An Open Letter about TMW Development
Posted: 12 Feb 2014, 16:46
An open response to a number of threads about Art & Content
Maps/Tiles/Tilesets/WorldMap
The old crypt maps were awful... sooo boring.
All my finished work is here.
https://github.com/wushin/tmwa-client-d ... ts-release
This also includes the Underworld Maps.
It was planned, couldn't get enough help to script.
I've also made the Castle Tiles fit Tonori and extended that area
https://github.com/wushin/tmwa-client-d ... i-Town-Wip
But recently I've been reworking old tilesets submitted to fit the game and broke up/changed some
Started working with a simpler templated form to increase the volume of sets and easier mapping.
https://github.com/wushin/tmwa-client-d ... hics/tiles
I've been thinking about tilesets and the structure of them for awhile.
First things that have change possibly since you mapped/arted last:
1) Ground & Over can have more then 1 layer
2) Fringe is for fringe only, it is not another ground layer
3) Wooden Staff & Christmas Tree hat can break fringe/overs
4) Arguing art quality is as simple as no worse than the worst set in game (I'd say a tie between the original cave sets)
5) Animated tiles show up in tiled
6) Animated tiles are getting a new xml set-up (bjorn & 4144 were working on this I believe)
7) Maps are now CSV
As for tiles/tilesets, We don't use near the features we could; Atlas, dynamic tilesets, tmOptimizer, tiled_baker, etc.
A lot of these methods would save us time and space in client-updates.
If you notice under skunkworks I've begun to pull layers apart and put them onto larger development sets.
It has very basic sets with all terrians combined. Why Combine terrains?
Whats Mud? Dirt and Water. Whats the difference between a beach and a desert? water. Whats the difference between a forest and plain? Vegetation quantities.
Throw some pebbles on sand you got gravel, throw some stones on it it's coarse gravel.
Everything in Pixel art is layers, layers and more layers. Whoaa!! Wushin stop, thats a lot of layers.
It is a lot of layers, but we could make a DevMaps and DevTiles folder and compile the client-data maps/ tilesets from that via tmxOptimizer and tiled_baker.
These are scripts available via tiled as suggested other projects. All can be found on github.
One other thing to make mapping easier is to constrain all maps to a standard size and lay them out in a grid pattern. and adjust naming conventions as such.
Thus the upper left corner becomes 001-001-1. Knowing the area of the playing field plus what I already mentioned has enormous benefits.
1) Map edging is far easier (and my skunkworks has a script that already does it)
2) Using a Dev set of simple "placeholder" tiles we can map the unmapped maps with placeholder tiles so when someone wants to map that have a world reference as to what it should generally be composed of.
3) Then we can create minimaps and stitch them all together to form a world map, throw an edge on it bam, Coordinated World Map that directly correlates to the maps contained in the content.
4) The World Map is "set" to some extent and we don't run into the "Tonori Problem" https://wiki.themanaworld.org/index.php/Geography
Note the difference in the Tonori over world map and the Tonori in-game maps.
Sprites/NPCs/Mobs
The player sprite should be finished for more attacks, poses & directions
either the 1.5 or a 2.0
1.5 Player sprite:
- Body, Underpants & Eyes can be dyed
- Races built from layering textures onto base sprite
- Additional Attack frames
- Additional Emote frames
2.0 Player Sprite
- Sprite 1.5 + the diagonals
Replace humanoid NPCs with sprites instead of static images.
This branch has examples of replacing all NPCs with base sprite.
https://github.com/wushin/tmwa-client-d ... sprite-1.5
This helps add a lot of consistency across the game. As the Tulimshar Bizarre is the perfect name for it.
Looking forward this makes:
1) NPCs Variations = # of item sprite combinations & dyes
2) NPCs Animations are as easy as using the player sprite base (Walk, default-attack, etc)
3) Mobs can be made the same way
4) Each Mob/Npc added from new equipment to a humanoid mob automatically create PC Gear
5) NPCs could be fully mobile
6) Racies made as equipment means; Transformation
7) Reduce the number of art request for NPCs
8) Increase the ease of creating a NPC for content
9) Ability to randomly generate NPCs
Script
Currently the server data and scripts are a hodgepodge of different standards from different years.
While there is some discussion to get a firm standard set for style.
The core code for the most part in server data does a lot of neat things, but the variations from script to script really bogs down the newbie and sometimes creates redundant scripts.
The randomness also creates hurdles when trying to do things like a Quest Log or a spell like Clairvoyance
Or even getting the ! ? marks to appear. (see Evol, and sob TMW)
When I look at the scripts I see a framework inside most of them.
Most scripts rely on a core set of similar functions.
The game is heavy grinding because questing gives so little Xp after a while because the quest dry out. Mainly due to the hurdles of quest writing. The biggest I noticed was bitmasking.
https://github.com/themanaworld/tmwa-se ... nuals/xmas
Thats what Xmas was really about. How-to write Script Frameworks in TMW.
