Some Thoughts About 9 Villager Maps

[T-West]

This post is intended mostly for fellow map scripting nerds, but would be great to get feedback from everyone.

The feedback I've seen for the 9-Villager start has been mostly positive, so I expect we'll be seeing more of it in tournaments in the future. We've now had 3 tournaments---DWL, Wallhalla, and Warlords---use slightly different variations on the start. I think there's still more tinkering that can be done to improve it, and I've read lots of interesting suggestions during the tournament. So let's discuss some of the things that are and are not possible currently in map scripts, and some of the ways the 9 Villager start might be improved in the future.

XS Scripts
It wouldn't be a post about the Definitive Edition if it didn't start with a bug report! In addition to random map scripts, DE has "xs scripts," which allow map scripters to use additional features in their maps. For example, we can use xs scripts to adjust resources individually for each civ, and could remove the -30 Wood penalty from the Huns, since they don't need the starting Houses.

The problem is that these scripts don't work for spectators. You know how a custom random map transfers when you enter a lobby? The xs script needs to transfer too. And right now they don't transfer properly. There are workarounds, such as manually copy and pasting files within the temporary folder or using additional mods for changes, but I wouldn't want to rely upon them for a large tournament. Having one single mod with maps for a tournament is all that should be necessary.

So there are adjustments we might want to do, and potentially could do in the future, that were not possible for Warlords. If 9 Villager starts become more common, tournament hosts will need to push the development studio to prioritize fixing the bug. It's a pretty big limitation to have such a powerful feature unusable due to this file transfer error.

Anyway, on to design questions about how to adjust the 9-Villager start.

Sheep
Standard games give players 4+2+2 Sheep, that is, 4 Sheep near the TC, and two other groups of 2 Sheep. DWL adjusted this to 4+2, 4 Sheep near the TC and another group of 2 Sheep. For Warlords we stuck with the standard 4+2+2. I wanted to avoid "nerfing" civs as much as possible. The Britons, Tatars, Gurjaras, and even Mayans would get a bit of a penalty if they were "missing" 2 of their Sheep.

Perhaps these civs could have some kind of adjustment made (extra food, or be given an extra sheep) to make up for this? I think a 2+2+2 pattern, with 2 Sheep under the TC and two groups of 2 Sheep at the normal distance so you still have to scout for them, would be the right way to do the start. But I'm not sure whether civs with Sheep bonuses should be adjusted, or what exactly those adjustments should be.

Starting Houses
The starting Houses can block off resources or just generally be rather annoying. And on some maps (e.g. Border Dispute, Yucatan, or Migration) the starting locations around the TCs are already very cramped, with the Houses just making this worse.

One suggestion I liked was to give the starting TC an additional 10 population space (this would take an xs script). This would be just the starting TC, with other TCs giving their normal 5. That way we wouldn't need to bother placing the Houses. Perhaps it makes the start lose too much "character" though, if the Houses aren't there.

This could affect the douche if a TC is deleted and then rebuilt with less pop space. But, well, I don't think that's too crucial of a concern.

Scouting
I've seen a concern that the loss of 2:30 at the start of the game gives too little time to scout. Of course in a normal game players will have scouted the area near their starting TC by then. Here are a few of the suggestions I've seen for addressing this:

• Give a larger revealed area around the starting TC.
• Use the same Line of Sight in the Dark Age for the Scout as in the Feudal Age (that's +2, and similarly Eagle Scouts would get their Tracking benefit from the start
• Start players with 2 Scout
• Increase Scout speed in the Dark Age.

Personally I think either a larger LoS around the TC or more LoS for the Scout would be reasonable. More TC LoS makes sense for Arabia, but for some more custom maps like Border Dispute or Socotra, it can be awkward to have a large explored area that overlaps somewhat with your opponent. It also removes some of the really early decision making about what to do with the Scout. Going instead for the increased Scout LoS means that players still need to Scout their own base, and still have a tradeoff between scouting immediately and forgoing scouting to go forward and try to lame (like on Socotra).

On the other hand, the lack of Scouting makes it riskier to go for a Deer push (where it seems Deer pushing may be more common now than it was in the past), making players choose between the Deer and Scouting more than in normal RM. Some people might actually prefer that.

