Our community module

[Rectifier]

Thought of this when I woke up today.

Its about Backgrounds. I was reminded of level 1 feats in DnD 3.0/3.5. In that setting, backgrounds were passive, simple, yet very powerful the entire game. Example: The "Luck of Heroes" feat gives +1 all saves and +1 armor class; no other defensive feat even comes close to this until Improved Toughness which has a prerequisite feat and stat requirement, whereas the former has no requirement other than it *has* to be taken at level 1.

In any case, I believe our module could use this approach on Backgrounds.

At level 1, the player is given a choice of backgrounds, with the first option being 'No Background'. The other backgrounds would be something simple like, "Blaster Master - +1 precision, attack, and damage on laser weapons." The 'No Background' would have a benefit as well, such as +5% experience gain.

Furthermore, each Background, while not ignoring any talents, would have "Recommended Talents" listed first, with the others right afterwards. The 'No Background' would have no bias.

This way we can create groups of talents that favor synergy without sacrificing adaptability, like a variety of ranged weapon talents our Blaster Master would favor nicely moved to the top of his talent choices.

Speaking of talent groups, it might behoove us to group talents into categories like Offense, Defense, and Utility which then could be sub-divided into Long Range, Short Range, Melee, and Self.
Examples of Each Type from Tome:
Berserk would be Offense/Self, Heal Defense/Self, Grace of the Eternals Utility/Self.
Flurry would be Offense/Melee, a cursory glance tells me Defense/Melee can be considered Defense/Self, and Dirty Fighting Utility/Melee.
Dominate would be Offense/Short Range, Bathe in Light Defense/Short Range, Infusion:Sun Utility/Short Range.
Flame would be Offense/Long Range, Displacement Shield Defense/Long Range, and Rune: Teleport Utility/Long Range

Tell me what ya think.

[darkgod]

Sounds fun

More code !

[tiger_eye]

Hey, I'd like to offer up a new field of vision and line of sight algorithm to add more tactical depth to the community module. It's a simplification of a generalized FoV algorithm that I have, and it is designed to work seamlessly with a modified bresenham line to use for targeting. Since we need a way to refer to it, let's call it "Savvy FoV". Savvy FoV has Symmetry and Variable permissiveness (SaV --> savvy), as well as no trick shots.

I believe symmetric targeting (if you can fire at an enemy, it can fire back at you) adds a degree of fairness to a game. The aspect that can make it less fun, however, is stepping into the line of fire of an enemy that you didn't see and it shooting you first (asymmetric targeting also has this issue, and the player can both abuse and be abused by it--being pinned by a master skeleton archer you can't see really sucks, but it's awesome if you're the one who pins an enemy that can't see you). So, targeting is symmetric, but vision need not be symmetric. For example, perhaps characters in the community module can attain a biological or cybernetic enhancement to vision to increase the amount they can see around objects. In the art of war, the side with the best information gains an edge; similarly for roguelikes, the actor with the best vision has an advantage and can choose the best tactics.

There is one more thing I have written into Savvy FoV that I would like to propose: cover! Cover is not a foreign concept to grid-based fighting games, but I don't think it has been used much in roguelikes. A target actor in a grid is "under cover" if the line of sight from source actor to target tile is partially blocked. This means hanging around pillars offers a tactical advantage, because you will probably be under cover w.r.t. enemies (which is reverse from traditional LoS whereby being next to pillars blocks your sight and makes you susceptible to fire from enemies you can't see). Furthermore, there could be "cover" terrain like tall grass or an upended table that provides cover, but doesn't block sight or movement. The advantage for being under cover is open for development.

I'm nearly finished with developing Savvy FoV&LoS. Since I claim perfect symmetry, I really need to perform a thorough consistency check to make sure everything behaves properly. My next week is pretty busy, though, so I make no promises for timeliness . Cheers!

[darkgod]

Funky =

[lukep]

There is one more thing I have written into Savvy FoV that I would like to propose: cover! Cover is not a foreign concept to grid-based fighting games, but I don't think it has been used much in roguelikes. A target actor in a grid is "under cover" if the line of sight from source actor to target tile is partially blocked.

Cover would fit with my proposed systems for vision and accuracy quite easily.

For combat, it could simply reduce the effective size of the actor. For example, a humanoid (size 25) in 50% cover would have an effective size for hit calculations of 12-13. (or, an easier/more consistent way to code the same result: make the new miss size be [2 * old miss size + old hit size], doubling the target area)

For sight, it would reduce the detection value of the monster. For example, a monster that you had 50 detection on (full knowledge) would be reduced to 25 detection (monster name, HP %) when under 50% cover.

