[http://sourceforge.net/p/ufoai/bugs/2976 Item 2976] imported from sourceforge.net tracker on 2013-01-28 19:54:51

The new RMA2 algorithm seems to be too greedy on using a tile.

Here is a few screenshots and tilesets:
01: -

02: CM_LoadMap: "-tropic/tr_ +h01 +h02 +h03 +h04 +craft_drop_firebird +craft_crash_scout +p02 +p02 +p02 +p02 +p02 +p02 +p02 +p02 +f07 +f07 +f07 +f04 +f04 +f04 +f04 +f01" "8 16 0 0 -8 0 8 -24 0 -8 -8 0 -16 -24 0 -16 8 0 -24 -24 0 -24 -16 0 16 -8 0 16 0 0 -24 8 0 -24 16 0 -8 0 0 8 8 0 16 8 0 -24 24 0 -16 24 0 -8 24 0 0 24 0 8 24 0 16 24 0 -24 -8 0"

03: CM_LoadMap: "-ufocrash/uc_ +craft_drop_firebird +craft_crash_scout +h02 +h02 +h02 +h02 +h02 +h02 +h02 +d03 +h02 +h02 +d03 +h06 +h05" "-8 0 0 -24 -8 0 -24 -16 0 -16 -16 0 -24 16 0 -16 16 0 -8 16 0 0 16 0 8 16 0 16 8 0 16 0 0 24 0 0 -8 -16 0 8 -16 0 24 -16 0"

04: CM_LoadMap: "-ufocrash/uc_ +craft_drop_firebird +craft_crash_scout +h02 +h02 +h02 +h02 +h02 +h02 +h02 +d03 +h02 +h02 +d03 +h06 +h05" "-8 0 0 -24 -8 0 -24 -16 0 -16 -16 0 -24 16 0 -16 16 0 -8 16 0 0 16 0 8 16 0 16 8 0 16 0 0 24 0 0 -8 -16 0 8 -16 0 24 -16 0"
Rerouted for RMA in 1.0s
checksum for the map '+ufocrash': 3339541002
ufo script checksum 2879660949

-geever
===== Comments Ported from Sourceforge =====

====== geever (2011-08-14 22:11:41) ======

example map 01
====== geever (2011-08-14 22:12:38) ======

example map 02
====== geever (2011-08-14 22:13:32) ======

example map 03
====== geever (2011-08-14 22:14:29) ======

example map 04
====== geever (2011-08-14 22:22:25) ======

05: CM_LoadMap: "-ufocrash/uc_ +craft_drop_firebird +craft_crash_scout +h05 +h05 +h02 +h02 +h02 +h02 +h02 +h01 +h02 +h02 +h02 +h02 +h06 +h06" "-16 -8 0 8 0 0 24 8 0 24 -8 0 -24 -8 0 -24 -16 0 -16 -16 0 -8 -16 0 0 -16 0 16 -16 0 8 -16 0 8 -8 0 16 -8 0 -24 0 0 -8 8 0 -24 8 0"

and I could add thousands more...

-geever
====== geever (2011-08-14 22:22:52) ======

example map 05
====== aduke1 (2011-08-16 19:55:08) ======

Accepted.
I know what happens and what to do about it.
But the solution would require a set of new seedlists :(
So let me work some more on getting rid of the seedlists first.
====== aduke1 (2011-09-08 20:52:27) ======

Now that we have gotten rid of the seedlists, I will take care of this as promised.

Can you tell me the *best* map+assembly to produce this effect (for testing purpose) ?
====== geever (2011-10-12 20:01:49) ======

savegame
====== geever (2011-10-12 20:03:23) ======

I've uploaded a savegame with a mission, that produced similar maps like:

CM_LoadMap: "-ufocrash/uc_ +craft_drop_firebird +craft_crash_scout +h02 +h01 +h01 +h01 +h05 +h02 +h05 +h02 +h02 +h02 +h01 +h01 +h02 +h01 +h02" "8 0 0 -8 -8 0 24 16 0 8 16 0 -8 16 0 -16 -16 0 -24 -16 0 0 -16 0 -16 -8 0 -24 0 0 -24 8 0 -16 8 0 -24 16 0 8 -16 0 24 -16 0 8 -8 0 24 -8 0"


-geever
====== aduke1 (2011-10-12 20:25:45) ======

Thx. This new mapstring perfectly shows the fix.
The problem is about equivalent tiles, that is, same size, shape and neighbours.

In the mapstring of the original post you can see that RMA2 used up the maximum allowed # of a tile before placing the first of the next equivalent.
Your latest mapsting shows that this is no longer the case.

Admittably the map still looks boring.
==> it's not RMA2 that is boring, it's the assembly.
====== aduke1 (2011-11-04 21:47:06) ======

tlh2000 fixed it partially by reducing the allowed # for h01-3 from 10 to 3.
But the map still can look somewhat boring because two of those houses are merely a mirror of each other.
Maybe a mapper could recycle some more houses from other maps ?

However, I do no longer consider this a bug.
====== aduke1 (2012-09-21 21:59:31.908000) ======

- **assigned_to**: Duke --> Holger 'ShipIt' Gellrich

====== aduke1 (2012-09-21 22:02:52.241000) ======

ShipIt, I bet you can produce some tiles that make ufocrash map a little less boring ;)
====== shipit (2012-09-30 08:17:23.741000) ======

The ufocrash map is kind of a backup plan, just in case there is no other map useable for a crashsite. The goal is to make it obsolete, not expand it.

Also, after the mapselection is fixed now, it should not appear that often anymore. At least not for the smaller UFO´s.

====== aduke1 (2012-12-08 22:17:52.299000) ======

Nate, afaik you have been playing a campaign recently.
Can you confirm that ufocrash doesn't appear that often anymore ?
====== natewr (2012-12-08 23:24:37.682000) ======

For small UFOs it rarely appears for me now. Maybe be different for large UFOs but to be honest hte most common maps are those without UFOs, so it won't really matter. Either way, I will close this because I think the fundamental issue has been fixed. If the ufo_crash map appears too much its from a lack of maps, not a coding issue.