What are people's thoughts here?

Straggler Trees
One small thing I noticed was that some other maps just added an additional straggler tree for a total of 6 around the TC. I kept this to a total of 5 for the Warlords maps, and also ensured the tree does not spawn adjacent to another tree.

The side of the TC the straggler spawns on can make a huge difference on some maps. It could lead to easier Boars on Border Dispute, or faster walls on BF. Usually this isn't an important detail, but on some maps it's important to control which side of the TC the straggler spawns on.

Nomad
We used a 9 Villager start for Nomad, while giving an additional 50 Wood so players could construct two Houses right away. Personally I feel like Nomad really benefits from the additional scouting and quicker start you get with all of the Villagers.

There was a suggestion that, instead of 9 Villagers, Nomad starts with 6 Villagers and 3 Fishing Ships. For the current DE version of Nomad with water on all 4 sides, that seems like a reasonable way to start as well, with the 3 Fishing Ships spread out around the outside. Would be interested in hearing thoughts about that.

Individual Civilization Adjustments
Assuming that the xs transfer bug is fixed and we can adjust resources individually for each civ, what adjustments should we make? The Chinese are the most obvious civ to address, but we could open up a can of worms with changing bonuses. Scrolling through the tech tree real quick, here are the changes I've come up with:

• Chinese Food and Villagers: DWL subtracted 100 Food from all civs, but this has the effect that the Chinese start with negative 100 Food. I think it's too gimmicky for a civ to have negative Food. With xs scripting I think starting with 0 Food and 11 Villagers might be the right compromise to make here, and most other civs could have 100 Food.
• Loom for Chinese and Mayans: Should these civs start with these techs already researched (and -50 Gold)? It doesn't matter much for the Chinese (who still won't have Food), but since the Mayans still get two Houses they don't start Housed. Mayans could start with 9 Villagers instead of 10, but get Loom instead of the Vil.
• Huns Wood: Huns should start with 100 Wood like normal.
• Malians: Malians save Wood on their Houses, they could start with extra 8 wood to make up for this. - Hindustanis: Should they receive the extra 30 Food they would have saved on their first 6 Villagers?
• Britons, Tatars, Gurjaras, Mayans: What compensation should they get for missing 2 Sheep (if the map has 6 Sheep in total)?
• Should Gurjaras even have 6 Villagers around a starting Sheep, or some different way to collect Food at the start so they can keep the Sheep in their Mill?
• Should the Mongols (or their team in a TG) get a bonus for the "extra" scouting time they missed?

There has been lots of discussion about how various civ bonuses should work at the start of Empire Wars.
The Chinese are clearly the most important to address.
But otherwise, I don't think these bonuses are too important here, but it'd still be nice to get them working consistently.

 

[Russeljones]

9 vils is the way to go.

 

[RobChang]

This post is intended mostly for fellow map scripting nerds, but would be great to get feedback from everyone.

The feedback I've seen for the 9-Villager start has been mostly positive, so I expect we'll be seeing more of it in tournaments in the future. We've now had 3 tournaments---DWL, Wallhalla, and Warlords---use slightly different variations on the start. I think there's still more tinkering that can be done to improve it, and I've read lots of interesting suggestions during the tournament. So let's discuss some of the things that are and are not possible currently in map scripts, and some of the ways the 9 Villager start might be improved in the future.

XS Scripts
It wouldn't be a post about the Definitive Edition if it didn't start with a bug report! In addition to random map scripts, DE has "xs scripts," which allow map scripters to use additional features in their maps. For example, we can use xs scripts to adjust resources individually for each civ, and could remove the -30 Wood penalty from the Huns, since they don't need the starting Houses.

The problem is that these scripts don't work for spectators. You know how a custom random map transfers when you enter a lobby? The xs script needs to transfer too. And right now they don't transfer properly. There are workarounds, such as manually copy and pasting files within the temporary folder or using additional mods for changes, but I wouldn't want to rely upon them for a large tournament. Having one single mod with maps for a tournament is all that should be necessary.

