Hey everyone, I’ve learned my lesson from the previous round and have been merciful to my F5 key (thus far). However, my curiosity is getting the better of me, and I’m curious to hear how you guys approached this year’s challenge. I’m guessing it’s safe to start discussing these kind of things, or do we need to wait until the results are final?
Happy to share from my side as well, just want to make sure I’m not missing something before I spill the beans 
This is last years and i havent seen bot changes been allowed for the final in the past, may have missed something.
I was actually hoping we could at least see peoples logs and scores from previous tournaments to try and figure out folk’s approaches.
From previous years, I thought I had a good enough bot, the closest I got was in the top 10 or so, hoping this year is better.
If I make it into the top 6/8 I am really hoping there’s a chance to make some changes and fixes. I reviewed my bot a bit yesterday and ran it against another users and came to the realization that I’m only building 1 kind of building, instead of rotating through them like it’s supposed to. There’s some other optimizations I didn’t get to and I couldn’t get to territory at all…
I personally would love a small window for refinements for the top 8,
even if its just 1 week.
Territory did drop pretty late and the team was crazy on their game with last week’s updates,
Contestants were also on their game.
But even then the last engine release came out around 5 days ago.
Version 3.5 was 8 days ago and that one was still a bit wonky 
I feel that if the team could this year as an exception give us a few days for bug fixing incase we are top 8 that could help a lot.
On that note I would say lets not get into strategy at all yet.
Maybe this year they give us that extra few days for top 8.
But it could be a gamble regardless.
I would say if they announced a window that could be cool.
Territory can be a double edged sword and my strat is in for certain.
So like I had a weaker bot in event 2 than I had in event 1 even though I was building.
If you had a decent bot in event 2 you can still be top 8 in event 3
This approach does not negatively impact users who already submitted.
Because they might have territory and buildings implemented.
So I personally feel that a small window for refinements if you are top 8 would be amazing.
This obviously depends on how fast the team can determine top 8.
I expect they already know placements very early September.
If they could however after top 8 has been determined give like 3 days or so (including a weekend)
That might ease many users. I was already restless yesterday because in my mind I have been simulating territory.
And though my territory strat is pretty good.
I can see some improvements.
I do feel the team will have those results in very early next week for placements and top 8.
If that is the case maybe they can give us till the very next week (between determining top 8 and actually running the top 8 matches their side so they can prepare.)
I feel myself and other contestants could appreciate this.
**Note: My strategy isnt incomplete at all. So im not saying this from a position of weakness.
Gains on my own entry would be very marginal. As I already have pretty advanced territory management in place.
But nobody wants to go to the finals and watch a last minute submission mistake for example.
My vote is once top 8 is established to give a small update window for fixes and updates
My village management style is very laid back, I envision my little dudes as chilled hunter-gatherers with dreadlocks and guitars.
There are some suggestions/guidelines for the community, but really they do what they feel like, even taking a few days off sometimes for quiet contemplation.
Modern conveniences like buildings are shunned, they prefer to live outside in nature. And they have no concept of territory. As long as they are fed and kept warm, they are happy.
In earlier times there was a “bronze age collapse” event that would occur around round 1500, where the community would decide that wood was no longer important. But that was a bit too chilled: the village elders have sorted that out, so at least people aren’t freezing to death anymore.
I agree with the standard approach, that bots are final after the 3rd phase.
Changing that around now puts additional pressure on the competitors (anyone who isn’t available during that week is at an immediate, unplanned disadvantage ) and on the team running the challenge (more admin to re-open submissions, less time to run matches, validate them, prepare for the finals, etc).
Yes and no:
A point on:
Changing that around now puts additional pressure on the competitors (anyone who isn’t available during that week is at an immediate, unplanned disadvantage )
Technically the same can be said about territory and the last week.
Imagine someone spent weeks on their bot for event 1 and event 2.
Then territory dropped and they had commitments this past weekend and work during the day.
Your statement is 100% true on the last week.
Cant say it impacted me much.
Your point is valid though.
From a team perspective it would mean
A) Submissions needs to be reopened (But im sure this is a one liner from a bot submission perspective)
B) They would need to also actually rebuild the top 8 bot pool
Beyond that theres not much added pressure.
**The slides and presentations can be built without even knowing who the top 8 are.
Im probably over paranoid.
As I am every year.
And highly perfectionistic.
I almost didn’t enter this time round due to time-constraints (since I’ve started studying again, in addition to a full-time day job). I did manage to submit something for the final round though - it’s not pretty, but I have my fingers crossed.
For the most part, I try to keep my villagers in this state.
Eish that building bug is a bummer 
Hopefully it’s not road. Once I had my bot submitted I refused to do any more runs/tests - if there’s a bug in my bot I don’t want to know about it
With @rfnel on this one, I also think it’s better to keep the bots final after the 3rd phase. With only ~4 weeks until Comic Con, if we assume the EC team has the results out after 1 week (last time it took 2 IIRC), and the finalists have 1 week for adjustments it leaves only 2 weeks for the EC team to again validate the matches + finalize everything for the finals where they left 5 weeks for for prep originally.
Strange, my bot also ran into the same bronze-age collapse (albeit a bit later than the year 1500AD). Maybe the EC team was trying to make a statement about deforestation leading to widespread collapse of civilization?
@EC Team - I’m not sure if you guys have posted anything in this regard yet, but can you confirm whether or not bots are final for the finals? If they are, I assume its safe to spill state secrets?
Seems like there is a general consensus on this part of management:
I felt a bit like an arsonist coding this bot 
Maybe the AI that ultimately ends the world is one of our bots that grows self-conscience and decides to start replicating its strategy on the environment…
If that happens then the Universe as we know it is doomed.
Want to know how we harvest Suns? That is how we harvest suns
I tried implementing AI in a python bot which I didnt have much joy with due to the delayed ticks, went back to JS for the finals, doubt my bot is taking over the world
Yeah I also switched from Python - happy I did (got to learn a new language), but since Python bots can easily choke up with large state updates something needs to be done for next year. Did you end up testing the wsaccell package thing I posted earlier in the year to see if it helped?
Also switched from Python, used Java for my final bot (after some effort updating all the DTOs to the latest version…)
I have this mental image of my bot where there are vast landscapes of desolate tree stumps, left over from the frenzy of lumberjacks cutting them down and towing flowing currents of logs into a single solitary fortress. All the while, flames blaze tall in the sky, illuminating everything within many kilometers of the fortress (and then somehow the flames get captured and stored?)
I tried a few things though I eas actually trying to pick up Python in this year’s challenge, didn’t have eniugh time to tinker around though so went with the safer option
