(Some structures didn't comply with my math, without these tests, the minRange would have been too low or too big.)

Build is green

Updating workspaces.
Build (release)...
Build (debug)...
Running debug tests...
Running cxxtest tests (302 tests)..............................................................................................................................................................................................................................................................................................................OK!

http://jw:8080/job/phabricator/274/ for more details.

Build is green

Updating workspaces.
Build (release)...
Build (debug)...
Running debug tests...
Running cxxtest tests (302 tests)..............................................................................................................................................................................................................................................................................................................OK!

http://jw:8080/job/phabricator/275/ for more details.

In D111#3725, @elexis wrote:

Increase minimum distance of defensive structures shortly beyond its footprint

You mean that currently the min range has in fact no effect, that's it?

In D111#3754, @fatherbushido wrote:

In D111#3725, @elexis wrote:

Increase minimum distance of defensive structures shortly beyond its footprint

You mean that currently the min range has in fact no effect, that's it?

Correct sir, thats why I originally set it to 0 in the prior patch (would leave a misleading tooltip), but why not make it a fun feature

In D111#3731, @elexis wrote:

(Some structures didn't comply with my math, without these tests, the minRange would have been too low or too big.)

I guess you want to let just the room for a normal unit, right?

In D111#3755, @elexis wrote:

In D111#3754, @fatherbushido wrote:

In D111#3725, @elexis wrote:

Increase minimum distance of defensive structures shortly beyond its footprint

You mean that currently the min range has in fact no effect, that's it?

Correct sir, thats why I originally set it to 0 in the prior patch (would leave a misleading tooltip), but why not make it a fun feature

(Thinking to that, with or without the change, the min distance tooltip will be as misleading or not misleading?)
EDIT: a bit less
EDIT: (I agree with both changes, 0 or murderhole, the second seems more interesting)

In D111#3777, @fatherbushido wrote:

In D111#3755, @elexis wrote:

In D111#3754, @fatherbushido wrote:

In D111#3725, @elexis wrote:

Increase minimum distance of defensive structures shortly beyond its footprint

You mean that currently the min range has in fact no effect, that's it?

Correct sir, thats why I originally set it to 0 in the prior patch (would leave a misleading tooltip), but why not make it a fun feature

(Thinking to that, with or without the change, the min distance tooltip will be as misleading or not misleading?)
EDIT: a bit less

Displaying X meters minimum range when all units are attacked would be misleading.
With the proposed patch applied, it will still be slightly unexpected: 25m minimum range when only units at the buildings footprint are attacked. On the other hand the tooltip doesn't lie, players should just know that the arrows are shot from the center (feels wrong to subtracting the footprint).

Build is green

Updating workspaces.
Build (release)...
Build (debug)...
Running debug tests...
Running cxxtest tests (302 tests)..............................................................................................................................................................................................................................................................................................................OK!

http://jw:8080/job/phabricator/276/ for more details.

Displaying X meters minimum range when all units are attacked would be misleading.
With the proposed patch applied, it will still be slightly unexpected: 25m minimum range when only units at the buildings footprint are attacked. On the other hand the tooltip doesn't lie, players should just know that the arrows are shot from the center (feels wrong to subtracting the footprint).

sure we agree.

In D111#3772, @fatherbushido wrote:

In D111#3731, @elexis wrote:

(Some structures didn't comply with my math, without these tests, the minRange would have been too low or too big.)

well I had reviewed the tooltip part with that funny tooltip 2x2 matrix. I will check it again.

For the min dist, it seems you was misleaded by obstruction vs footprint? Moreover I have some checks to do as most of your structure are square.

Build is green

Updating workspaces.
Build (release)...
Build (debug)...
Running debug tests...
Running cxxtest tests (302 tests)..............................................................................................................................................................................................................................................................................................................OK!

http://jw:8080/job/phabricator/279/ for more details.

As we discussed together, there are some issue with square shape. It depends of what you want to do but you could use for example a formula like sqrt( (w/2)^2+(h/2)^2) + something.

In D111#5287, @fatherbushido wrote:

As we discussed together, there are some issue with square shape. It depends of what you want to do but you could use for example a formula like sqrt( (w/2)^2+(h/2)^2) + something.

