… | |
… | |
4 | TODO: apartment-descritpions for gotisch1/2 |
4 | TODO: apartment-descritpions for gotisch1/2 |
5 | TODO: it is possible to inventory-lock items on the floor, and *spomehow* possible to drop them. |
5 | TODO: it is possible to inventory-lock items on the floor, and *spomehow* possible to drop them. |
6 | TODO: think about naking the rhyzian amulet useful - makes little sense in deliantra? |
6 | TODO: think about naking the rhyzian amulet useful - makes little sense in deliantra? |
7 | TODO: cast invisible... |
7 | TODO: cast invisible... |
8 | |
8 | |
9 | TODO: update_stats delay for players, between commands etc. |
9 | - (maps) add gorokh_final mapset from crossfire. |
10 | TODO: get_nearest_player slow |
10 | - (maps) block most vault-type mlab locations, open up |
11 | TODO: speed_left dumping very slow |
11 | imperial converters for anybody to use. |
|
|
12 | - (maps, arch) replace all permanent stat potions by random |
|
|
13 | variants. |
|
|
14 | - (maps) spread gorokh demon styles more evenly over 1..100. |
12 | - TRIGGER events got passed a garbage pointer instead of a |
15 | - TRIGGER events got passed a garbage pointer instead of a |
13 | real player object on player changes. |
16 | real player object on player changes. |
|
|
17 | - diseases and maybe other things could sometimes reduce the speed |
|
|
18 | to below MIN_ACTIVE_SPEED, deactivating the monster and causing |
|
|
19 | much grief. |
14 | - shop specialisation calculation was totally broken. |
20 | - shop specialisation calculation was totally broken. |
15 | - nrof/itemcount/volume limits when dropping to the floor are |
21 | - nrof/itemcount/volume limits when dropping to the floor are |
16 | now enforced. |
22 | now enforced. |
17 | - fix magic mapping, which only worked for inate spell casting. |
23 | - fix magic mapping, which only worked for inate spell casting. |
18 | - godgiven items will now be explicitly described as such. |
24 | - godgiven items will now be explicitly described as such. |
19 | - materialnames would change "randomly" on every load, only |
25 | - materialnames would change "randomly" on every load, only |
20 | set the material when the materialname was not set before. |
26 | set the material when the materialname was not set before. |
21 | - balancing change, halve rod recharge speed (instead |
27 | - balancing change, halve rod recharge speed (instead |
22 | of 4.8s, the standard rods now recharge in ~10s. not a solution, |
28 | of 4.8s, the standard rods now recharge in ~10s. not a solution, |
23 | but still. |
29 | but still. |
24 | - reduce cpu time used for map saving by >40% by avoiding |
30 | - reduce cpu time used for map saving by >>50% by avoiding |
25 | the costly flasg-checking loop in most cases. |
31 | the costly flag-checking loop in most cases and using a simple |
|
|
32 | repeat-cache for speed_left. |
26 | - always delay stats update for players on insert/remove till |
33 | - always delay stats update for players on insert/remove till |
27 | next tick/command, which saves a lot of cpu cycles on drop/take. |
34 | next tick/command, which saves a lot of cpu cycles on drop/take. |
|
|
35 | - add hint about havign to run to attach unaggressive npcs. |
28 | - pet monster level was not displayed correctly in "showpets index". |
36 | - pet monster level was not displayed correctly in "showpets index". |
29 | - clean up floating point conversions and many similar nitpicks. |
37 | - clean up floating point conversions and many similar nitpicks. |
30 | - fix hanuk altar. |
38 | - fix hanuk altar. |
31 | - overhaul material system to be much faster, and having less bugs, but |
39 | - overhaul material system to be much faster, and having less bugs, but |
32 | still having a lot of cruft :/. |
40 | still having a lot of cruft :/. |
… | |
… | |
42 | - fixed a small uncleanlieness in item converters, which might |
50 | - fixed a small uncleanlieness in item converters, which might |
43 | have led to weird item multiplications. |
51 | have led to weird item multiplications. |
44 | - store hash value in shstr, not used by anything, but doesn't use much memory. |
52 | - store hash value in shstr, not used by anything, but doesn't use much memory. |
45 | - fixed crash bug in learn_spell command. |
53 | - fixed crash bug in learn_spell command. |
46 | - remove all traces of FLAG_NO_FIX_PLAYER. |
54 | - remove all traces of FLAG_NO_FIX_PLAYER. |
|
|
55 | - avoid sv_derived_from calls (lots of hash lookups) when converting |
|
|
56 | from perl to c++. |
47 | - implemented least_significand_bit and for_all_bits_sparse_32 utilities |
57 | - implemented least_significand_bit and for_all_bits_sparse_32 utilities |
48 | and use it in some "inner" loops. |
58 | and use it in some "inner" loops. |
49 | - rename flags to attachable_flags to avoid confusion :=). |
59 | - rename flags to attachable_flags to avoid confusion :=). |
|
|
60 | - fixed jeweler experience output to not be floating point. |
|
|
61 | - added lock/unlock commands, thanks for the patch go to Shawn Robinson! |
|
|
62 | - (arch, maps) introduce inactive_shop_floor and fix some shops with it. |
|
|
63 | - try to prefetch memory inside process_objects, which can give |
|
|
64 | enourmous speed gains. |
|
|
65 | - make some small optimisation to the process_objects loop. |
|
|
66 | - move facedata chksum calculation into cfutil, make server more flexible |
|
|
67 | handling different chksum lengths. |
|
|
68 | - check for hash collisions in cfutil. |
|
|
69 | - reduce face checksum size from 6 to 4 bytes. |
|
|
70 | - don't crash when a map uses empty sound/sound_destroy/face. |
|
|
71 | - in general, don't crash on value-less resource file lines |
|
|
72 | anymore, due to get_str never returning a 0-pointer. |
|
|
73 | - code cleanup in move_symptom. |
50 | |
74 | |
51 | 2.90 Sat Nov 7 15:06:06 CET 2009 |
75 | 2.90 Sat Nov 7 15:06:06 CET 2009 |
52 | - some Andrew Madloch wanted, despite being specifically told |
76 | - some Andrew Madloch wanted, despite being specifically told |
53 | multiple times that the Deliantra parts additions are under AGPL, |
77 | multiple times that the Deliantra parts additions are under AGPL, |
54 | lock up the source code of the server within his company. |
78 | lock up the source code of the server within his company. |