10/16/2019
(Toady One) Interrogation work has continued as promised with some additions to the conversation system. The first part was to add some new per-character variables to every conversation - how much a person wants to stay, whether they are pleased with things, and how confident/dominant/composed they are feeling, basically (this is more or less experimenting with valence/arousal/dominance emotional theory stuff, to see how it complements the myriad circumstance-based emotions we use already.) The second was to add tones/tactics attached to utterances, so that a skill like 'intimidate' and a skill roll can piggyback from the menu selection to the character's 'move' through to their utterance in the conversation and finally to a reaction in the listener.

Those two additions allow us to set up an interrogation game that's vaguely analogous to existing aimed attacks. The subject's relationship with you and their personality (and their skills, etc.) determine how their transient conversation attitudes are affected (successful intimidation breaks their composure, for example), and this state determines whether they answer your question. Each utterance gives some feedback on how it all went, on top of the responses themselves, and you don't need to use tactics on people that are already predisposed to reply. The game has a little preview window now that'll tell you about what to expect from a given question+tactic pair, based on how good you are at judging character (which relies on empathy, social awareness and intuition), similar to how aimed attacks show chances. In this way, characters balanced toward these skills and attributes can have a significant advantage in this area.

(Incidentally, most of this took place with dingo people that had moved into a test village, rather than the Manifestation, for reasons of convenience.)

There's more to do, as usual, but there's some hope it seems of having something that feels like an investigation in opposition to the villains, and we'll be expanding this out now until we nab enough of our goals to move on to the fort.

(Comment on this)

10/10/2019
(Toady One) As mentioned last time, we went down to the Roguelike Celebration! We didn't have a talk this year, but there were plenty of speakers that will be of interested to our players. This year's talks are available here, and here's a schedule you can use to navigate a day's video.

There wasn't much time left over, but in Dwarf land, we worked on the new interrogation conversations a bit. We should have something to report on our conversation experiments for next time.

(Comment on this)

10/02/2019
(Toady One) Okay, we've started interviews and interrogation! It seemed like a fair entry point to investigations, since you'll be confronting villains, and the conversations here need to work, regardless of other evidence. Naturally, our debugging friend, the Manifestation, made a return, wandering towns and taverns, clotheslessly peppering the criminal underworld with various queries. They don't currently push back or refuse to answer questions, but that's up next.

We're going to experiment with some new conversation modifiers, like the ability to ask a question while also using 'intimidation' or 'persuasion' or 'flattery' as the principle conversation skill. We'll also be using the new relationship variables ('loyalty', 'trust/distrust', 'fear', 'love/hate', and 'respect') as well as the existing reputation and personality information. Hopefully we can at least get a basic feel of a back-and-forth going during these important moments in the adventure. The ability to get somebody to yield through the combat system is always available to such characters, and somebody in such a scary situation is likely to balance that fear vs. their fear of and loyalty to their master, and this will go in your favor oftentimes, but it'll also lead to quicker retaliation from the villain or organization at large. Using other conversation styles beyond intimidation and violence should present a viable safer option in many cases - at least this is what the plan is!

The Roguelike Celebration is this weekend, so this coming week's work might be a little light, but it is always wholesome to meet with the Roguelike people, and some DF players should be there as well! Zach and I will both be attending.

(Comment on this)

10/01/2019
(Toady One) To start this month, we have

the report

as well as

the Future of the Fortress

.

(Comment on this)

09/25/2019
(Toady One) Alright, I did the last two things I wanted to do in this network rewrite: having plots that last longer (using the existing civ-based spies as a model), and allowing plots to pass even further down the criminal organization networks (so that more evidence will be created while making the leaves of the org chart more active.) This has been good, and I think we're ready for investigation now! Hopefully! I also discovered a few problems with my org chart displayer, which is another way to make the networks look even more rich, ha ha (it wasn't printing certain assets.) This has been a long detour, but we should be to the heart of the adventure mode update now. Once that all works, fun for forts.