So there are adjustments we might want to do, and potentially could do in the future, that were not possible for Warlords. If 9 Villager starts become more common, tournament hosts will need to push the development studio to prioritize fixing the bug. It's a pretty big limitation to have such a powerful feature unusable due to this file transfer error.

Anyway, on to design questions about how to adjust the 9-Villager start.

Sheep
Standard games give players 4+2+2 Sheep, that is, 4 Sheep near the TC, and two other groups of 2 Sheep. DWL adjusted this to 4+2, 4 Sheep near the TC and another group of 2 Sheep. For Warlords we stuck with the standard 4+2+2. I wanted to avoid "nerfing" civs as much as possible. The Britons, Tatars, Gurjaras, and even Mayans would get a bit of a penalty if they were "missing" 2 of their Sheep.

Perhaps these civs could have some kind of adjustment made (extra food, or be given an extra sheep) to make up for this? I think a 2+2+2 pattern, with 2 Sheep under the TC and two groups of 2 Sheep at the normal distance so you still have to scout for them, would be the right way to do the start. But I'm not sure whether civs with Sheep bonuses should be adjusted, or what exactly those adjustments should be.

Starting Houses
The starting Houses can block off resources or just generally be rather annoying. And on some maps (e.g. Border Dispute, Yucatan, or Migration) the starting locations around the TCs are already very cramped, with the Houses just making this worse.

One suggestion I liked was to give the starting TC an additional 10 population space (this would take an xs script). This would be just the starting TC, with other TCs giving their normal 5. That way we wouldn't need to bother placing the Houses. Perhaps it makes the start lose too much "character" though, if the Houses aren't there.

This could affect the douche if a TC is deleted and then rebuilt with less pop space. But, well, I don't think that's too crucial of a concern.

Scouting
I've seen a concern that the loss of 2:30 at the start of the game gives too little time to scout. Of course in a normal game players will have scouted the area near their starting TC by then. Here are a few of the suggestions I've seen for addressing this:

• Give a larger revealed area around the starting TC.
• Use the same Line of Sight in the Dark Age for the Scout as in the Feudal Age (that's +2, and similarly Eagle Scouts would get their Tracking benefit from the start
• Start players with 2 Scout
• Increase Scout speed in the Dark Age.

Personally I think either a larger LoS around the TC or more LoS for the Scout would be reasonable. More TC LoS makes sense for Arabia, but for some more custom maps like Border Dispute or Socotra, it can be awkward to have a large explored area that overlaps somewhat with your opponent. It also removes some of the really early decision making about what to do with the Scout. Going instead for the increased Scout LoS means that players still need to Scout their own base, and still have a tradeoff between scouting immediately and forgoing scouting to go forward and try to lame (like on Socotra).

On the other hand, the lack of Scouting makes it riskier to go for a Deer push (where it seems Deer pushing may be more common now than it was in the past), making players choose between the Deer and Scouting more than in normal RM. Some people might actually prefer that.

What are people's thoughts here?

Straggler Trees
One small thing I noticed was that some other maps just added an additional straggler tree for a total of 6 around the TC. I kept this to a total of 5 for the Warlords maps, and also ensured the tree does not spawn adjacent to another tree.

The side of the TC the straggler spawns on can make a huge difference on some maps. It could lead to easier Boars on Border Dispute, or faster walls on BF. Usually this isn't an important detail, but on some maps it's important to control which side of the TC the straggler spawns on.

Nomad
We used a 9 Villager start for Nomad, while giving an additional 50 Wood so players could construct two Houses right away. Personally I feel like Nomad really benefits from the additional scouting and quicker start you get with all of the Villagers.

There was a suggestion that, instead of 9 Villagers, Nomad starts with 6 Villagers and 3 Fishing Ships. For the current DE version of Nomad with water on all 4 sides, that seems like a reasonable way to start as well, with the 3 Fishing Ships spread out around the outside. Would be interested in hearing thoughts about that.

