Friday, February 4, 2011

TGTIWLTP: Implimentation

It looks like I need to be careful not to add too many complex features that make a game impossible to create. So the main points would be:
Probability functions
dynamic world (aka dwarf fortress)
unlocks (simple to start with, also allows easy upgrade)
information (no need for external theory crafting)
constructive (not exclusively, kill everything that moves)
degenerative health (cannot function at 100% when only 1HP remains)
Polish (simple graphics that move smoothly)
small (can play on older PCs)
free (open source, donations maybe)
single player
no elves, orks, trolls. mostly original classes. (Real animals allowed)
not locked into specific class stereotypes (practice results in progress)
communities (play both sides of the coin)
sentimentality ( to explain where artefacts come from)
realistic death (with consequences to the world)
realistic backpack (not using Judo while carrying 3 sets of plate mail armour)
Rogelike
barter system (some people want gems, some want milk)
scales & proficiency rather than level ups (real life is not digital)
That is all, probably too much actually.

Version 1 should have:
1 class (woodcutter)
1 race (trinitarian)
1 homestead (3 people)
1 skill (axe)
4 items: axe, wood, food, water
1 backpack (things always carried)
1 Loot sack (dropped when performing actions)
0 economy (free items from mum)
0 enemies
0 unlocks (they have not been made yet)
dynamic world (let graves be made for each death)
single map, no over world.
Probability functions (using axe on tree, woodcutting skill)
sentimentality (for axe, full development of this one item)
realistic death (no rez, but most things will leave you for dead rather than kill)
free
single player
Rogelike
Plot (available for thickening)
2 quests (fetch wood for fireplace, do not get eaten by wolves)

This should be sufficient to create a workable platform on which the rest of the game could be built.
After this everything else can be considered unlockable, ie wolves, clothes.
For me version 1 would be the hardest part of making a game. The theory and IP will take the longest and if the game ever starts getting made will probably be ongoing like minecraft. But thinking about what to add is easy "if you can cook a meal with words, I will promise you an ocean of butter" is an Arabic saying that I remember.
Version 2 will not nearly be so hard. I have been able to mod morrowind/puzzle quest/Red Alert #?/Fallout 3/titan quest/AOE1 without many hassles but in these games the framework was already there to develop new quests items, monsters, races, campaigns. From my uni programming days I know that it is a different kettle of fish altogether when you are starting from scratch.

In the meantime as this game is not made I will continue to roll a halfling thief who wields a battleaxe and is too nice to steal. Or maybe I will multiclass a necromancer healer.

I remember one rogelike that I really enjoyed playing except that it really only had 2 quests. I could see that the game went a lot further than that and that there were a lot of race/classes awaiting deployment but not enough to keep me playing. I decided to definitely come back to that variant in a year’s time. But now I have forgotten what it was called. The version one I have described here would probably keep me entertained for half an hour.
So I would probably keep it in alpha until there was an infinite dungeon that generates a new level each time the stairs are used. This will allow any character to play with the scalars and proficiencies without any plot interruptions, nice and simple hack and slash.
As far as enemies goes I suppose there needs to be a bank of unintelligent animals with which to populate the starting world.
I didn’t actually realise how complicated the resistance math was. There are a lot of extra things to consider. Then there are dozens of ways to alter the values. No wonder this iteration of tome (Tome4) had no spell casters in the beginning. The resistance buffs need to take into account a lot of different variables.
Then things need to be tested for balance, since I have a different damage calculation this may be not very easy. I am starting to see why nearly all RPGs use a d20 system.

Each race needs a normal default:
physiology
Preferred class
Preferred skill
Preferred home
Unique active
Passive
Preferred food

Each class needs:
starting location
guild (formal or informal)
pre learned skill
item + clothes

1 comment:

  1. Polish (simple graphics that move smoothly)
    small (can play on older PCs)
    free (open source, donations maybe)
    ...Roguelike
    ...like minecraft

    Maybe try something like this: Java Roguelike Engine -
    http://roguelike-eng.sourceforge.net/

    Would let it run on any computer (though technically against the small footprint goal since java uses more resources than a dedicated c++ would), is free and hopefully easy to implement, leaving you more time for all of the databases you would need to conceptualize to store all the variables etc XD

    Not sure if the mechanics are locked in on it though, so it may not be doable with the combat mechanics you have in mind

    ReplyDelete