/[projet1]/users/chema/TINE/todolist.txt
Defence Force logotype

Contents of /users/chema/TINE/todolist.txt

Parent Directory Parent Directory | Revision Log Revision Log


Revision 199 - (show annotations)
Tue Jan 19 22:13:21 2010 UTC (10 years, 2 months ago) by Chema
File MIME type: text/plain
File size: 10017 byte(s)
Several suggestions by Dbug added:
some kind of in-screen instructions in some pages
arrow cursor support (partial)
solved the planet size jump thingy
added Dbug's optimization to stars.s. The easy one, the add/sub thing.
Music is correct now. Plays all patterns.
1
2 - Enemies allways fire you at the same position in the frame
3
4 - Adapt the ship diameter (size) from the data to the scaled-down versions generated by modelimport (see out.txt in the modelimport folder).
5
6 - PROBLEM: Seems that rotating 3 degrees is too much for easy targeting the enemy!!! That requires modifications in the engine for the player!!!
7
8 - Cannot add color if the frame is in pixel 5. If it is not, the frame should be drawn at every screen dump?
9
10 - Implement a save/load screen
11
12 - Check and manage fuel
13
14 - Buy/manage equipment
15
16 - Create enemies dynamically
17
18 - Missiles should not destroy planets... (should never target them!). Is this solved now?
19
20 - Complete AI
21
22 - Implement ECM for missiles
23
24 - Remove inflight messages afer jumping/docking...
25
26 - Update ship's energy, in a timely manner and to maximum when dock (or before leaving the planet)
27
28 - Scooping is implemented, but it seems that randomly creating the loot harldy creates anything (not even plates).
29 Don't know what is going on here... should check. UPDATE: Seems OK now (but still prob is low).
30
31 - When your ship is destroyed, sometimes everything seems to hung. No apparent reason, but maybe too many objects?
32 Seems not quite probable. Hard to make this repeatable.
33
34 - damage_ship needs a look... it is quite a disaster. DONE. Also modified damage_player to be sure to decouple it...
35
36 - I am not sure when, but sometimes increment_kills is called with wrong parameters and fails... Target the ADDER and destroy it when nearby to see. Maybe it was because of the lack of memory? reducing MAXSHIPS to 10 seems to have solved this. SOLVED?
37
38 - The first time we go back from a text screen to space view, the three lines of text go black paper. Also sometimes @ chars appear... Too low on memory?, but why if we are not creating new objects? Maybe we are copying more lines than necessary for drawing the cockpit?
39
40 - Our explosion is not quite good... be sure to launch splinters, at least... Ok now it allways launches splinters, but need more MAXOBJS to work. 10 are too low.
41
42 - If we are to use Fore and Aft shields, we need to modify damage_ship so something (like the carry flag)
43 indicates if the damaging ship is in front or not. It is necessary because of missiles (in which the register Y
44 indicates the launcher).
45
46 - Avoid scooping alloys when dying.
47
48 - Rotation of stars is wrong. ROTATEP and ROTATEN rotate just one step. Need to loop g_alpha times.
49
50 Dbugs idea here:
51 let say that from a particular orientation and velocity you can compute a vector
52
53 [13:55] <_Dbug_> this vector is a fixed point signed thing, that represents the increment in X/Y/Z to apply to a particular position to get to the new position in this particular direction at this particular speed
54 [13:56] <Chema> That is what it must be doing, if I am not wrong, for adding that vector to the object's centre position.
55 [13:56] <_Dbug_> Then there is no need for computing this vector if the angle or velocity does not change
56 [13:56] <_Dbug_> just need to add it each frame
57 [13:56] <_Dbug_> newpos=oldpos+incrementpos
58 [13:57] <_Dbug_> you need to recompute the move vector only if the orientation or the velocity changes
59 [13:58] <Chema> Something like keeping the vector for each ship and a flag indicating if it is still valid or not. If a ship rotates or accelerates/deccelerates it will become invalid. Else you apply the same calculation again.
60 [13:58] <_Dbug_> and if the velocity only changes, you can probably just scale the vector
61
62
63 - Modified back tinefuncs.s MoveCurrent so it does
64 jsr check_maxman
65 sta _rotXXXX,x
66
67 This was commented out, I don't know why. This is the preferred behaviour in Elite (from TNK: rotate 1 amount per turn, and keep track of how much is left).
68
69 Commented it back, so rotXXX keeps the value until told otherwise. Seems enemies are much more agile this way... Need to check.
70
71 - Flahses in red and yellow also happen when in screens different than FRONT_VIEW. This is buggy.
72
73 - Models seem too big. The first intro screen with the rotating COBRA starts overflowing after several minutes!
74
75 - After adding energy refill, speed gets back > 0 when dying. Need to add a game_over or player_dead variable to keep track of all this.
76
77 - Had to set HIRES mode from the INIST, as the copy of the charset was destroying data.
78
79 - Check missiles. When fired they seem to jump quite far from the launcher. Maybe speed is too high or they are moved too far.
80
81 - player-launched Missiles are too near sometimes, models are incorrectly drawn
82
83 - Compass is only updated if x,y changed, which is incorrect if changing from hollow to filled and
84 viceversa. Also when comming from other screens it is sometimes not updated. Same reason?