Individual Civilization Adjustments
Assuming that the xs transfer bug is fixed and we can adjust resources individually for each civ, what adjustments should we make? The Chinese are the most obvious civ to address, but we could open up a can of worms with changing bonuses. Scrolling through the tech tree real quick, here are the changes I've come up with:

• Chinese Food and Villagers: DWL subtracted 100 Food from all civs, but this has the effect that the Chinese start with negative 100 Food. I think it's too gimmicky for a civ to have negative Food. With xs scripting I think starting with 0 Food and 11 Villagers might be the right compromise to make here, and most other civs could have 100 Food.
• Loom for Chinese and Mayans: Should these civs start with these techs already researched (and -50 Gold)? It doesn't matter much for the Chinese (who still won't have Food), but since the Mayans still get two Houses they don't start Housed. Mayans could start with 9 Villagers instead of 10, but get Loom instead of the Vil.
• Huns Wood: Huns should start with 100 Wood like normal.
• Malians: Malians save Wood on their Houses, they could start with extra 8 wood to make up for this. - Hindustanis: Should they receive the extra 30 Food they would have saved on their first 6 Villagers?
• Britons, Tatars, Gurjaras, Mayans: What compensation should they get for missing 2 Sheep (if the map has 6 Sheep in total)?
• Should Gurjaras even have 6 Villagers around a starting Sheep, or some different way to collect Food at the start so they can keep the Sheep in their Mill?
• Should the Mongols (or their team in a TG) get a bonus for the "extra" scouting time they missed?

There has been lots of discussion about how various civ bonuses should work at the start of Empire Wars.
The Chinese are clearly the most important to address.
But otherwise, I don't think these bonuses are too important here, but it'd still be nice to get them working consistently.

Some things sound familiar here? How about 2+3×2 sheep in the outer sphere if the base?

How about having tcs only have chinese los and maybe scouts get +1 los?

To combine two my personal hobbies, how about reducing stone start to 150 freeing up some stone for outposts to be made at the start so scouting the own base gets pretty fast done.

 

[Rorrik]

Another suggestion for the scouting could be to use map revealers as part of the map script to try to approximate the first couple minutes of scouting: loop around the TC, near one of the wood lines and the berries, possibly a boar.
The additional scouting on water from having fishing ships on Nomad that need to return to where the dock is being built would be interesting. It would make water aggression more effective in general, since you are more likely to know where the opponent's dock is without giving up fishing time to go looking for it.

 

[T-West]

Another suggestion for the scouting could be to use map revealers as part of the map script to try to approximate the first couple minutes of scouting: loop around the TC, near one of the wood lines and the berries, possibly a boar.
The additional scouting on water from having fishing ships on Nomad that need to return to where the dock is being built would be interesting. It would make water aggression more effective in general, since you are more likely to know where the opponent's dock is without giving up fishing time to go looking for it.

That's another idea for scouting. Instead of just a big circle of vision around the TC, you could very specifically reveal woodlines, Boars, or other important objects. It might be awkward to have a system that would vary by map if there are different woodlines or resources though.

 

[LowEloNobody]

Re: Scouting
Increased TC line of sight seems to be the easiest solution IMO. More scout LoS/Speed/More scouts also affects other parts of the game (2 scouts is a massive difference for killing vils etc, more speed changes laming / being able to chase down villagers, more LoS gives you the ability to scout opponents buildings behind walls)

 

[LmScar12]

I don't think we should necessarily be hyper-focused on making the maps have the same play or same balance as a normal start where we just skipped the first 2.5 minutes. That's already out the window because of the flexibility that 9 vills allows you on your build. You can go anywhere from 3 all the way to all 9 of your initial vills on food and maintain villager production. If you do 6 on food, 3 on wood intially, you will always be able to go up at least 1 pop earlier because you don't have those periods of negative and break-even food accumulation at the start of the game. Minor things like giving the Malians +8 wood can be ignored.

 

[T-West]

Re: Scouting
Increased TC line of sight seems to be the easiest solution IMO. More scout LoS/Speed/More scouts also affects other parts of the game (2 scouts is a massive difference for killing vils etc, more speed changes laming / being able to chase down villagers, more LoS gives you the ability to scout opponents buildings behind walls)

I feel like more Scout LoS, only restricted to the Dark Age, is a fairly minor change. More speed or attack would be a really big change which I don't think would be a good idea (I've seen it suggested though).