(Comment on this)

09/18/2019
(Toady One) This week I mostly pushed forward the ideas from last time. Plots can propagate out into dedicated criminal organizations from the non-criminal position holders (often through intermediaries), and criminal organizations can also expand out into other cities, forming branches much as the merchant companies do, where they then try to muscle out and subordinate local groups. These together have led to even more pleasantly rich and geographically-varied networks. We'll take a few more steps along these lines before trying to implement player-led investigations, but we're almost ready to give them a shot.

(Comment on this)

09/11/2019
(Toady One) I am mostly recovered from the traveling and am back to work. As mentioned in the log before the travels, it turned out that we needed more work to link up networks and to make sure every link was generating some evidence. As stated there, we decided to focus on criminal organizations, and the linkages are much better now! For villains without brighter ideas, doing some petty crime with a few like-minded individuals is a start, and then these groups can fuse and otherwise associate, with various skimming and tribute and so forth as some of them grow more powerful. This allows the standard anti-bandit/criminal quests to lead into evidence network crawling, as the most successful groups can draw back to a more villainous status (ie they need discovering rather than generating direct quests.) The next steps will include linking the more important, position-holding villains to these groups in a few ways, and also bringing in the bandits and perhaps certain hearth lords into the mix as criminal organization analogues. I'm much happier with how the network graphs look now, but we'll need to see an actual investigation to know if it's sufficient.

The individual network links will also need to be strengthened by adding more long-term jobs for the important villains; the big assassination/etc. plots resolve too quickly, and so feel a bit like old news when you start adventuring. This can be solved by treating them a little bit more like the existing civ-based agents, with some work to be done over some in-game months before any job can be pulled off, and by making position-holder corruption being easier to accomplish when some relationship groundwork has been laid. After this tweaking, we should have an evidence-rich environment that doesn't feel as artificially urgent.

Here's a recording of the live play of DF at PAX with Nate Crowley of RockPaperShotgun and Victoria Tran of Kitfox, along with Zach and I. And for people that haven't seen the systems panel at PAX, here's a recording on Twitch.

(Comment on this)

09/04/2019
(Toady One) And here's the delayed Future of the Fortress:

part 1

,

part 2

. I'm almost back together and will be programming again starting tomorrow.

(Comment on this)

09/01/2019
(Toady One) Here's

the report.

PAX is underway. We've met a bunch of people at the Kitfox booth so far, and we'll also be there tomorrow from 3-6 (#6120). People that haven't seen the original Steam trailer (the ten second transformation from ASCII to graphics, as well as some screenshots) can be quite surprised by it! After five days of what turned out to be a very intense academic conference (I must have seen 40-50 paper presentations), Future of the Fortress will be delayed a few days, as expected. I feel like I could sleep for many years currently, but one more day of PAX first, then some sleep, and I should be back to adventure mode investigations!

(Comment on this)

08/21/2019
(Toady One) We started by writing up the adventure log 'intrigue' tab where most evidence will be stored. This is divided into actors, organizations and plots. The organization view comes complete with graph visualizations showing names and roles and associations, like those string conspiracy boards, and you'll be able to watch your view of a given organization fill in over time as you collect information, interview witnesses, and interrogate suspects.

The new info screen has highlighted several defects in the plots and villainous behavior. We'll have to do some additional work to get more organizations merged or at least linked (there are too many isolated failed organizations with one person bossing one other person around for decades), and to make sure that more of the lower level members have their fingers in some evidence-creating cookie jar. Currently, too many of the low-level people try one gig and otherwise sit quiet for years. Our planned solution is to use the existing criminal organizations and bandit gangs; we put that off previously, but it'll have to be done now in some abridged fashion that addresses these issues. At the same time, it was really cool to see a 25 member gang all laid out, from the leader hidden in an abandoned monastery, to their lieutenant goblin childhood friend that actually did all the organizing, to the low-level members, to various thieves and snatchers they'd hired for jobs down at the leaves, along with one embezzler and a highly-placed asset connected to the tree through their handler.