85
86 - Refilling energy and shields in hyper jump is not a good idea... maybe filing shields or x2 to both could do
87
88 - Need to /2 the legal status at each jump or something...
89
90 - Energy seems to get a value >128, which becomes negative confusing the cockpit panel
91
92 - Bad initialization of planet_dist, permits a jump inmediately after launching
93
94 - Front shield (maybe rear too?) is not updated correctly (after jump). Going to another screen and getting
95 back updates it.
96
97 - Are shields recharging?. Nope. If energy is negative (greater than 128) they don't recharge. Keep player maxen<128 is mandatory
98
99 - After galmap population race is wrong (change between galchart and sysinfo)
100
101 - At XEESLE you appear too far from planet (problems with wrapping). Seems solved now Zhigh is between $5 and $35, instead of $10 and $40
102
103 - Make stick tops larger
104
105 - Again after destroying player system hangs... not enough memory to create ships? Seems corrected again.
106
107 - Still can jump right after launching...
108
109 - Fuel circle seems somewhat wrong, when it is 3.0LY planets at 2.4LY are outside. (?)
110
111 - When frame rate goes too low, skip frame drawing
112
113 - Ships still attack from too far. Lasers incorrectly drawn.
114
115 - Seems to be an error where a shield may become less than 0 or something like that,
116 and is never decremented again. Load saved game and let the viper hit you. You should
117 see it.
118
119 - Implement Escape Pod
120
121 - Escape pod should clean your status
122
123 - Check prices... ECM and Escape pod seem interchanged, according to what is
124 said on the manual.
125
126 - After moving the laser origin and destination points for drawing to page 4, as they are not initialized, four erroneous lines appear at the first time space is draw.
127
128 - Check cirlce drawing. Uses tmp3 to decide if needs clipping or not, but internal vars use up to tmp7 ???
129
130 - Check the escape pod flags. When killing the ADDER it seems that _flags FLG_INNOCENT is not set (?)
131
132 + Model in the init screens get further and further slowly... ? Ok. No. The accumulation routines make
133 a roundoff error that actually makes the model smaller with time. Not sure what routine, should check.
134 If you make the orientation matrix again $40 0 0 0 $40 0 0 0 $40 0 0 0 0 0 0 0 0 0 it gets the correct
135 size again. This was not ocurring before????
136
137 Ok. There are roundoff errors. With the original sindelta/cosdelta tables, effects are different, but
138 exist. Should check if normalizing matrices is needed.
139
140 + Optimizing drawing of radar? It is calling pixel_address several times, also code is shared between erase and draw and it is inlined...
141
142 + Bounties added, but it is quite a waste to use 2 bytes for just a couple of cases... Anyway the inflight message with
143 the bounty appears maybe too fast? Maybe even before the ship explodes!
144
145 + Implement a Cargo screen
146
147 + AddSpaceObject needs tmp0 pointing to a coordinate structure, but it is allways _PosX.
148 Should we change this, or at least add another routine that loads the tmp0 pointer?
149
150 + Elite TNK uses the current galactic seed (6bytes) to create planets. As this seed may be corrupt (whenever used for creating
151 a galactic chart or to find a planet), I will use the system's seed of 4 bytes (base for goatsoup).
152
153 + When saving the laser vertex there is no check to see if it is not a ship (something like a planet)
154 beq normal
155 cmp #5 ; is it 1,2 or 4 ? Then planet, sun or moon
156 bcc checkmin ; skip max check
157 normal
158
159 - When drawing planets, if they are too close, maybe it is better not to draw them // if 0x6000/Z > 0xff then return (too close to draw?)
160 Now they can't be
161
162 - A certain hysteresis would be needed to avoid planets appearing and disappearing as we rotate.
163 Changed to $ffff/(Z/2).
164
165 + Look in tineloop.s
166 notdocked
167 jsr clr_hires
168 jsr load_frame
169 jsr _DrawFrameBorder
170 jsr _DoubleBuffOn
171 jsr _FirstFrame
172
173 If DrawFrameBorder is called before load_frame there is a bug the second time we switch from screen2 to frontview (the frame is erased at the bottom).
174
175 + Maybe implement dumping cargo?
176
177 + Mission scripting
178
179 + Use of changing the object that sets the point of view?
180
181 + Buying other ships?
182
183 + Missiles still eat up CPU cycles. Too many of them, I think.
184 Limit possible missiles at the same time, moreover for enemies,
185 which launch 3-4 missiles in a row. "Missile Jammed".
186
187 + When frame rate goes too high don't simulate all the time, or make a waiting loop
188
189 + Snapping to planet does not always work in short chart
190
191 + In the manual it is said that we start with 100 Cr, as in the original version.
192 However I am starting with 200, such as in the speccy version?
193
194 + Need some feedback of danger!!!
195
196 + Adding rear laser should be piece of cake
197
198 + Add a new sfx for hit when no shields?
199
200 - It's a bit quiet out there, isn't it... secure planets are quite... secure.
201 Now pirates are created even with police around.
202
203 + Force deleting a slot before being able to save to it ?
204
205 + Add prepare for landing at Diso, launching off messages to the cockpit.
206
207 + Arrows for lists?
208
209 + snow on the scanner when readings erratic
210
211
212
213
214

  ViewVC Help
Powered by ViewVC 1.1.26