Yes, your review was posted in the lobby on the day the diff was updated the last time and it encompassed the fact that units can reach the potentially square obstruction shape of the building while the arrows shoot in a circular distance.
Hence we can't have the same distance from the obstruction shape to the min attack range and
have to indeed use minRange = sqrt((width/2)^2 + (height/2)^2) , where width and height are the obstruction size figures extended by the clearance of the units we care about (siege engine size 4 from pathfinder.xml) * 1.5 for example.

Even though every single building was tested manually, a spreadsheet would help making this more consistent.

Also Grugnas and Hannibal noticed that the civic center is also affected and they don't like it. Me neither on first sight, but on second sight I felt it was (a) more consistent and (b) allows players to certainly defeat a player in phase 1 (as the building can be surrounded once all defenders are killed).

Build is green

Updating workspaces.
Build (release)...
Build (debug)...
Running release tests...
Running cxxtest tests (305 tests).................................................................................................................................................................................................................................................................................................................OK!
Running debug tests...
Running cxxtest tests (305 tests).................................................................................................................................................................................................................................................................................................................OK!

http://jw:8080/job/phabricator/618/ for more details.

Forget all what I said before, I didn't really understand.
EDIT: the table I had uploaded was for a minimal min distance
If there is no problem with unitAI or things like that (I don't know, that's the thing to check), my feeling is that it's not misleading to have a 0 minRange displayed. Something misleading could be to have a number like 4.0 with a 20.0x20.0 square. I
IIrc the min range is just used in unitAI and buildingAI.
In buildingAI the distance constraint is given by the RangeManager (it's center to center iirc)
So you want to display in the tooltip that you won't fire if a unit is inside the obstruction square? It seems imo useless and anyway it's not possitble as you will avoid the sides (if you take the inner circle) or you will add 4 lunes if you take the outer circle.

The confusing fact is also that those two min range ised in BuildingAI and UnitAI are not the same (despite the fact that they use different range check range manager vs unit motion):

• the unit one gives a constraint to the distance where the unit should be before attacking
• the building one gives a constraint to the distance at which targets should be before being attacked

(one could say it's the same, but not so much).

• the tower siege and the ship ones should be even more confusing in that case :p (with our current system)

Just elements about that.

@fatherbushido @elexis : Regarding Fatherbushido's comment above: this is exactly why I changed this in D13. Now the range checks are in the obstruction manager and anybody can call them with arbitrary shapes, which would allow re-opening this discussion.

IMO right now we just need a temporary fix or something.

In D111#14729, @fatherbushido wrote:

Forget all what I said before, I didn't really understand.
EDIT: the table I had uploaded was for a minimal min distance
it's not misleading to have a 0 minRange displayed.

It doesn't display 0 meters, so thats not an issue.
But the civic center tooltip currently displays 8 meters minimum range IIRC but the footprint is always bigger, so _that_ is misleading.

So you want to display in the tooltip that you won't fire if a unit is inside the obstruction square?

This is not a tooltip fix, this is a gameplay change. Murder holes tech should not only apply to the tower but also to fortress and potentiall other buildings.

you will add 4 lunes if you take the outer circle.

Doesn't matter, we already have that for murder holes. The idea is that you must research murder holes tech to reach units standing under the footprint.

Destroying CCs in age 1, eh? I think that won't change strategies too much... after all, kill what's outside the CC and the player is dead already. I think the bigger balance change would be that CCs and military colonies no longer can hold a territory by themselves without murder holes, making it more difficult to take an age II expansion or build a CC near the enemy. Maybe with this change Murder Holes should be an age 1 technology (researchable at the sentry tower and the guard tower) instead of an age 3 tech.

Note: the Crenellations guard tower tech tooltip says it installs "crenellations and murder holes" which is false.

• Currently a min distance which is not an actual min distance (as it's always checked) is displayed. So we don't need it or we need to make it real. I would be for the 1st thing for the moment. So setting 0 for cc.
• Making murderhole usefull for fortress is an option. So the tech is ok.
• Then for the fortress, using only one min distance is imo sufficient (with some really slight benefits for some fortress). A value above 25 will be good (and ok for all fortress, taking into account their obstruction squares). Using the same value will make maintenance easier. EDIT: a value above 22 will be good, something like 25-30 sounds nice.
• For the roman camp, see comment, your choice.
• I checked wall towers, the min range 12.0 is ok (checked cart things, ...)

I can t care of that