|
|
1 | http://www.vionline.com/sound.html |
|
|
2 | http://www.stonewashed.net/2.html |
|
|
3 | |
| 1 | http://pyopengl.sourceforge.net/documentation/manual/glutEnterGameMode.3GLUT.html |
4 | http://pyopengl.sourceforge.net/documentation/manual/glutEnterGameMode.3GLUT.html |
| 2 | |
5 | |
| 3 | 12:17 <@schmorp> i do list all libs manually for pclient |
6 | 12:17 <@schmorp> i do list all libs manually for pclient |
| 4 | 12:17 <@schmorp> will have to do that for gce too |
7 | 12:17 <@schmorp> will have to do that for gce too |
| 5 | 12:17 <@schmorp> remind me of it |
8 | 12:17 <@schmorp> remind me of it |
| 6 | 12:17 <@schmorp> qwhen i next bundle it |
9 | 12:17 <@schmorp> qwhen i next bundle it |
| 7 | |
10 | |
| 8 | - IMPORTANT: containers must be arrays, not hashes, for floor ordering |
11 | - IPORTANT³: server crashes |
| 9 | - IMPORTANT: slider should clip to valid range properly |
12 | >In the ruins of Narcopin lies the tomb of Aljwaf. |
| 10 | - IMPORTANT: user interface (window positions etc.) should be saved, but |
13 | ... |
| 11 | a) switching modes clears/recreates all widgets (has to, as parameters |
14 | .. |
| 12 | are not resolution-independent) |
15 | many dangers, great rew< |
| 13 | b) widget layout is nontrivial as to not obscure map |
|
|
| 14 | c) need better way to center map then middle-button, maybe map overview |
|
|
| 15 | (but updates will be slow UPDATE updates are not at all slow). |
|
|
| 16 | - IMPORTANT: animate faces in inventory etc. |
|
|
| 17 | |
16 | |
| 18 | - remove wrap_mode hack from texture and make it a per-texture operation |
17 | |
|
|
18 | - IMPORTANT²: if the metaserver dialog is open while save confgi was used, the client |
|
|
19 | crashe son every startup. |
|
|
20 | - IMPORTANT: running over or pickung up large stacks is etxremely slow. |
|
|
21 | - IMPORTANT: bind to key should always be verbose (DONE) |
|
|
22 | (elmex: what is meant by 'verbose' ?) |
|
|
23 | (^ whats being bound to a key, see other, similar, menu entries?) |
|
|
24 | - IMPORTANT: please look over the whole keybinding thing, there are myriads of little bugs (elmex: which bugs for example?) |
|
|
25 | (documenting as found: the binding window is behind other windows, since it forcefully grabs focus, |
|
|
26 | - it should always be topmost (z => 1000 or so) |
|
|
27 | - IMPORTANT: armour might not be used in the protocol anymore, maybe resist_physical should be used as armour? |
|
|
28 | (or maybe even better: not have an Arm value anymore if its not used in the protocol?) |
|
|
29 | - IMPORTANT: texture reload not working on scaras machine (maybe now) |
|
|
30 | - IMPORTANT: gauge size adjustment wreaks havoc, probably needs reconfigure calls |
|
|
31 | - IMPORTANT: discuss topmost menu, minor issues |
|
|
32 | |
|
|
33 | should be solved/investitaged before BETA release: |
|
|
34 | - binding keys is too complicated in the general case => disucss |
|
|
35 | - maybe the topmost fancyframe border should be larger (for close button and title to be more visible?) |
|
|
36 | - the binding dialog is sometimes behind all windows |
|
|
37 | - maybe support mb4/mb5 on selected widgets to facilitate scrolling |
|
|
38 | (inventory, spell list, messge log...) => marc |
|
|
39 | - add lmb mouse = cast, mmb2 = invoke or so to spell list. |
|
|
40 | - maybe open the help viewer on the first start, or a simple dialog that |
|
|
41 | initially lists keybidings for gcfclient converts. |
|
|
42 | - binding window and editor layout broken for long recordings |
|
|
43 | - minor: skill list which shows XP and lets u bind stuff |
|
|
44 | - Attempt to free unreferenced scalar: SV 0x11f5d98 during global destruction. |
|
|
45 | (after starting, press quit) |
|
|
46 | - mapmap (overview) - scroll visible area by clicking/dragging |
|
|
47 | - maybe move window managing functionality into toplevel: better window management (graphical feedback) |
|
|
48 | - the player weight somehow updates strangely when dropping/picking up |
|
|
49 | (update: is this still the case?) |
|
|
50 | |
|
|
51 | should be solved/investigated before 1.0 release: |
|
|
52 | - spell list => scroll down => "Pickup" => scrollbar not updated |
|
|
53 | - what happened to the dialog being shown when modifiera are pressed, showing |
|
|
54 | keybindings for this modifier? |
|
|
55 | - add menu itemds for inventory items for UNcommon actions: mark/inscribe etc. |
|
|