[edge2054]

I haven't read this thread in depth but I've skimmed it and read a bit so maybe this should be taken with a grain of salt.

Speaking as someone that's just now joining the discussion the gene-splicing idea sounds really cool to me and I think has potential to be expanded into an entire game premise. Cyberware, as much as I love my cyberpunk, has been done to death and I feel doesn't need to be included. I guess what I'm trying to get at is that I feel the thread is going the direction of kitchen sink syndrome. In other words, gene-splicing was a cool idea and I think you guys have started running with cyberware too because that's the next logical thing to add after that.

As to gene-splicing, I realize you guys are still very much in the mechanics phase but I think it's also important to consider premise. You have a what already but a why could help to solidify those ideas into something fun and help guide future decisions so they fit and make sense.

So here's my idea on why. The character could be sort of a genetic borg. He (or she) could belong to a race that's seeking genetic perfection by splicing DNA from other alien species. I'm sure Darkgod remembers the Far Scape episode about that. Or he could be sort of a collector of genetic subtypes, which would fit with the modular talent design yufra proposed. Basically what I'm getting at is if you know the what the why can help guide and focus your ideas. Who is the character, what's he doing, why is he doing it? We know the what but we still don't know the who or the why. These three things can work together to flesh each other out.

Take the mad scientist approach for example. The genetic collector stealing DNA from other life forms to put them to his own ends. This is the guy that would also have lasers and other technological gizmos because he's a scientist after all.

But if you go the other route, the gentic borg, as in you will be assimilated borg, (he could even be a shapeshifter so he's limited to only manifesting so many splices at a time), he may be a lot more reliant on his own abilities. The entire gun play system could revolve around the NPCs shooting at the player with lasers and pistols while the PC shoots flames out of his mouth or spikes from his tail or whatever.

But yeah, grain of salt and all of that. Just had a couple of ideas of who and why and figured I'd drop them into the thread.

edit: 2000th post.. Maybe I should make a new account

[yufra]

Congratulations edge, and thanks for the post. We miss you...

I agree that the premise will drive the correct mechanics, and trying to make a premise fit mechanics can result in some awkwardness. To that end I want to remind everyone that I am scheduling an IRC meeting for all interested in this project for next Saturday, January 7th. I have set up a Doodle scheduling poll here: http://www.doodle.com/ffackuscigzfgfmf

I tried to offer a variety of options that I knew I would be available for, and they span at least the early evening to way-too-early/late morning for Eastern Europe so I think we should be able to include the variety of time zones. If you think a different option should be added, please just let me know.

Why meet? Well I think realtime communication can help us push through some of these decisions on setting and mechanics. We have had a lot of brainstorming going on in the thread, which has given me time to play around with some engine modifications that are nearly done. I am creating an agenda for the IRC meeting on the wiki here: https://github.com/mfajer/Gene-Revoluti ... 01-07-2012

Feel free to add to that agenda, although I will edit/cull it just before the meeting to make it manageable. Cheers!

[bricks]

Resource usage is something that still seems really vague. Defining the premise will help shore up what these resources mean, but from a gameplay perspective there is some that can be said. Here are a few ideas, some my own, some picked from the discussion. They aren't necessarily mutually exclusive.

Single-resource system
Apart from health, there is a single "energy" meter that is used for skills. It recovers over time.

• Easiest to interpret
• Only offers a small diversity of play styles

Multiple primary resources
Two (or more) resources. Skills consume one or the other, rarely both.

• Still very simple
• Makes certain skills more distinct and groups them naturally

Low-throughput resources
Resource pools are large but recover slowly.

• Tactical choice between exhausting a resource or using it up slowly
• Players can "spam" a skill unless it has a cooldown
• Punishes players for unavoidably long fights
• Lots of "resting"
• Very difficult to balance for NPCs (due to their short lifespans)

High-throughput resources
Resource pools are small but recover quickly.

• Players are encouraged to constantly use resources
• "Spamming" is largely limited
• Small distinction between short and long fights
• Little resting

Failure-based resources
Using a skill causes the resource to increase. A larger pool means a greater rate of failure. Decreases at a moderate rate over time.

• Places a soft limit on skill "spamming"
• Luck-dependent

Diminishing-strength resources
Using a skill causes the resource to increase. A larger pool means reduced skill effectiveness. Decreases at a moderate rate over time.

• Also places a soft limit on skill "spamming"
• Effects are deterministic
• Some skills are harder to scale back, require greater design consideration (i.e; what is "half" of a knockback effect?)

