| Time |
Nick |
Message |
| 00:11 |
|
EvergreenTree joined #minetest-dev |
| 00:24 |
|
lumberJ joined #minetest-dev |
| 00:27 |
|
Lymkwi joined #minetest-dev |
| 00:50 |
|
Player-2 joined #minetest-dev |
| 01:11 |
|
Lone-Star joined #minetest-dev |
| 01:16 |
|
EvergreenTree joined #minetest-dev |
| 01:39 |
|
Tmanyo joined #minetest-dev |
| 02:28 |
|
Lymkwi joined #minetest-dev |
| 02:53 |
|
Lymkwi joined #minetest-dev |
| 03:31 |
|
ssieb joined #minetest-dev |
| 04:14 |
|
Foz joined #minetest-dev |
| 05:36 |
|
nerzhul joined #minetest-dev |
| 07:33 |
|
troller joined #minetest-dev |
| 07:56 |
|
longerstaff13-m joined #minetest-dev |
| 08:55 |
|
Lunatrius joined #minetest-dev |
| 09:08 |
|
Gael-de-Sailly joined #minetest-dev |
| 09:35 |
|
Fixer joined #minetest-dev |
| 09:35 |
|
ANAND joined #minetest-dev |
| 09:55 |
|
Beton joined #minetest-dev |
| 10:21 |
|
troller joined #minetest-dev |
| 10:31 |
|
proller__ joined #minetest-dev |
| 10:40 |
|
indiana joined #minetest-dev |
| 11:25 |
|
proller__ joined #minetest-dev |
| 11:31 |
|
proller joined #minetest-dev |
| 11:38 |
|
YuGiOhJCJ joined #minetest-dev |
| 11:48 |
|
proller joined #minetest-dev |
| 11:50 |
|
Beton_ joined #minetest-dev |
| 11:55 |
|
Darcidride joined #minetest-dev |
| 12:02 |
|
Lymkwi joined #minetest-dev |
| 12:12 |
|
Megaf joined #minetest-dev |
| 12:19 |
|
lumberJ joined #minetest-dev |
| 12:40 |
|
nerzhul joined #minetest-dev |
| 13:12 |
|
Beton joined #minetest-dev |
| 13:21 |
|
Beton_ joined #minetest-dev |
| 13:26 |
|
Beton joined #minetest-dev |
| 14:09 |
|
Player-2 joined #minetest-dev |
| 14:09 |
|
prawnsalad joined #minetest-dev |
| 14:11 |
|
ANAND joined #minetest-dev |
| 14:27 |
|
octacian joined #minetest-dev |
| 14:27 |
|
octacian joined #minetest-dev |
| 14:47 |
|
proller joined #minetest-dev |
| 14:59 |
|
Player-2 joined #minetest-dev |
| 15:07 |
|
Lymkwi joined #minetest-dev |
| 15:15 |
|
Wuzzy joined #minetest-dev |
| 15:23 |
|
Fixer joined #minetest-dev |
| 15:25 |
|
ssieb joined #minetest-dev |
| 15:35 |
|
Krock joined #minetest-dev |
| 15:38 |
|
antims joined #minetest-dev |
| 15:48 |
|
Lymkwi joined #minetest-dev |
| 15:52 |
|
Lunatrius` joined #minetest-dev |
| 15:56 |
|
proller joined #minetest-dev |
| 16:12 |
|
EvergreenTree joined #minetest-dev |
| 16:29 |
|
proller joined #minetest-dev |
| 16:31 |
|
Fixer joined #minetest-dev |
| 16:51 |
|
EvergreenTree joined #minetest-dev |
| 17:09 |
|
nerzhul joined #minetest-dev |
| 17:10 |
|
EvergreenTree joined #minetest-dev |
| 17:25 |
|
Lunatrius joined #minetest-dev |
| 17:59 |
|
Gael-de-Sailly joined #minetest-dev |
| 18:07 |
|
Icedream joined #minetest-dev |
| 18:34 |
|
EvergreenTree joined #minetest-dev |
| 18:50 |
|
proller joined #minetest-dev |
| 19:04 |
|
longerstaff13-m joined #minetest-dev |
| 19:23 |
|
paramat joined #minetest-dev |
| 20:02 |
|
ssieb joined #minetest-dev |
| 20:04 |
|
proller joined #minetest-dev |
| 20:13 |
|
EvergreenTree joined #minetest-dev |
| 20:15 |
|
proller joined #minetest-dev |
| 20:20 |
|
proller joined #minetest-dev |
| 20:32 |
|
indiana joined #minetest-dev |
| 20:55 |
|
Cornelia joined #minetest-dev |
| 20:57 |
|
JDCodeIt joined #minetest-dev |
| 21:02 |
|
EvergreenTree joined #minetest-dev |
| 21:02 |
|
DI3HARD139 joined #minetest-dev |
| 21:12 |
|
YuGiOhJCJ joined #minetest-dev |
| 21:29 |
|
EvergreenTree joined #minetest-dev |
| 21:36 |
JDCodeIt |
Hi all - a performance question in FileCache - any particular reason the buffer size is 1024 on a read? Would performance increase with a larger buffer? FileCache::loadByPath... char buf[1024]; |
| 21:38 |
sfan5 |
using e.g. 4096 might increase performance, but not significantly |
| 21:45 |
JDCodeIt |
I was also looking at the reconnect speed - joining a world you've already been to. If nothing has changed, is there maybe a shortcut that the protocol could pass along as a first check - such as the SHA1 of a file full of all the SHA1's? This ahead of exchanging all the SHA1's one-by-one to see if any one or more changed. Of course one would need to calculate this SHA1 and associate it with an address/port of the previously visited |
| 21:47 |
JDCodeIt |
...and cache the list of SHA1 |
| 21:48 |
JDCodeIt |
...for each connection |
| 21:48 |
sfan5 |
sounds like lots of trouble for little gain |
| 21:53 |
JDCodeIt |
I increased the FileCache read buffer to 64k and the load off disk changed from 12 seconds to 4 seconds. Relatively significant? |
| 21:55 |
JDCodeIt |
Changing from stack-allocated to the heap made no difference |
| 22:05 |
sfan5 |
hm, loading a file 1 KB at a time usually isn't that slow |
| 22:05 |
|
proller joined #minetest-dev |
| 22:05 |
sfan5 |
maybe it's the std::ostream it's being pushed into? |
| 22:06 |
JDCodeIt |
It's a world with 3000+ inventory items, so it can be a while |
| 23:16 |
|
Lone-Star joined #minetest-dev |
| 23:41 |
|
EvergreenTree joined #minetest-dev |
| 23:46 |
|
Tmanyo joined #minetest-dev |
| 23:59 |
|
Lone-Star joined #minetest-dev |