56 | - split up gauges into: head, body and foot, so that contents begin in the interior of gauges |
| 19 | - enter runmode when cursor-key repeats |
57 | - enter runmode when cursor-key repeats (maybe not?) |
| 20 | - mapmap (overview) |
58 | - better focus management: put focus onto active windows, remove it again later |
| 21 | - convetr to real window |
59 | (mainly to speed up gui interaction when map isn't important). |
| 22 | - draw rectangle around visible area |
60 | - cairo/win32 looks like shit (premultiplied alpha bug?) |
| 23 | - scroll visible area |
61 | (despite horrendous efforts, this is likely not fixable except by building a custom |
| 24 | - make size configurable/reesizable |
62 | libcairo for win32). |
| 25 | - use real mapmap size in map caching, not hardcoded 250 size |
63 | (update: disabled antialiasing, this looks bad, but much better than antialiasing) |
|
|
64 | - the binding editor should have a append-record mode or something like that |
|
|
65 | - make a real "table" for the spell list (not long strings like lvl: xxx sp: xx) |
|
|
66 | and add column titles (also for other tables like inventory and bindings) |
|
|
67 | - sliders do not change size after reconfigure. |
|
|
68 | - use more graphical frames/separators to make dialogs more eye-pleasing/clean. |
|
|
69 | - investigate fill widget option. |
|
|
70 | - win32, fow_tetxure sometimes nukes other textures |
|
|
71 | (mostly fixed, but still there: displays list and texture uploads do not go well together) |
|
|
72 | - get rid of ScrolledWindow special case and use draw_xywh rectangle inside TextView for optimisation. |
|
|
73 | - completer should know more about arguments, e.g. cast summon pet monster, |
|
|
74 | or that some commands do not take arguments ("drop all"). |
|
|
75 | - textview should not snap to bottom on resize. |
|
|
76 | - alt+cursor == diagonal |
|
|
77 | - look into extendedmapinfo |
| 26 | - save only dirty cached maps, and save dirty maps != current map |
78 | - save only dirty cached maps, and save dirty maps != current map |
| 27 | - elmex: split up gauges into: head, body and foot, so that fillings begin in the interior of gauges |
79 | - try to synchronize local animation speed with server updates to save on screen refreshes. |
| 28 | - the stats for the gauges which get into the stats_update function are somehow weird |
80 | - player list from server for tell etc. commands |
|
|
81 | - examine etc., should use extendeddrawinfo or sth. similar |
|
|
82 | - maybe use outline fonts/effect for statusbox and floorbox (and health gauge stats etc.). |
| 29 | - fow intensity => fow color (greyscale, just rename), add fow darkness slider, too |
83 | - fow intensity => fow color (greyscale, just rename), add fow darkness slider, too |
| 30 | - message window layout broken (entry too small). |
|
|
| 31 | - mesa-rendering of textview shows only garbage and/or another texture |
|
|
| 32 | - label needs "expected char width" |
|
|
| 33 | - toplevel should enforce that windows always be partially visible |
|
|
| 34 | - maybe move window managing functionality into toplevel |
|
|
| 35 | - look into extendeddrawinfo and extendedmapinfo |
|
|
| 36 | - add extenstive tooltips to stats window, basically like a manual :) (turn undead resistance missing) |
|
|
| 37 | - performance: use texture collections for upstream server data instead if gobs |
84 | - performance: use texture collections for upstream server data instead if gobs |
| 38 | of small textures. |
85 | of small textures. |
| 39 | - when mipmapping upstream data files, run a fatten pass so pixels with alpha=0 don't negatively |
86 | - when mipmapping upstream data files, run a fatten pass so pixels with alpha=0 don't negatively |
| 40 | affect neighbouring visible pixels (border bleeding). |
87 | affect neighbouring visible pixels (border bleeding). |
|
|
88 | - pango fontsize measure and decrease to achieve real pixel height |
| 41 | |
89 | |
| 42 | TEMPORARY SERVER TODO: |
90 | TEMPORARY SERVER TODO: |
| 43 | - palyer peaceful setting should be independen of game peaceful setting |
91 | - palyer peaceful setting should be independen of game peaceful setting |
| 44 | (i.e. one should be able to become hostile against guards but still |
92 | (i.e. one should be able to become hostile against guards but still |
| 45 | be peaceful - the palyer peaceful would be toggled by priests and shown |
93 | be peaceful - the palyer peaceful would be toggled by priests and shown |