My ideal setup
Three resources. Bioenergy, a high-throughput resource. Fidelity, a diminishing-strength resource for genetics skills. Sync, a diminishing-strength resource for cybernetics skills. (Note that these work backwards and have finite limits, unlike ToME's Equilibrium/Paradox). Skills would not have cooldowns. Most actions would use resources, even bump attacks (though those would be much smaller than regular skills). Sustained skills could work a few ways; some may drain bioenergy constantly or based on a specific action, or they could decrease the attainable maximum of the appropriate diminishing-strength resource (more sustains = weaker active skills). Every part would have a Fidelity or Sync contribution to the max. The larger the pool, the less you suffer from the effects of low Sync/Fidelity, so there would be an inherent benefit to stacking one type of part or the other (and no real penalty for using a hybrid). Apart from extremely poor effectiveness, there would be no penalty or restriction on using skills at minimal Sync/Fidelity, so long as you had the requisite Bioenergy.

[yufra]

Bricks, that is a fantastic summary and should really help that section of the discussion along. I think it would be nice to have two high throughput resources, one for each side (cybernetic and biological) with methods of interconverting between them. I don't envision the inverconversion following the Anorithil model where you actively move positive energy into negative energy, but more sustained effects that would steal either regeneration or maximum pool size from one resource and move it to the other. Think "Biocapacitors" or "Nano proton pumps".

[Dekar]

The ideal setup sounds really nice.

[bricks]

Thanks, Yufra and Dekar! I think one or two primary resources would both work fine; I avoided the latter because 1) four resources seems like a lot to me and 2) with the Sync/Fidelity split, there's already a distinction between the two part categories. Talents and parts that exchange between the two (passively or actively) would add a lot of depth, so I can see reason for it. Of course, the same could be applied to Sync/Fidelity (damage one to repair the other).

I have a few loose thoughts on Sync/Fidelity. "Reduced effectiveness" can be interpreted in a lot of ways. Take a laser blast attack, for example. Poor Sync could cause damage reduction, accuracy penalties, bioenergy cost increase, or a chance of failure. Hypothetically you could have four versions of the talent, each "failing" in a different fashion. High-risk players might opt for talents that don't lose damage with poor Sync, instead utilizing those that suffer accuracy penalties or plain failures. Low-risk players may prefer to take the damage reduction version of the talent, since it's more reliable. Or, sustained talents/active effects/part egos could modify how actions interact with Sync/Fidelity.

If talents are designed with a chance to fail, I think the failures should be more specific to the talent/part. Guns jamming are a simple example; infravision systems could perhaps "fail" in the sense that they cause blinding or afterimages. Failures could also be more neutral or situationally practical, like ToME's anomalies. I could totally see a teleporter "failure" resulting in a Fearscape-style scenario.

[yufra]

One of the ways to reduce effectiveness that you did not mention is to limit the "level" of the genes/cyberpart. The level of a specific gene/cyberpart would slowly increase towards the global fidelity/sync through use or time. The level would also be limited to the value of fidelity/sync, so talents could be lost if the fidelity/sync is lowered sufficiently. I think this is more aggressive than the reduced effectiveness, but still more deterministic than failure rates. Thoughts?

[yufra]

Oh, and the github code requires the attached patch to be applied to the TE4 engine code before it will run.

[bricks]

Oh, that's very interesting. And it makes logical sense. I'm not sure if I'd like all items to have that mechanic, but there could be some extremely powerful attacks that essentially disable themselves (which would be a hidden cooldown). In fact, the player could go for a high-throughput Sync/Fidelity build, with one "tactical nuke" part and the rest geared towards rapid recovery of Sync/Fidelity. Perhaps some story artifacts would work this way; extremely powerful, hard to regularly used, and occasionally very important for progressing the story.

Thanks for the patch. I've begun working on some inventory code, which currently has amounted to marginally more than copying the instructions off the wiki (big thanks to whomever wrote that up). I'll probably play with learning talents through equipping items today, and if that gets anywhere, I may even get a basic melee part put together.

Edit: Oh my god why have I never used GitHub this is so cool.

[yufra]

The masses have spoken and we have a meeting time: 6:00pm UTC/GMT, 12:00pm Central Standard Time, 7:00pm Central European Time, etc.

We will be meeting on the #tome IRC channel (details: http://forums.te4.org/viewtopic.php?f=38&t=23463).

The agenda will still be tweaked around a bit, and hopefully we can cover the majority of it to get some consolidated vision. You can find the agenda here: https://github.com/mfajer/Gene-Revoluti ... 01-07-2012

Cheers!