As you notice most of what is in Xmas is controlled by a config file and series of functions. The core of the Quest is left interchangeable and the requirements easy to sub in/out or change completely while losing none of the code.
Also hooks are added in so GMs could perform various actions from botcheck.
As it would be fun if GMs could be like the Wizard of Oz.
https://github.com/wushin/tmwa-server-d ... i-Town-Wip
I've made further improvements based on what I learned with the starting tutorial but I'm not done with that script yet. (I'm also not opposed to someone picking it up.)
As the more I looked at it the more quests fell into a few categories:
Gather X, Take X to Y, Cindy/Candor/Rossy Battle, End Game (illia), Mini-game or other.
From what I've learned:
1) Script needs clean up
2) Scripts need a complete framework for questing
3) Base functions can be made for the simpler quest types
4) Adding more Quests offsets the problem with the grind
5) Encourages more levels of experience to help
6) Allows Dialogue to be developed Separately with ease.
Note: If we could use something like https://github.com/tweecode/twine on TMW-Script writers/scripters/collaborators would have an easier time.
Wiki/ContentDisplay/ContentSearch
Currently all the differences in the above code make it harder to dynamically generate information in a human (player, not dev) standpoint.
The scripts used prior were outdated and even after running them and fixing them I was still not satisfied.
Often what is on the wiki falls behind if not directly related to a quest, item thats important to players. Even then what the player see and what the code says confuses them. For Example, XP on the https://wiki.themanaworld.org/index.php ... _Reference displays the base XP before the mutations. It's silly we parse it instead of possibly either using a github page and referencing it in on that page or finding a plugin to read the data from git itself and making wiki templates as the items, mobs, etc. will still need a Mythos to them.
The less hand tinkering writers have to spend creating uniform pages the more they can write. It's why it's a shame this page https://wiki.themanaworld.org/index.php/Dev:Main doesn't have a wiki section.
TMW Governance
http://en.wikipedia.org/wiki/Cooperative
Why we pick anything other than a Co-operative operating style baffles me.
As a Content Dev I'm kinda amazed at how I feel into it. We elect Gms but not the Devs. Lead Positions would need to be filled so I'd say anyone could run for a position and we could let the population vote based on X,Y,Z criteria. Making sure each candidate understands it's a year long commitment and a new election can occur if they abdicate the position.
and the Road goes on and on...
Fix these things and we will have an awesome game. Until then, I'll bump this when we are back at this point again. Where adding new content only increases the number of problems we have later.
As I've said this all before.
viewtopic.php?f=2&t=17834
Maps/Tiles/Tilesets/WorldMap
The old crypt maps were awful... sooo boring.
All my finished work is here.
https://github.com/wushin/tmwa-client-d ... ts-release
This also includes the Underworld Maps.
It was planned, couldn't get enough help to script.
I've also made the Castle Tiles fit Tonori and extended that area
https://github.com/wushin/tmwa-client-d ... i-Town-Wip
But recently I've been reworking old tilesets submitted to fit the game and broke up/changed some
Started working with a simpler templated form to increase the volume of sets and easier mapping.
https://github.com/wushin/tmwa-client-d ... hics/tiles
I've been thinking about tilesets and the structure of them for awhile.
First things that have change possibly since you mapped/arted last:
1) Ground & Over can have more then 1 layer
2) Fringe is for fringe only, it is not another ground layer
3) Wooden Staff & Christmas Tree hat can break fringe/overs
4) Arguing art quality is as simple as no worse than the worst set in game (I'd say a tie between the original cave sets)
5) Animated tiles show up in tiled
6) Animated tiles are getting a new xml set-up (bjorn & 4144 were working on this I believe)
7) Maps are now CSV
As for tiles/tilesets, We don't use near the features we could; Atlas, dynamic tilesets, tmOptimizer, tiled_baker, etc.
A lot of these methods would save us time and space in client-updates.
If you notice under skunkworks I've begun to pull layers apart and put them onto larger development sets.
It has very basic sets with all terrians combined. Why Combine terrains?
Whats Mud? Dirt and Water. Whats the difference between a beach and a desert? water. Whats the difference between a forest and plain? Vegetation quantities.
Throw some pebbles on sand you got gravel, throw some stones on it it's coarse gravel.
Everything in Pixel art is layers, layers and more layers. Whoaa!! Wushin stop, thats a lot of layers.
It is a lot of layers, but we could make a DevMaps and DevTiles folder and compile the client-data maps/ tilesets from that via tmxOptimizer and tiled_baker.
These are scripts available via tiled as suggested other projects. All can be found on github.
One other thing to make mapping easier is to constrain all maps to a standard size and lay them out in a grid pattern. and adjust naming conventions as such.
Thus the upper left corner becomes 001-001-1. Knowing the area of the playing field plus what I already mentioned has enormous benefits.
1) Map edging is far easier (and my skunkworks has a script that already does it)
2) Using a Dev set of simple "placeholder" tiles we can map the unmapped maps with placeholder tiles so when someone wants to map that have a world reference as to what it should generally be composed of.