I realized after the last dev log that I should probably clarify what I meant by FDG-PAX extravaganza. People found some intriguing acronyms. For us, this means the Foundations of Digital Games conference down in San Luis Obispo at the end of the month, and the Penny Arcade Expo in Seattle overlapping with it. I'll be giving a keynote down at FDG, and generally talking to the academic people about systems and emergent narrative and whatever else. At PAX, I'll be on a systems panel (which should be livestreamed), doing a Rockpapershotgun stream with a live DF game, and Zach and I will also be there bravely staffing our own tiny DF section of the Kitfox booth on Sunday and Monday.

So yeah, no dev log next week, as there won't be any DF work until the month changes. Hopefully I can get the Future of the Fortress up on time between the events. It'll be exciting to get the first evidence-investigation conversations in the game soon.

(Comment on this)

08/15/2019
(Toady One) Ha, the prediction we'd be easing into adventure mode villainy was correct. Work was mostly on paper this week, as we reconciled various knowledge and investigatory/evidence data structures with what we want them to be able to accomplish. Knowledge was already stored in partial fragments due to how interrogation works in world generation, but we need to tease apart every little piece even more finely if we want you to be able to feel like you're actually piecing clues together (whether that's in adventure mode, or in the fort mode justice system.) Real-life family stuff came up as well (everything's fine, now.) But we should have some more tangible action to report next week before the whole FDG-PAX extravaganza begins.

(Comment on this)

08/07/2019
(Toady One) The necessary map work I was set to do this week is now mostly completed. That included bandit and mercenary forts, merchant company counting houses, guildhalls, monasteries, and improved necromancer towers. The necros still need a bit of work, but overall it's in good shape. As usual, actually creating the maps and populating them brought various bugs and nonsense to the surface, some of it very old. I fixed the problems that I found, with the occasional collapsed roof and infinite pit to the underworld along the way.

We've also been convinced by this process that doing the map rewrite as part of the Big Wait is necessary. For instance, I was able to get guildhalls and counting houses to appear in non-player dwarf forts, but it was an iffy thing, and there are bandaids all over the place. The inability to extend map code smoothly is going to inhibit a lot of what we want to do with magic and beyond, so we'll definitely be tackling that. We were leaning that way anyway, since planes and portals and multiple cameras (e.g. seeing off-site dwarf battles while running your fort) are all pretty cool, but now it feels compelled. There are a ton of benefits to this, but the Big Wait will indeed be Big, even if the map rewrite is most of what we do at first.

(As a reminder, the Big Wait is the development time for the first myth/magic release. This will happen after the Steam/itch release, and after any immediate improvements that follow Steam/itch - candidates are listed on dev, like improved sieges, adventurer mode medical improvements, etc. The length is unavoidable due to the interlinking of systems, but we'll try to make it as short as we can.)

We'll be easing into adventure mode villainy now! Then to adv investigations, and fort mode. How we handle evidence and interrogation and so forth in adventure mode will directly inform how the player deals with villains in fort mode.

(Comment on this)

08/01/2019
(Toady One) August's report . The Future of the Fortress: part 1 , part 2 .

As we finished off the month, which is mostly an administrative period, there was also a last bit of tangentry and excitement over in necro and demon land. Summons were expanded to include larger nightmarish beings. They last for a short time and can't be called often, but they are trouble. Raising intelligent undead was expanded to include ghosts. These can act as plotters and agents (you can put them down with their names, and likely that'll also come up when we get to interrogating them about their role in plots.) Both of these should also contribute to fun in the fort when we arrive at that part.

Finally, certain necromancers and certain demons can also magically experiment on or otherwise corruptly transform the citizens and livestock of cities that they capture. This leads to a variety of humanoids and quadrupeds and others (like little failed experiment winged blobs), some of which can escape into the wilderness and perhaps even rarely reintegrate into society (and thereby possibly become playable in adv mode and available as fort travelers and migrants.) Collections of citizens can also be amalgamated into monstrous giants.

