OK, how do I get started?

[DavidMedley]

I looked around briefly but didn't see a guide for contributing code. Here's relevant info to my situation:
- I was a professional software developer for 17 years (Perl, Flash, and core web dev tools & languages)
- I've never contributed to open-source, only proprietary code
- I'm mainly interested in gameplay upgrades, which are often very subjective. I don't want to run an unnecessarily high risk of rejection / wasted effort, hence this post seeking advice. I'll happily throw in minor fixes and documentation, too, when I see a need.
- I am presently working an a completely original CRPG system. I might put much of my system into an open-source roguelike as an intermediate step to the full game. That's not what I'm posting about here, but that's my long-term goal.
- My short-term goals as far as code contribution are to help improve the game(s), gain a bit of recognition in the community, gain experience tinkering with game mechanics.
- I'm most familiar with Vanilla, but also quite intrigued by Sil. I've played some Frog, but that seems like too much of a morass for me to chip in. I've also played some Dungeon Crawl and ToME, but I don't foresee contributing to those. Haven't tried ADOM yet. So mainly Vanilla and Sil, I think.

Thanks in advance for the help!

[DavidMedley]

Examples

I have posted a lot of helpful (?) suggestions in the two months since signing up. If you are curious about what kinds of improvements (??) I would try to make, you can look through my posts. That's probably where I'll start, depending on the process feedback I get here.

[Pete Mack]

The code is on github. Nick will accept reasonable code changes posted as requests there.

github.com/angband/angband

[Nick]

Briefly, what Pete said. I'm very occupied by other stuff at the moment, but will try to answer a bit more fully when I get the chance.

[DavidMedley]

The code is on github.

Yep, I am familiar! Maybe the best way to start is to make some non-controversial improvements.

[Gwarl]

The best way to start is undoubtedly to make your own variant.

Given your experience with web tools, and your initiative in compiling spreadsheet info, I'd invite you to help me in creating an online wiki - something I plan to get around to starting on myself, perhaps this month but there are gains to be made in parsing datafiles to automatically populate information about various version and variants..

But given that you would like to actually work on a game, giving yourself free reign to do it in your own variant is doubtless the most expedient course. I am happy to add new variants to angband live, game development is more rewarding when you can engage with your audience.

[DavidMedley]

I don't want to create and maintain my own variant, tho. I don't want Medband or whatever. I'd like to get some improvements into the mainstream that will hopefully live on. I could try making a variant but if most of the changes didn't get merged back in then I'd feel like it was a waste. But maybe I'm not understanding your point.

[DavidMedley]

Possible Vanilla Projects

Here are some possible areas I'd like to work on in Vanilla. I'm posting this list to get some feedback before attempting anything. Some of these sub-items are not crucial, I'm just noting them here because I posted about them previously. These are things I believe I could realistically take on, make improvements that the majority of players would really like, and that I think I would enjoy working on.

Loot system (more about that next post)

Refining Necromancer

• Would be so fun to be somewhat responsible for this new class!
• Lots of improvements to the way their spells are annotated
• A couple spells need reworking, especially the highest level ones
• Vampire Strike on creeping adamantite coins
• Vampire Strike "target is shielded"
• Vampire Form Discrepancy

Few Paladin improvements

• Make Protection from Evil usable at all levels but not OP
• Improve messaging on Smite Evil and others

Others

• Get rid of 18/xxx notation
• Contribute to documentation/wiki
• projection.txt header documentation
• Semi-persistent levels
• Reorganize monster recall
• Black Market misnomer

[Gwarl]

I will just say that a new face turning up and making gameplay changes directly to mainline angband is unprecedented as far as I'm aware. Not to say it won't happen but these goals:

gain a bit of recognition in the community, gain experience tinkering with game mechanics.

are usually accomplished via variant-creation, and invitation to work directly on vanilla probably comes later, built on the back of that experience and recognition.

Again you're better off waiting for nick's answer, but from inside the community unwillingness to make one's own variant seems odd.

[Gwarl]

• Semi-persistent levels

I could write whole essays on why non-persistence of levels makes angband what it is. It's the pivot on which the whole rhythm and style of the game and all of its many children turn.

Uncontroversial gameplay changes will be slim pickings in a game of this longevity.

[Pete Mack]

His list isn't bad, per se. Some of those changes are modest UI changes or improving measagw clarity, and he already said he will stsrt with those. I agree that monster lore is not easy to read as it is now--a briefer format at least as an option would help. For weapons and armor, it'd be particularly valuable on smaller screens, and in charactet dumps.

I will point out that I never made a variant, but contributed a few thousand lines back when Takkaria first took over, to improve UI and make a first stab at rationalizing the code (along with the Sidwells and pelpel.) V 3.0.9 was the result.

