Auto-translated
Well, by the way, Teamwork is more suitable for mirroring, because the generator places 4 zones on a plane more adequately than 3. And there isn't a strict requirement that "it should be like before."
New here? I'm Roha — want a quick tour?

Mirror image of mt_Teamwork.
1) Mirrored it 1 to 1 (blocked zones 2, 6, and 4 with Guard Towers) in the center, with 2 portals on another plane.
Mirror Teamwork 1 to 1
2) Reduced version (divided the template in half and mirrored it, with 2 portals in the center).
Reduced Mirror Teamwork
I was also thinking of making one for 8 players for a 4v4 game (it might be useful with the appearance of lobbies and Online Hotseat), but the generator still sets it to 4 players. If needed, I'll work on that version too. It will take more time.

Understood. Well, if needed, I'll change what needs to be changed. To make the zones less awkward, I can simply add extra zones without passages. I can also remove the Gold Ore (GO) or set it to 3000-6000 just to prevent players from scouting the opponent's treasure.

For fans of 8mm6, I made a mirrored version of 8mm6!
The template was taken, divided in half, and the left part was mirrored. Paths to the opponent go through the same zones 19 and 12. The cross-shaped portals work.
Mirror mt_Teamwork.
2) Reduced version (divided the template in half and mirrored it; there are also 2 portals in the center).
Reduced Mirror Teamwork
Look at how the mirrored 8mm6 is made in Warlords (the scheme is in the mirrortemplates folder). There are 6 connections between the planes. This particular template is not very suitable for Soda, because the buggy passages will break the structure, but it is perfect for illustrating the principle. And the principle is simple: the transition between planes should not divide the template structure into two parts, unless this is a feature of the structure itself, as in h3dm1, for example. That is, the transition to another plane during the game should occur as early as possible. I think that the same mirrored diamond could have been made with two transitions to another plane from the starting zone, but instead, there is only 1 transition, and for some reason, the guard on it is stronger than on the intra-plane ones. As for the mirrored 8mm6, it would be quite suitable for HotA if the buggy passages were fixed. The fact that it supposedly didn't work out is simply because no one really tried it. I played it a couple of times - in my opinion, it is much more interesting than h3dm1. In terms of size, as already noted, XL mirrored doesn't make much sense, because a large size is a way to achieve balance, which is already present in the mirror by default.

The principle is simple: transitioning between planes shouldn't divide the template structure into two parts, unless it's a feature of the structure itself, like in h3dm1, for example. That is, the transition to another plane during the game should happen as early as possible.
I think the same mirrored Diamond could have been made with two transitions to another plane from the starting zone, but instead, there's only one transition, and for some reason, the guard on it is stronger than the guards within the same plane.
That's why I specifically placed stronger guards at the teleport on the mirrored Daimond, near the spawn point – so that people would primarily rush into the zones of their own plane, and not get used to the idea that they can easily enter the opponent's plane for free in the first week.
I believe that in the future, a software mechanism will be implemented for mirror maps to prevent crossing into the opponent's plane before turn 121. This will fundamentally solve the problem of early contact. Experience has shown that even with a replay of the turn, this can be fatal for the blue player.
Therefore, on the mirror Diamond map, I specifically placed a stronger teleport guard near the spawn point – so that players would primarily focus on exploring their own plane, rather than getting used to the idea that they can freely enter the opponent's plane in the first week.
Well, in this matter, opinions can differ. If we take Jebus and similar templates with exits in the center, then you can connect them however you like. But if we take 8mm, 6lm, 8xm templates, then connecting them results in the template not being exactly the same as the original scheme, but some passages change, and the template becomes based on the original. But on the other hand, connecting allows you to move to other zones with non-repeating objects from this zone. I can do both, and it doesn't make a difference to me. It's a matter of personal preference.
In my opinion, it hardly makes sense to tailor all templates to this principle. Some, like Jebus Cross, h3dm1 - yes.
In the given template example, there is actually no structural change whatsoever. The computer colors are different only for technical reasons. What you call crossing is the only acceptable method of mirroring: mirrored zones should not be connected to each other, and if this is unavoidable, they are connected by two pairs of teleports crosswise, as on Jebus Cross. It was already explained by Vik above why that is.

That's why I create several template variations with connections at once. Everyone chooses the option they consider the best. You can't impose your point of view on others. I don't see the point in placing 2-4 portals in each zone, except for the central treasure rooms (a bunch of portals to the same zones, additional blocks due to their protection, other road junctions, less space for objects, etc.), and crossing connections, for example, on one of the 8x8 maps, can significantly prolong the game due to the increased distance to the opponent in terms of the number of zones, and it turns into a different template. I am in favor of speeding up the gameplay, but if mirroring significantly changes the template scheme, I'd rather call the template by a different name than keep calling it the same. The 8x8 map mirrors perfectly, and crossing connections are not as significant there, especially on similar symmetrical templates with the same number of zones to the opponent and AI.
