Hmm, some more issues:
linflas wrote:i played a moment with the ClockDivide value and there's no unique value that can be realistic for everything.
What I got from this list is that you feel thrown objects seem need to go a lot faster, attack recovery times need to be a little lower, and door opening and closing feel right the way it is... is that right?
You play mostly RTC, right? I based DSB's timing on CSBwin and memories of DM on an overburdened CPU... I think thrown objects in RTC go at a full 6 ticks/sec so it seems faster (still not as fast as ClockDivide=2 though!) Right now thrown things move one square per game "tick" so there's not a lot of flexibility there. Attack recovery times, more so, though, as far as I can tell, the values are straight out of DM... at least the numbers are.
linflas wrote:i press 7 to turn left and immediately after/at the same time, i press 8 to go forward but it seems that DSB can't cumulate succesive key typings while CSBwin or RTC can.
Wow, I didn't realize people played/typed this fast!
Ok, I'll take a look at this.
linflas wrote:is there an option to play DSB windowed ? with scaling ?
Yes to windowed, no to scaling-- but it's rather experimental. I hacked it in quickly and forgot about it, to be honest. If you'd like to play with it, though, add "Windowed=1" to your "Settings" section of dsb.ini.
Joramund wrote:Zo Kath Ra spell has "nil" mass -> bug when throwing (in util.lua)
No, for me, the mass is always initialized to 0, which is valid, and I can't reproduce this...
Joramund wrote:Item comments are not parsed as scroll text ( / doesn't mean new line)
It automatically breaks these into newlines. Good/bad?
Joramund wrote:Potential bug : poison cloud visible when thrown on hidden fake wall tile.
Clouds are visible on walls to make the Vi Altar work, and because that is (as far as I know) the way DM works too.
Joramund wrote:affect "class" attribute to objects that have particular behaviors but come in different types : eg for all potions have : class = "POTION" etc.
Ok, I'll do this.
Joramund wrote:dsb_get_cell(lvl, xc, yc) return cell_state (0 or 1)
dsb_set_cell(lvl, xc, yc, state) where state = 0 or 1 (square and wall)
It returns "true" or "false" (Lua booleans) but otherwise "dsb_get_cell" is in already. "dsb_set_cell" isn't, but I'll add that.