Shandalar Clone

[bladescape]

Admonition Angel:W Exile/Angel Deck.

Code: Select all

BasicLandPercentage=34

[Group MaxCnt=4 Percentage=32]
Admonition Angel
Iona's Judgement
Swords to Plowshares
Path To Exile
Journey to Nowhere
Oblivion Ring
[/Group]

[Group MaxCnt=34 Percentage=28]
Serra Avenger
Serra Angel
Thunder Spirit
Suntail Hawk
Elite Vanguard
Battlegrace Angel
[/Group]

[Group MaxCnt=2 Percentage=5]
Platinum Angel
Elspeth, Knight-Errant
Leonin Sun Standard
[/Group]

End

[DennisBergkamp]

Hmm, as opposed to generating general themes for each color (and color combinations), I'll just use the current deck generator code.
Then as more themes are generated, more specific archetypes can be used... I just finished most of the code that decides what creature is played against exactly for the random encounters.

[friarsol]

Dennis, how close do you think you are to submitting your graphic quest code? Even if it's incomplete, it may be better to have it in the SVN while you can continue to work on it at the very least so it's backed up.

[DennisBergkamp]

Pretty close actually, it still needs a bunch of cleanups... but I could even do those after committing.

[DennisBergkamp]

Actually, I'm kind of scared to commit (it will add dirty code all over the place and *might* even break things on our beloved SVN ), but I will just release the code with the next test version of this map.

Do we have a 4 or 5 color deck generator (using the new deck generation algorithm) by the way?

[Rob Cashwalker]

Not yet. I'm getting back to that stuff soon..

How much of the Map version affects the core of Forge? I know the map stuff is a separate executable, that launches Forge matches.

[DennisBergkamp]

Ok, no probs, for now we could just use the old ones.

I haven't added that much to the core of Forge, but I added code to (at least) these classes:

AllZone
CardListUtil
EndOfTurn (<--- not sure why)
GameActionUtil (can't remember why either)
GuiDisplay3 (??)
GuiDisplay4 (??)
Gui_WinLose
ImageCache
QuestData_BoosterPack
QuestData
ReadBoosterPack
ReadQuest_Assignment
WinLose

Not too bad, I suppose.

[DennisBergkamp]

Had a few hours of spare time to work on some of the "random encounter" code.
This works now:

1. Once a random encounter triggers, what kind of colors the opposing creature will be gets calculated by looking at the surrounding tiles (closer tiles have more "weight").
2. As soon as the creature card gets selected based on what kind of colors, first an attempt is made at finding a matching theme for the creature (this could use a lot of work - currently it will just go through all of the themes and attempt to match by the name of the creature, then if nothing has been found, it will try creature types).
3. If no matching theme was found, generate a deck based on the color(s) of the creature.

Rob mentioned something about meta data for the themes... this is probably the best way to go. We could just have some kind of keywords that map a creature to a theme (creature type, name, etc.).
What do you guys think this meta data in the theme files should look like?

[Sloth]

1. Once a random encounter triggers, what kind of colors the opposing creature will be gets calculated by looking at the surrounding tiles (closer tiles have more "weight").
2. As soon as the creature card gets selected based on what kind of colors, first an attempt is made at finding a matching theme for the creature (this could use a lot of work - currently it will just go through all of the themes and attempt to match by the name of the creature, then if nothing has been found, it will try creature types).
3. If no matching theme was found, generate a deck based on the color(s) of the creature.

If this doesn't slow down the game, I don't think we need meta data to choose a theme. It was bothersome to modify three files whenever I added a new quest deck.

[DennisBergkamp]

Hmm, the meta data could be contained within the theme file though.

[Sloth]

Hmm, the meta data could be contained within the theme file though.

That would be ok then. What should it contain? A list of creatures that would have a chance to trigger this theme? A difficulty level?

[DennisBergkamp]

For a start, yes it would contain either a (small, hopefully) list of creature names, and/or a list of creature types. Not sure what else we could stuff in there...

[Jaedayr]

While waiting as patiently as I can for this I have been thinking of how to link creatures and themes together. This may or may not work so it is just ideas to kick around.

The first link is domain, or the terrain type where the encounter takes place. Each creature and each theme file would have one or more domains out of the five possible. The second link is keyword which would start with a defined list where every keyword is represented one or more times by a creature and one or more times in a theme.

For Example:

Code: Select all

Creature        Domain  Keywords
Elf             G       Elf, Warrior, Druid, Shaman
Druid           G       Druid
Beetle King     G       Beetle

Theme
Green           G       Elemental, Elf, Treefolk, Warrior, Druid, Shaman
Beetle          BG      Beetle
Generic Elf     G       Elf, Warrior, Druid, Shaman
Druid           G       Druid
Human Druid     GW      Druid


When an encounter is generated, the creature would be chosen based on the domain currently occupied or nearby, and then the creature domain and keyword(s) would be used to find an appropriate theme. Selection of both creature and theme could be random or weighted as desired.

Dennis Forgewalker is roaming through the forest one day and triggers an encounter. If it is a Beetle there is one theme that could be chosen. If it is an Elf, there are two themes and if it is a Druid there are four themes.

Keyword selection would be the critical area for this. Some creatures may have several themes possible while others would always play the same theme. Each theme could also be possibly made into three difficulties, an easy, medium and hard, with the same cards in each, just different groups and percentages. Difficulty would be determined from quest progress and added to the creature keywords for theme selection.

When a new theme is added, if it uses existing keywords it would require no changes to any other files, just the addition of the theme file to the proper folder. If a new creature is added using existing keywords then no other changes would be required. Only when adding new keywords would changes be needed to both the creature and themes.

[DennisBergkamp]

Yup, this would be a good way of doing it, and it would allow for multiple themes to be picked.

[friarsol]

Hooray for reviving Dead Topics.

Dennis, I'm not sure if you've noticed that we've moved off Google Code SVN to Git in the last several days. The great thing about Git is that it allows you to clone the Master Source Control and have your own safe sandbox. If you can get your Adventure code merged into everything you can safely commit code to your Clone without so you don't have to worry about merging a bazillion changes at a later date.

Then when you are confident the merges won't break anything it's relatively straightforward to Push your changes Upstream to the Master Repository.

I'd recommend doing this as soon as you can, as the longer you wait the more changes between the final SVN version and the current GIT version you'll have to wade through. Even better than that once you have your personal repository up, other people can grab a Read Only version of the Adventure Mode, allowing additional testing and suggestions to the code base.