ERROR: too many cell->vertex links in pathfinding database!

[zacharias]

Anyone had this..?

All of a sudden Drom does not want to finish the pathfinding. Pretty weird really as what i've added since i last did pathfinding isn't that complex.

Here's some monolog:
=================
Setup 2309 brush level (3611 brushes total)...
Inserted 2308 brushes.
RAW cell count: 35458
WARNING: Memory allocation suspiciously high (67126140 bytes)!
WARNING: Memory allocation suspiciously high (67121224 bytes)!
WARNING: Memory allocation suspiciously high (67116832 bytes)!
WARNING: Memory allocation suspiciously high (71357404 bytes)!
WARNING: Memory allocation suspiciously high (75566092 bytes)!
WARNING: Memory allocation suspiciously high (88255924 bytes)!
WARNING: Memory allocation suspiciously high (100664500 bytes)!
WARNING: Memory allocation suspiciously high (113246316 bytes)!
WARNING: Memory allocation suspiciously high (125835532 bytes)!
40 extra bsp planes created
18860 cells, 127844 portals, 47465 polygons; 490 splits from large polys
20972 polys merged; 75049 colinear vertices deleted; 17 vertices max
4150 unique planes
Bytes used for internal portal structures: 42811976 bytes.
Polyhedra 285880 Polys 6318592 Portal PolyEdges 14567696 PolyEdges 21639808
Portalized...portal_cells_intersecting_sphere: too many cells @ (-36. -124. -12.)
portal_cells_intersecting_sphere: too many cells @ (-38. -92. -12.)
portal_cells_intersecting_sphere: too many cells @ (94. -68. -12.)
portal_cells_intersecting_sphere: too many cells @ (68. -32. -12.)
portal_cells_intersecting_sphere: too many cells @ (114. -68. -12.)
portal_cells_intersecting_sphere: too many cells @ (152. -56. -12.)
portal_cells_intersecting_sphere: too many cells @ (172. -228. 4.)
portal_cells_intersecting_sphere: too many cells @ (172. -208. 4.)
portal_cells_intersecting_sphere: too many cells @ (172. -248. 4.)
portal_cells_intersecting_sphere: too many cells @ (142. -260. 2.)
portal_cells_intersecting_sphere: too many cells @ (96. -136. -12.)
portal_cells_intersecting_sphere: too many cells @ (234. 22. -12.)
portal_cells_intersecting_sphere: too many cells @ (116. -38. -12.)
portal_cells_intersecting_sphere: too many cells @ (264. 100. -14.)
Computing optimized BSP... this may take a while.
Bytes used for internal portal structures: 3311040 bytes.
Polyhedra 18304 Polys 464000 Portal PolyEdges 942912 PolyEdges 1885824

Building rooms...
136 rooms, 426 portals, 3.1323529 average portals per room
Generating cell-room lists...
0% completed
2% completed
5% completed
8% .....
...etc...etc.
..97% completed
16946 cells, 41.183824 average cells per room
Calculating path find cells..........................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................ERROR: too many cell->vertex links in pathfinding database!
Cannot enforce quality!
=====================

Anyone know if this stems from the "portal_cells_intersecting_sphere: too many cells" error? (Or is that not a critical error?)
Any help would be appreciated..

[Caradavin1]

I think that the problem lies with the error messages:

Portalized...portal_cells_intersecting_sphere: too many cells @
(-36. -124. -12.)

And I remember someone mentioned elsewhere in the forum that the three numbers (E.g. 68. -32. -12.) may be coordinates.

Are you stacking solids on solids to create things? I used to do this when I first started out, and it seemed to cause problems. When I learned that I could use air brushes to carve out of the solid, I did so. It got rid of most of my problems during pathfinding and portalizing, and it speeded my pathfinding up.

[zacharias]

Sorry for the lateness of this reply..i was locked out of the forums for the last couple of days since i posted this.

Yes, those numbers are co-ords.

Well, i've solved my problem for the moment by doing the old trick of 'area brush where ai's go, optimize, compute pathfinding, then 'show all' and re-optimize, recalculate rooms etc.'.
It turns out that i'm still getting a slew of 'Portalized...portal_cells_intersecting_sphere: too many cells @
(-36. -124. -12.) error messages even when i do this, so i guess that, while my problem *is* somehow related to cell complexity, this error message alone isn't a fatal error in terms of pathfinding..btw it comes up as part of optimisation, so it may not be related to pathfinding at all(?)
The too many cell->vertex links is the fatal error message.

[amievil?]

i had the same error. to fix, i optimized first and then ran the database. no more errors for now :) what i added after the error wasnt complex either, it has to be with the number of pathable cells or something. thats why optimzing worked for me to help cut down on the cells. if you havent tried it, optimze first and then run the database. may work, may not who knows.

[zacharias]

Thanks..but i was optimizing and still getting the errror. (My mission won't portalize (the whole thing) anymore, it crashes.)