[DavidMedley]

Loot System

I came up with a system for easily managing loot in my yet-untitled CRPG. I think it's something I could port right into Angband and improve quality of life. Basically, it breaks down anything that's valuable but not practical adventuring gear into broad loot categories that step up in value by a factor of 4 or 5. From my notes:

This is a system for describing valuable items that are of no use to the adventurers other than to sell them. An arrow, a potion of healing and a vorpal longsword +5 would all fit on this table somewhere, but would be described individually instead of using this system because they have use on the adventure. A pile of gold coins, on the other hand, is only useful once the adventure is over. Using this system you can describe the loot adventurers get in very general terms. If their carrying capacity is limited then it's easy to determine what they take and what they leave behind. In addition, Objects which might have value to some adventurers, like a magical sword, but don't have value to the character(s) that actually find it, like a wizard who already has a weapon that suits him better, those things can be de-itemized and placed into this system instead.

For Angband, first we'll leave all money in town. Next, all the precious metals and stones you find in the dungeon have some weight - but not a lot. We'll give 1 inventory slot to this generic "loot" per 20 lbs or so - same as Iron Shots in your quiver. Anything you find in the dungeon that you don't want to use you can "loot" or "put in loot" or "scrap" and it'll be broken down into the 2 closest categories of loot. If you get overburdened it'll be easy to dump the lowest value loot first, and we can also put in automatic options, too. When you return to town, you'll automatically fence all your loot. You can sell individual items to shops if you want to have the option to purchase them later or need them ID'ed or recharged, but if you just want the fair value you can loot it anywhere in the town and it'll be sold automatically. Alternatively, players could be required to go to the black market to fence their loot, but they'd get a fair price.

I'd also like to add a bit of thematic flair where Rangers can skin animals, and if that goes over then we can add more stuff like that for other classes.

Obviously this causes some shockwaves in other systems... monsters that pick your pocket will have to change. It'll shine a light on the value of stuff, too, which will be a pain in the ass but good long term. Like, the value of a Ring of Free action, per pound, is much higher than many artifacts. But I think it can be a great system, especially for games like Frog where managing loot is such a hassle.

I've put up two new tabs in my public Angband spreadsheet. One is "Loot System" which is copied straight from my game development sheets. While the loot table found here is not finalized, it's pretty good. You can also see many of the sources I've used to form these values. The second tab is "Selected Items" where I took 40 items (at time of writing) from object.txt and artifact.txt. I ranked them by value density and then calculated how much "loot" you'd get for each item. (REALLY rough normalizing factor used.) E.g., a 2.2 lb torch gives 2 lbs of "Necessities" and .2 lbs of "Low Quality Equipment." 0.5 lb Scroll of Word of Recall gives 0.2 lbs "High Quality Equipment" and 0.3 lbs "Treasure." 13 lb Elvagil (the lowest value density artifact in the sheet) gives 10.6 lbs of Silver and 2.4 lbs of "Precious Treasure" which is a poorly named category between Silver and Gold. The One Ring is off the top of my chart at the moment.

This system, like everything in my game, will be released under a CC BY-SA license (or something similar). If somehow you want to use it soon for a published work I'll get that licensing part figured out asap.

https://docs.google.com/spreadsheets/d/1nAVk2AodMHloqhLBzZm7p_s4mne3My4Kyo9wODJWAFQ/edit?usp=sharing

[DavidMedley]

I will just say that a new face turning up and making gameplay changes directly to mainline angband is unprecedented as far as I'm aware.

OK. I'm new. I'm asking around. I'm not at all trying to call the shots. I'm definitely not expecting to come in out of the blue and change the fundamentals of Angband by fiat.

[DavidMedley]

I could write whole essays on why non-persistence of levels makes angband what it is. It's the pivot on which the whole rhythm and style of the game and all of its many children turn.

For sure! There's already a "Persistent Levels (experimental)" birth option. That's where I assumed I would work on my ideas for semi-persistent levels, if the feedback I got made me want to pursue this.

[Gwarl]

His list isn't bad, per se. Some of those changes are modest UI changes or improving measagw clarity, and he already said he will stsrt with those. I agree that monster lore is not easy to read as it is now--a briefer format at least as an option would help. For weapons and armor, it'd be particularly valuable on smaller screens, and in charactet dumps.

I will point out that I never made a variant, but contributed a few thousand lines back when Takkaria first took over, to improve UI and make a first stab at rationalizing the code (along with the Sidwells and pelpel.) V 3.0.9 was the result.

Yes, UI and code improvements have traditionally been accepted by all and sundry. But the configuration of spellbooks and the spells avaiable to the casting classes in the 3.x era was field tested in GSNband before being adopted by vanilla.

I'm seeming argumentative at this point, perhaps, I just wanted to clarify.