The creature description includes the conditions under which the creature first appeared, and their creature name often includes the name of the necromancer or demon (e.g. humanoids called Tura's hands), and they are uncommon enough that I think it's a relatively safe step in terms of exposition and potential confusion. That is, while elves, goblins, animal people, etc. are easy enough for most players to parse, along with the occasional forgotten beast, we were worried about throwing tons of procedural beings in before myth generation can cohere them. With the experiments though, despite being random, they still occupy a controlled position that seems to work in the contexts where they appear.

I think the tangents are finally out of my system now, for the time being, and we can get these maps done at long last. These new additions have been very good for the villain theme, though, and it'll be wholesome to see them out in the worlds.

(Comment on this)

07/25/2019
(Toady One) I began work on the in-play maps of new locations, the last major bit to do before all the villainy comes to play. The new evil regions were first, making sure the ones that spread out from necromantic activity properly killed plants and so forth. Working on this led to a number of changes. Evil that spread now fades away over many years if there's no new activity in the region. The primordial evil areas don't fade. Certain demons can now spread evil from their towers (or disaster forts), and the properties of the evil areas are related to the spheres of the demon. That is, a blight demon might kill vegetation, while a deformity demon can spread evil creature populations and wormy eyeball grass and so forth. This was a natural segue into restricting bogeymen to certain primordial evil regions and to nightmare demons, which puts an end to that little years-long experiment in annoying people. To compensate, bogeymen have been given some new powers, but they are way more rare overall.

These considerations led (there is a lot of leading this time) to undead lieutenants being given a variety of new powers. Many of them are based on existing syndromes, like the ability to temporarily blind and so forth, but I added some new effects as well. For instance, some of these undead can raise a heavy fog (when outdoors), and others can propel people away from them. I also added a summon power, and new tags to make polymorph-type abilities more versatile. Demons with the knowledge of life and death (that is, death demons, mainly) will now use their lieutenant raising property. They don't animate zombies because animated zombies aren't friendly to their goblin and other buddies, but the lieutenants fit in just fine. I made sure the goblins don't chase them away for being unnaturally immortal, he he he.

Having escaped from all these tangents, I dutifully went to work on shrine maps. They are now distributed throughout sites based on their historical prevalence, with the dominant religions claiming e.g. large plots, while the minor religions might just have a single altar on a single curbside. Being full of tangents this week, that of course meant adding dice, for divination. With face sets (images, words, numerals or dots) and sides according to their shape (platonic or 'long die' prisms), and certain divination practices working with combinations of dice, and the ability to roll either one die or all the dice you are holding (which matters if the divination practice requires two dice, say.) And, um, the divinations doing things sometimes. I went to a shrine, rolled a tin icosahedral die and it was a bad roll and I was cursed to be a snowy owl for a week. So I flew up to a nearby rooftop and pondered tangents. But I will relax and finish my maps now. Working on worldgen villain plot details for months made an unwind like this overdue, I think. At least when I finally do get to tavern games and gambling, dice will be ready.

(Comment on this)

07/17/2019
(Toady One) I started with some character sheet issues this time around. To keep up with some promises from old Future of the Fortresses for the new adventurers, I allowed the purchase of item quality in chargen, and the selection of religion, including the organized ones, depending on what town you start in and its history. I also changed the 'hearthperson or not' choice to allow for a variety of regular professional backgrounds, which mostly don't do anything in adv mode yet, but they give free skills which'll allow you to have some useful skills if you decide to retire in or move to a fort. At some point we'll get to wider skill use in adventure mode, but not yet!

I also finished the tactical mode for parties of adventurers. Once you enter tactical mode, you can either control all of your characters manually, or designate a subset you want to control and let the rest be controlled like regular companions. You can change how each individual is controlled on the fly, which can be convenient if there's, for instance, a straggler you don't want to manually walk into the fray while everybody else is already in action, or an archer you don't want to keep popping back to (or vice versa with a frontline fighter.) If all of your manually-controlled characters are incapacitated, it'll give you a little announcement and pop you over to one of your automated ones. Now that the basic implementation is finished, there are a lot of different ways to add to it, naturally, but I'll wait for some feedback on this new system before going to far with it now.