 

[Lokalo]

I feel like more Scout LoS, only restricted to the Dark Age, is a fairly minor change. More speed or attack would be a really big change which I don't think would be a good idea (I've seen it suggested though).

I personally after trying both warlords and DWL, think DWL explored area is good. It often takes me quite a bit to find sheeps still, but at least I don't have to play blind for first minute and search where to build lumber camp. And if there are few maps where that is bad, maybe those few maps should be changed instead. At least for me it seems easiest and making most sense solution. Although I have to admit I did play only arabia. But I think even though I like 9 vill start I felt big difference between warlords and dwl start, so you can be sure many people especially in lower level would be unhappy about having to look where to place their lumber camp. Every other option like increasing scout speed will lead into some aspect of the game changed like making to block vilalgers with drush easier etc.

 

[JayGould]

Yeah I think just increasing the "autoscouted" part at the start would be best. Maybe an additional 2-3 tiles or so. Just enough to see the edge of your nearest woodline(s) on Arabia (and your whole walled section on Arena).

 

[Festivus]

What about adding a horse (as opposed to a second scout) or something like that to help the scouting situation. Its a lot of fun for the viewer to see how much some of the high level players can do with 2 starting scouting options. If it hasn't been experimented with yet, I think it'd be a fun way to help the scouting issue in a more entertaining way than just increasing LOS, but without giving 2 scouts and the offensive nature that comes with that.

 

[TheMadCADer]

If you don't like the house, you can just give the pop space for free, similar to how Huns bonus works, just with 5 pop space instead of 200 pop space. No xs required.

Not so much of a suggestion, but I'm curious. Assuming xs scripts were working properly, what was your plan for changing the starting vills for Chinese? It's easy enough to disable the spawning of extra villagers, but to change extra vills from 3 to 2 sounds tricky.

 

[T-West]

If you don't like the house, you can just give the pop space for free, similar to how Huns bonus works, just with 5 pop space instead of 200 pop space. No xs required.

Yeah, I had considered that also. I think the only practical difference would be in the "instant douche" situation like DauT did, where you'd still have population but zero buildings. I'd prefer to avoid that if possible, but otherwise that's also a good solution.

Not so much of a suggestion, but I'm curious. Assuming xs scripts were working properly, what was your plan for changing the starting vills for Chinese? It's easy enough to disable the spawning of extra villagers, but to change extra vills from 3 to 2 sounds tricky.

That's a good point, even if we can check the civ is the Chinese I don't think the "amount" field of the tech is editable within the xs script currently. That might take some tricks to get working too. Or maybe we could get the number of Villagers refactored to a player attribute/resource so it can be adjusted in a future patch. I still want to collect ideas that could be done by pushing for some other slight changes too (like how random number generation had been added to xs scripts).

What the xs script can do is subtract 100 Food from all civs except for the Chinese (to prevent negative food). But that by itself just buffs Chinese even more relative to the other civs.

 

[RobChang]

Yeah I think just increasing the "autoscouted" part at the start would be best. Maybe an additional 2-3 tiles or so. Just enough to see the edge of your nearest woodline(s) on Arabia (and your whole walled section on Arena).

Yeah one should know the nearest woodline and have sheep under tc to have working vills. Maybe starting scout can start at a woodline and then you automatically see it

 

[paradox303]

I actually ran two test brackets for Nomad.

6 vills + 3 fish and 9 vills + 3 fish.

Two 8 player brackets. So not a tonne of games but a small and decent sample size to view and gather some feedback.

9 vills + fish, too much vision and just an FC fest. The 6 + 3 was nice as there was still a window for Feudal and everyone having a dock and fish meant that the civ bonuses mattered less. Personally, I really like the 6 + 3 start from what I played and viewed.

 

[T-West]

I actually ran two test brackets for Nomad.

6 vills + 3 fish and 9 vills + 3 fish.

Two 8 player brackets. So not a tonne of games but a small and decent sample size to view and gather some feedback.

9 vills + fish, too much vision and just an FC fest. The 6 + 3 was nice as there was still a window for Feudal and everyone having a dock and fish meant that the civ bonuses mattered less. Personally, I really like the 6 + 3 start from what I played and viewed.

