| Time |
Nick |
Message |
| 01:17 |
|
Taoki joined #minetest-dev |
| 02:56 |
|
ssieb joined #minetest-dev |
| 03:28 |
|
nepugia joined #minetest-dev |
| 04:14 |
|
fluxflux joined #minetest-dev |
| 04:58 |
|
fluxflux joined #minetest-dev |
| 05:37 |
|
vesper11 joined #minetest-dev |
| 05:38 |
|
YuGiOhJCJ joined #minetest-dev |
| 08:01 |
|
Beton joined #minetest-dev |
| 08:38 |
|
ShadowNinja joined #minetest-dev |
| 09:05 |
|
proller joined #minetest-dev |
| 09:43 |
|
karamel joined #minetest-dev |
| 10:37 |
|
df458 joined #minetest-dev |
| 10:48 |
|
calcul0n joined #minetest-dev |
| 10:48 |
|
proller joined #minetest-dev |
| 11:20 |
|
Qiangong2[m] joined #minetest-dev |
| 11:24 |
|
Fixer joined #minetest-dev |
| 11:43 |
|
kollaps[m] joined #minetest-dev |
| 11:43 |
|
texmex joined #minetest-dev |
| 11:53 |
|
Fixer_ joined #minetest-dev |
| 12:02 |
|
proller joined #minetest-dev |
| 12:09 |
|
absurb joined #minetest-dev |
| 12:54 |
|
df458 joined #minetest-dev |
| 13:29 |
|
fluxflux joined #minetest-dev |
| 14:02 |
|
erlehmann joined #minetest-dev |
| 14:21 |
|
proller joined #minetest-dev |
| 14:32 |
|
erlehmann joined #minetest-dev |
| 14:35 |
|
CBugDCoder joined #minetest-dev |
| 14:43 |
|
proller joined #minetest-dev |
| 14:46 |
|
Taoki joined #minetest-dev |
| 15:06 |
|
erlehmann joined #minetest-dev |
| 15:07 |
|
twoelk joined #minetest-dev |
| 16:22 |
|
behalebabo joined #minetest-dev |
| 16:22 |
|
Unarelith joined #minetest-dev |
| 16:48 |
|
sys4 joined #minetest-dev |
| 18:09 |
|
Wuzzy joined #minetest-dev |
| 18:21 |
|
calcul0n_ joined #minetest-dev |
| 19:00 |
|
erlehmann joined #minetest-dev |
| 19:03 |
|
erlehmann joined #minetest-dev |
| 19:07 |
|
erlehmann joined #minetest-dev |
| 19:27 |
|
proller joined #minetest-dev |
| 19:41 |
Krock |
Wuzzy: https://github.com/minetest/minetest/pull/9339#issuecomment-593357498 |
| 19:41 |
Krock |
that's because they have to be const refs |
| 19:41 |
Krock |
i.e. for (const v3s16 &direction : directions) |
| 19:41 |
Wuzzy |
oh crap |
| 19:42 |
Krock |
s/oh crap/til/ |
| 19:43 |
Wuzzy |
yet i somehow managed to snatch an approval... ? |
| 19:47 |
|
TC01 joined #minetest-dev |
| 19:52 |
|
mizux joined #minetest-dev |
| 19:55 |
Wuzzy |
Krock: ok its actually intentional that direction is changable |
| 19:55 |
Wuzzy |
cuz I do "direction.Y = cost.y_change" later |
| 19:55 |
sfan5 |
pushing http://sprunge.us/ghmpNB?diff in 10 minutes |
| 19:56 |
Krock |
in this case it cannot be const |
| 19:56 |
sfan5 |
hm actually I'm gonna reword my commit message |
| 19:57 |
Krock |
okay, you ninja'd me on that one |
| 19:57 |
sfan5 |
the isGone() change with SAOs really exposed some messy things |
| 19:58 |
|
erlehmann joined #minetest-dev |
| 20:02 |
Krock |
hmmm. If the path finder destination is unreachable it takes 20ms. that's quite a lot |
| 20:02 |
Krock |
but at least it works |
| 20:05 |
Wuzzy |
yeah, this PR will not be the last pathfinder PR... |
| 20:05 |
Wuzzy |
this is just the beginning |
| 20:05 |
Wuzzy |
Krock: did you use A*, A*_noprefetch or Dijktra? |
| 20:06 |
Wuzzy |
Dijkstra is very slow (note I didn't touch it) |
| 20:07 |
Krock |
A*_noprefetch |
| 20:08 |
|
ssieb joined #minetest-dev |
| 20:09 |
Krock |
1s for Dijkstra |
| 20:09 |
Wuzzy |
unsurprisingly |
| 20:11 |
sfan5 |
future idea: create a copy the affected map area, run the pathfinder in a separate thread (or multiple, worker-style) and let the mod retrieve the result when it's done |
| 20:15 |
|
erlehmann joined #minetest-dev |
| 20:17 |
Krock |
async pathfinder requests. like the HTTP API |
| 20:18 |
sfan5 |
Krock: does #9455 look okay to you? |
| 20:18 |
ShadowBot |
https://github.com/minetest/minetest/issues/9455 -- Two fixes by sfan5 |
| 20:18 |
Krock |
Yes. |
| 20:19 |
sfan5 |
thanks |
| 20:21 |
|
erlehmann joined #minetest-dev |
| 20:26 |
Krock |
sfan5: #9454 |
| 20:26 |
ShadowBot |
https://github.com/minetest/minetest/issues/9454 -- minetest.get_connected_players() returns recently disconnected players. |
| 20:26 |
sfan5 |
ah yea |
| 20:26 |
sfan5 |
just saw |
| 20:33 |
|
troller joined #minetest-dev |
| 20:49 |
|
behalebabo joined #minetest-dev |
| 20:57 |
|
twoelk left #minetest-dev |
| 21:11 |
rubenwardy |
async pathfinder is eugh |
| 21:11 |
rubenwardy |
for smaller distances, it would be unnecessary overhead |
| 21:11 |
rubenwardy |
for larger distances, maintaining a graph could be a good idea |
| 21:13 |
rubenwardy |
I forget the name of the algorithms, but there are techniques for maintaining graphs for large distance pathfinding, where the graphs indicate connections and weightings |
| 21:14 |
rubenwardy |
I guess it may not be so appropriate for a game with a lot of map changes and complexity |
| 21:16 |
sfan5 |
I think having some overhead is much better than locking up the server for 100ms because 10 mobs wanted to find some paths |
| 21:16 |
rubenwardy |
ah, HPA* |
| 21:23 |
|
absurb_ joined #minetest-dev |
| 22:04 |
|
erlehmann joined #minetest-dev |
| 22:06 |
|
erlehmann joined #minetest-dev |
| 22:15 |
|
bionicpup64_2519 joined #minetest-dev |
| 22:20 |
|
Borgvader123 joined #minetest-dev |
| 22:54 |
|
erlehmann joined #minetest-dev |
| 23:06 |
|
BorgVader joined #minetest-dev |
| 23:06 |
BorgVader |
hello |
| 23:52 |
|
erlehmann joined #minetest-dev |
