:crazyeye:... ...and i can already tell you, from experience that the difference between teams who use a "sandkasten" and these that wont, is pretty big speaking of performance...
 
I would like to think I have the optimal play memorized by now :p

But just in case I am a little bit off.... see if you can get the rights to it Snaat....
 
The sandkasten thing sounds cool! :)

I think we must have enough people and a slow enough timer to get that right?

Not very experienced with WB. How easy is it to edit in things like toroidal wrap or slightly different mapsizes if our initial guesses are off?
 
The sandkasten thing sounds cool! :)

I think we must have enough people and a slow enough timer to get that right?

Not very experienced with WB. How easy is it to edit in things like toroidal wrap or slightly different mapsizes if our initial guesses are off?

its not that difficult and timeconsuming... ...map-type and size and other civs we should know before we start. then we have to estimate our starting position (left/right of equator), distance to poles. then we start with a rough simulation of only our land, adding neighbours and cities as we find them (takes about half an hour, i can make the first sandkasten if needed). adapting the sandkasten turn by turn is done in some mins. cause you only have to add the new infos and tiles and move the units according to the turnlog (i guess we will have a turnlog, describing all unit movement?)

later when exact position in known, we have to restart the sandkasten and put all details and our exact position in (this is more timeconsuming, but luckily, you only have to do it once). once the actual standing is reached, work drops to the few mins mentioned above.

how can a sandkasten be coordinated in a big team an with a turn a day:
easy:
one starts with a first map/plan of round X, posts save and short description, next who has time picks save and continues for round Y. wb-saves are not difficult to edit and if i can lay hands on the small program i named (expanded WB options), we can add all the small stuff also turnwise (food stored in cities, partresearched techs...). its nice to have that also, cause your own research is much better planned, when you have put in the actual techstanding of all neighbours.

from my experience, both is needed:
- sandkasten for planning ahead like 20 turns, saved every turn, then
- spreadsheets (openoffice would be best) to do a detailed turn by turn plan off every minor detail to reach the planned goal (therefore you need the save of the sandkasten for every turn, or you go crazy planning food, prod, tiles for every city and all worker actions and unit movements):crazyeye:

this helps the player executing the actual turn a lot and avoids mistakes also (mistakes still will happen, but at least we can reduce them)
 
Top Bottom