There's also the familiarity of 6 Villagers being used for the old Chinese start on Nomad (before the bonus was changed to once their TC is up).

 

[Rissen]

Starting Houses
The starting Houses can block off resources or just generally be rather annoying. And on some maps (e.g. Border Dispute, Yucatan, or Migration) the starting locations around the TCs are already very cramped, with the Houses just making this worse.

Is it possible to pack/unpack yurts like a trebuchet (similar to that Kotyan Khan mission)?

 

[fenrir5]

9 vil start definitely feels amazing and thank you for all the work you put in for making it possible T-West.. Honestly your presence in this community is as important as the pro players! Kudos to you for all your efforts!

On the discussion topics..
1. Villager Placement:
Every civ starts with 6 vils around one sheep which in a way makes the chinese start quite easy. I wonder if this is a good thing or needs further thinking. This vil placement is why I believe 9 vil starts can actually be better for all rank levels.

2. Nomad/Land Nomad/ Bedouins/African Clearing/Steppe & Other future Nomad start maps:
I think just looking at Nomad might not be fair to the other nomad maps. Yes the 6 vils + 3 fish sounds good but resource wise, thats 225 wood free. This might make civs like Japanese and Dravidians a bit too strong (not against it tbh :D) on nomad and can even pave the way for different strats like start with a dock and lumbercamp. On the other hand, if we think about it for ranked, Chinese and Huns could struggle a lot. My suggestion for this is to have 6 + 1 or 2 fish. The 9 vil start in nomad however is not for all rank levels. Its a real struggle to gather all the vils and scout well with them so as a norm for all nomad type maps(maybe exception Steppe), I'd say 6 vils be the limit.

3. Scouting
The extra LOS for the TC sounds a bit too OP for civ like Chinese who could end up being able to see the entire base. I'd say scouts having same LOS as in feudal feel like a better option. Or just straight up just give +1 to scouts. There should always be a compromise between deer luring and scouting. Feudal LOS makes it a bit too easy.

4. Other thoughts/ Balance Changes:
Given there is the regular RM, EW and if this becomes a standard, there needs to be a lot of considerations for balances. Some civs will be stronger because of the shorter up times.

• Mongols might not have as much of an advantage.
• Chinese could become quite OP.
• Gurjaras sheep bonus will be affected(Yay!).
• Hindustanis wont save as much going up to feudal(also Yay!).
• Bengalis could benefit from faster uptime to get their extra vils.
• Mayans as you discussed might need a rethink
• Vikings will get wheelbarrow faster
• Burgundians might not be able to research feudal techs in dark age.

And none of these are necessarily overpowering but may need to be considered.
In all honesty, I'm all for the new RM gameplay with more vils! Dark age to feudal transition is something that can be learnt easily but hard to master. This change will make the entry point easier while still keeping the skill ceiling. Lets go Microsoft! Do it well and do it right!

 

[T-West]

Is it possible to pack/unpack yurts like a trebuchet (similar to that Kotyan Khan mission)?

Oh, so players could move them if they spawn blocking resources? That's an interesting idea, I'll look into it.

 

[T-West]

Assuming xs scripts were working properly, what was your plan for changing the starting vills for Chinese?

Figured out a way to do this. Place a flag under the TC in the rms file, and in the xs script do the following:

Set the Flag to have 0 HP
if the player's civ is not the Chinese:
    set the Flag's dead unit ID to a Villager
(leave the dead unit ID unset if the player is the Chinese)

The flag dies immediately and "disappears" for the Chinese but becomes a Vil for all other civs. I'll post a MWE in the RMS Discord.

Thanks to Alian for help coming up with this.

 

[Huehuecoyotl22]

Is it possible to pack/unpack yurts like a trebuchet (similar to that Kotyan Khan mission)?

Oh, so players could move them if they spawn blocking resources? That's an interesting idea, I'll look into it.

I did not play that mission, but wouldn't that lead to the same issue as with packed tcs, that most of the times you don't place them perfectly on the grid and then it leads to <1 tile-unbuildable spaces?
And since houses often are used for walling, there could be some strange holes. If the hole is <0.5 tiles it could even lead to that some units can get through (smaller collision size), while other units can't (bigger collision size).