Finally this week, I made a change to aquifers. Most of them will now run about 1/500th as slow as they used to, allowing you time to come up with a strategy to wall or channel or pump or whatever you like, while remaining a fairly immediate flooding danger. The old-style near-artesian aquifers are still in the game, but they make up about 5% of the total and are clearly marked on embark and in the site finder, so people can play around with the aquifer-busting methods that have been invented or just have an easy flowing water supply. The new slower aquifers can still be used for wells, but if overused, there is more danger now that they won't fill fast enough, in which case you just need to make an additional well or make the well excavation larger. All old forts loaded into the new version will have the old type of aquifer to avoid breaking existing fluid-mechanical setups, but the old worlds will have the new aquifer distribution for any future embarks.

(Comment on this)

07/10/2019
(Toady One) I finished up the basic features with pets and mounts. You can set the mount speed (which lets you see their gaits, like gallop, canter, trot, etc.), and party members will speed up to try to catch up with you. Of course, that doesn't mean anything if they don't use their own mounts, so they do that now. They will move to mount their own mountable pets after you do, and they will dismount when you dismount, if there isn't an ongoing combat or something like that drawing their immediate attention.

I also made sure that necromancer adventurers can ride undead mounts that they raise. This was flumoxed at first since the undead creatures don't have souls and so weren't taking the commands properly, but that's fixed now, though the nature of undead animation magic is still not pinned down (and won't be until much later.)

You can pet animals now. Necromancers can pet their zombie animals too, and the living can try too if they are foolish. Hmm, I think a necromancer might also be able to pet their zombie humanoids, due to how it does the detection.

Next up is the tactical party combat and other party issues for next week. Then either map updates or adventurer mode investigations and villainy.

(Comment on this)

07/04/2019
(Toady One) As a quick start for this month, I finished the other major plots with hireable moving pieces (sabotage, abduction, and artifact theft), up to their world generation level - sabotage, for instance, still just affects an abstract economic number, since there's nothing else for it to do until we get to fort mode, where it'll likely have a starring role which you'll all enjoy very much.

I'm finding the work a tad too abstract now that we're out of world gen, so I'm going to bring adventure mode into the mix now. This will start with completing mounts and tactical party combat to clear the plate, and perhaps the necessary map updates, and then we'll finally be ready to attach all of this plotting to an actual played game.

(Comment on this)

07/01/2019
(Toady One) Here's

the report

to start off July. And a short

Future of the Fortress

as well. I think the transition from world gen to detailing the same matter for post world gen has sparked less questions, but that's to be expected. Things should pick up again when we get to the actual play changes.

(Comment on this)

06/27/2019
(Toady One) This week I continued on with the in-play plot updates. The first thing I did was to get cutouts/handlers to work with assassins. Just as in world generation, the villains no longer need to hire assassins and other agents themselves. In the case where a villain or a handler is duty-bound and important, where a journey that might take several days would seem inappropriate, I've allowed them to send messages more abstractly, over the same period of time. These will still form the basis of evidence etc. during the investigation period. For villains and handlers that aren't so fortunate, they actually have to make the relevant journey. This will be particularly important once we get to fort mode, since this is where they'll be sneaking into your tavern and so forth. I also spent a bit of time with the other (non-assassination) plots, though the transfer there is not yet complete.

The supporting features from world generation have also needed to make the transition to play. In particular, I spent some time moving over the new cultural/population religious information that was stored in a more expedient form for world generation. I'll also have to do this with some of the new relationship data.

Finally, I took a first look back at the unfinished adventure mode stuff I'd left behind last year with parties and mounts, and fixed up some movement and announcement issues there to refresh my memory.

(Comment on this)

06/19/2019
(Toady One) I went to the Interplanetary Festival in Santa Fe, where I was part of a game design panel. They've already posted a video (should start around 4:40:15). The audience was seated out in the grass and there were many dogs (and a wolf from a wolf sanctuary!), so it was much more relaxed than the background would have you believe, which is why I perhaps do even more odd dancing than usual. Should have more normal sorts of news for next week.

(Comment on this)