| 46 | in who, the game peaceful steting would be toggled by the peaceful command |
94 | in who, the game peaceful steting would be toggled by the peaceful command |
| 47 | but not otherwise shown). |
95 | but not otherwise shown). |
| 48 | - run flawfinder on cf |
|
|
| 49 | - disable old socket mode in server |
|
|
| 50 | - implement suicide command for server |
|
|
| 51 | - cavehippo wants hashstring to be changed to hash name |
|
|
| 52 | - pippij wants playerstealing to work between hostile players |
96 | - pippij wants playerstealing to work between hostile players |
| 53 | - schmorp wants to rework the who output format |
|
|
| 54 | - attempt_jump tries to kick jumped-into monster,s but doesn't work |
97 | - attempt_jump tries to kick jumped-into monster,s but doesn't work |
| 55 | - 1) If you hide, and someone can see you trying to hide, you'll get a |
98 | - 1) If you hide, and someone can see you trying to hide, you'll get a |
| 56 | message, even when you can't see that other person. |
99 | message, even when you can't see that other person. |
| 57 | - 2) hiding exp is always 1, independant of how difficult it is to hide |
100 | - 2) hiding exp is always 1, independant of how difficult it is to hide |
| 58 | - 3) jumping into monsters does no damage, even though it's supposed to |
101 | - 3) jumping into monsters does no damage, even though it's supposed to |
| 59 | be an attack (mentioned before). You also get no exp for this |
102 | be an attack (mentioned before). You also get no exp for this |
| 60 | - 10) (feature request): bug/typo/idea commands to automatically log |
103 | - 10) (feature request): bug/typo/idea commands to automatically log |
| 61 | your comments, with the mentioning of the current map (and perhaps an |
104 | your comments, with the mentioning of the current map (and perhaps an |
| 62 | item, if you do 'bug <item>'). |
105 | item, if you do 'bug <item>'). |
| 63 | |
|
|
| 64 | "workarounded" but not solved: |
|
|
| 65 | (- IMPORTANT: move widgets to integer coordinates - fix containers!) |
|
|
| 66 | (- pango fontsize measure and decrease to achieve real pixel height) |
|
|
| 67 | |
106 | |
| 68 | set_face 4783 => 70 |
107 | set_face 4783 => 70 |
| 69 | libpng warning: Ignoring gAMA chunk with gamma=0 |
108 | libpng warning: Ignoring gAMA chunk with gamma=0 |
| 70 | allocating id 71 |
109 | allocating id 71 |
| 71 | OK |
110 | OK |
| … | |
… | |
| 78 | allocating id 73 |
117 | allocating id 73 |
| 79 | OK |
118 | OK |
| 80 | set_face 4784 => 73 |
119 | set_face 4784 => 73 |
| 81 | libpng warning: Ignoring gAMA chunk with gamma=0 |
120 | libpng warning: Ignoring gAMA chunk with gamma=0 |
| 82 | |
121 | |
|
|
122 | To achieve a more complete blending, you can have your drawing engine rasterize each transparent object more than once, altering in each pass the blending mode, object alpha channel, and buffer write masks. The first pass should perform RGB blending. Accordingly, you should disable writing any alpha channel or z buffer data during this pass. |
|
|
123 | /*first pass*/ |
|
|
124 | glColorMask(TRUE, TRUE, TRUE, FALSE); /*disable alpha channel*/ |
|
|
125 | glDepthMask(FALSE); /*disable Z buffer*/ |
|
|
126 | if (premultpliedTransparency) |
|
|
127 | glBlendFunc(GL_ONE, GL_ONE_MINUS_SRC_ALPHA); |
|
|
128 | else |
|
|
129 | glBlendFunc(GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA); |
|
|
130 | /*render the object here*/ |
|
|
131 | |
|
|
132 | On the second pass, you should set the frame buffer alpha channel value to (1- a s )(1- a d ). To do this, you need to render the object again, with a different alpha value, as follows: |
|
|
133 | /*second pass*/ |
|
|
134 | glColorMask(FALSE, FALSE, FALSE, TRUE); /*enable alpha channel*/ |
|
|
135 | glDepthMask(FALSE); /*disable Z buffer*/ |
|
|
136 | glBlendFunc(GL_ONE_MINUS_DST_ALPHA, GL_ZERO); |
|
|
137 | /*render the object with alpha replaced with 1-a*/ |
|
|
138 | |
|
|
139 | Finally, the third pass should replace the value in the alpha channel with the final value 1-((1- a s )(1- a d )). To do this, you need to render the object again, with its alpha value set to 1, as follows: |
|
|
140 | /*third pass*/ |
|
|
141 | glColorMask(FALSE, FALSE, FALSE, TRUE); /*enable alpha channel*/ |
|
|
142 | glDepthMask(TRUE); /*enable Z buffer*/ |
|
|
143 | glBlendFunc(GL_ONE_MINUS_DST_ALPHA, GL_ZERO); |
|
|
144 | /*render the object with alpha replaced with 1*/ |
|
|
145 | |
|
|
146 | |