3) Then we can create minimaps and stitch them all together to form a world map, throw an edge on it bam, Coordinated World Map that directly correlates to the maps contained in the content.
4) The World Map is "set" to some extent and we don't run into the "Tonori Problem" https://wiki.themanaworld.org/index.php/Geography
Note the difference in the Tonori over world map and the Tonori in-game maps.
Sprites/NPCs/Mobs
The player sprite should be finished for more attacks, poses & directions
either the 1.5 or a 2.0
1.5 Player sprite:
- Body, Underpants & Eyes can be dyed
- Races built from layering textures onto base sprite
- Additional Attack frames
- Additional Emote frames
2.0 Player Sprite
- Sprite 1.5 + the diagonals
Replace humanoid NPCs with sprites instead of static images.
This branch has examples of replacing all NPCs with base sprite.
https://github.com/wushin/tmwa-client-d ... sprite-1.5
This helps add a lot of consistency across the game. As the Tulimshar Bizarre is the perfect name for it.
Looking forward this makes:
1) NPCs Variations = # of item sprite combinations & dyes
2) NPCs Animations are as easy as using the player sprite base (Walk, default-attack, etc)
3) Mobs can be made the same way
4) Each Mob/Npc added from new equipment to a humanoid mob automatically create PC Gear
5) NPCs could be fully mobile
6) Racies made as equipment means; Transformation
7) Reduce the number of art request for NPCs
8) Increase the ease of creating a NPC for content
9) Ability to randomly generate NPCs
Script
Currently the server data and scripts are a hodgepodge of different standards from different years.
While there is some discussion to get a firm standard set for style.
The core code for the most part in server data does a lot of neat things, but the variations from script to script really bogs down the newbie and sometimes creates redundant scripts.
The randomness also creates hurdles when trying to do things like a Quest Log or a spell like Clairvoyance
Or even getting the ! ? marks to appear. (see Evol, and sob TMW)
When I look at the scripts I see a framework inside most of them.
Most scripts rely on a core set of similar functions.
The game is heavy grinding because questing gives so little Xp after a while because the quest dry out. Mainly due to the hurdles of quest writing. The biggest I noticed was bitmasking.
https://github.com/themanaworld/tmwa-se ... nuals/xmas
Thats what Xmas was really about. How-to write Script Frameworks in TMW.
As you notice most of what is in Xmas is controlled by a config file and series of functions. The core of the Quest is left interchangeable and the requirements easy to sub in/out or change completely while losing none of the code.
Also hooks are added in so GMs could perform various actions from botcheck.
As it would be fun if GMs could be like the Wizard of Oz.
https://github.com/wushin/tmwa-server-d ... i-Town-Wip
I've made further improvements based on what I learned with the starting tutorial but I'm not done with that script yet. (I'm also not opposed to someone picking it up.)
As the more I looked at it the more quests fell into a few categories:
Gather X, Take X to Y, Cindy/Candor/Rossy Battle, End Game (illia), Mini-game or other.
From what I've learned:
1) Script needs clean up
2) Scripts need a complete framework for questing
3) Base functions can be made for the simpler quest types
4) Adding more Quests offsets the problem with the grind
5) Encourages more levels of experience to help
6) Allows Dialogue to be developed Separately with ease.
Note: If we could use something like https://github.com/tweecode/twine on TMW-Script writers/scripters/collaborators would have an easier time.
Wiki/ContentDisplay/ContentSearch
Currently all the differences in the above code make it harder to dynamically generate information in a human (player, not dev) standpoint.
The scripts used prior were outdated and even after running them and fixing them I was still not satisfied.
Often what is on the wiki falls behind if not directly related to a quest, item thats important to players. Even then what the player see and what the code says confuses them. For Example, XP on the https://wiki.themanaworld.org/index.php ... _Reference displays the base XP before the mutations. It's silly we parse it instead of possibly either using a github page and referencing it in on that page or finding a plugin to read the data from git itself and making wiki templates as the items, mobs, etc. will still need a Mythos to them.
The less hand tinkering writers have to spend creating uniform pages the more they can write. It's why it's a shame this page https://wiki.themanaworld.org/index.php/Dev:Main doesn't have a wiki section.
TMW Governance
http://en.wikipedia.org/wiki/Cooperative
Why we pick anything other than a Co-operative operating style baffles me.
As a Content Dev I'm kinda amazed at how I feel into it. We elect Gms but not the Devs. Lead Positions would need to be filled so I'd say anyone could run for a position and we could let the population vote based on X,Y,Z criteria. Making sure each candidate understands it's a year long commitment and a new election can occur if they abdicate the position.
and the Road goes on and on...
Fix these things and we will have an awesome game. Until then, I'll bump this when we are back at this point again. Where adding new content only increases the number of problems we have later.
As I've said this all before.
viewtopic.php?f=2&t=17834