Ultima Ratio Regum (0.8 released after five years!)

[UltimaRatioRegum]

After last week’s post, I was very pleased with the kind of variation in styles of speaking I was getting, but something just didn’t feel quite right. It took a bit of time and quite a bit of feedback to figure out what it was, and that then triggered a second realisation about something else that needed changing, and in turn, a third. And a fourth and a fifth!

First, length and clauses. Many of these are simply too long, and there are too many additional clauses. I think the extra clauses will become just a little bit annoying after a certain point, especially if you’re talking to lots of people from one civilization close together. As such, I’ve made some sentences shorter, and reduced the likelihood of clauses being added to sentences. This also meant a change to sentence complexity: I’ve decided to actually make quite a substantial change here and change sentence complexity from a value/modifier embodied in a civilization to a value embodied individually in each person you encounter. I realized that since you’ll spend a decent length of time in each nation at once, you’re probably going to be spending a while talking to a range of people from the same background, and so I want to boost variation between people more than I want to boost variation between civilizations. As a result, each person now has a sentence complexity modifier specific to them. This has five variations, which I’m currently just storing as 1-5; that’s an x/15 chance that each sentence they say might have a clause. This means even the most wordy person will add in extra detail clauses in 1/3 sentences, but for most it will be far rarer. I’ve tried this out a little and I’m much happier with the kinds of sentences it produces (along with the other changes). Equally, it means that that kind of extra info will come to the player more gradually, which I think is also beneficial.

Secondly, punctuation. The end of sentences now varies according to the mood of the person speaking them; a hostile NPC will end with “.” or “…” to everything they say, a neutral NPC will end with “.”, whilst an NPC very positively disposed towards you will end sentences with a mix of “.” and “!”. Again, a very simple thing, but it’s already making a big different to the ebb and flow of conversation.

Thirdly, and related to the above – formality. Many of the sentences were just too formal and rigid and didn’t have the kind of flow that you would expect sentences to have. To adjust this, I’ve made some changes to quite a few of the sentences, added in the possibility for things like “Hmm…” to appear at the start of sentences, or things like “Let me think”, or “That’s hard to remember”, or “I don’t know much, but I’ll tell you what I recall”, things like that – I’ve also added in more use of “I’m” instead of “I am”, “I’ve” instead of “I have”, etc, and so on. The combination of these various alterations seems to be working quite nicely, and gives a far more conversational feel to the whole thing.

Fourthly, visuals. I felt that using the skin tone for the sentences on both sides of the conversation is making the actual conversation window look a bit too thick and same-y; instead, the player’s comments will always be the same colour as their @ symbol, i.e. white, whilst those from other characters will change according to skin tone. This breaks up the visuals and seems to make conversations a little easier to parse, which is obviously very important!

Fifthly, stories – see below in the example for precisely what I mean by that.

As such, this is how conversations should look from this point onwards:



There’s still a lot more to be done, of course – like adding “Hmm…” and all the other stuff described above, but I think this is a great step forward.

More next week!

[Untrustedlife]

Much better prose then dwarf fortress. Good job.

[Aleksanderus]

Much better prose then dwarf fortress. Good job.

I was banned from bay 12 games... It was inevitable

[UltimaRatioRegum]

Thanks - and ha! Good work. What got you banned?

[UltimaRatioRegum]

To make up for last week’s rather short entry, here’s a bit one. I’m pleased to report that a very pleasing volume of new stuff has been done this week, and things continue to look very exciting indeed. Read on!

Default Conversation Options

At this point almost all the “default” conversation options are done! This means everything that you can ask every character, and also the class-specific questions (like “What are you guarding?”, “What nation do you represent?”, “How long have you ruled this nation?”, etc). This was a totally huge volume of work I’ve been putting together over the last fortnight and it’s great to see it all nearly done,. and it’ll certainly be all done by this time next week. Here are some examples of conversations with people from specific classes, in this case with a priest, guard, and diplomat:







Default “Expansions”

“Expansions” are what I’m calling bits of a sentence that certain NPCs will add to certain words in a sentence. For example, as we’ve seen rather than just saying “What is the greatest battle your nation has fought?”, someone from a particularly militaristic nation might be “What is the greatest battle, that glorious arena in which all are equal, that your nation has ever fought?”. Similarly, rather than asking “Are there any mountains nearby?”, someone from a particularly exploration-focused nation might ask something like “Any there any mountains, those peaks that inspire us to great deeds, nearby?”. Almost all of the expansions for the default conversation options are now in place, and wow, there are a truly huge number. There are close to a thousand expansions I’ve written for various ideologies with various words, and much like everything else, the inclination of a person towards using expansions is contingent upon their sentence complexity value – a higher sentence complexity means more expansions. As a result of making conversations more human as we discussed a fortnight ago, these are now far rarer, but here are some examples from the generation system, which gathers all possible expansions for a certain word for every nation into one place, then if a random number is rolled that is lower than the nation’s sentence complexity, it selects an expansion for that nation for that term and stores it. Again, by this time next week these will be totally finished.



NPC Personalities

NPCs are now generated with four personality modifiers that will structure how they behave. These will now be explicitly visible, although they will be noticeable from the kinds of replies you get, and also in some cases I suspect I’ll get other NPCs to tell you about the personalities of other relevant NPCs.

Sentence Complexity – this has been shifted from nations to people, and reflects how much people give in answers they’re willing to give. By this, I mean let’s say you ask someone about whether they know about any nearby towns, and they know about five towns. Will they tell you all five, or just one, or somewhere in the middle, and will they comment on what they think about those towns? This modifier will determine that, but it’s different from whether they’ll reply at all – replying is a binary, and then the detail of the reply will be structured by these metrics.
Education – this affects how much someone knows, i.e. how often they’ll wind up saying “I don’t know!” to a question. Naturally this will vary according to class, and expected educational levels, and so forth, and therefore certain questions will be more likely to give you an answer from certain NPC classes, because others simply won’t know the answer!
Patience – this is a modifier for how much an NPC is willing to speak with you. This will combine with modifiers for each civilization, and also for each NPC class and the player’s current status (so a guard will be disinclined to talk to a random person automatically, whereas a merchant will naturally be more inclined to talk to a random person in the hope of making a sale).
Privacy – this modifier affects how willing the person is to talk about their own personal details, and probably how willing they are to tell you information about their culture, religion, etc. This won’t vary too much, but some NPCs of the same class will certainly be more or less concerned about telling strangers their thoughts.
These four metrics are being generated for each NPC, but I can’t say yet whether they’ll make it into the conversation system in 0.8 – I’m really trying to get an absolute core conversation system done, with the kind of scope and variation I want, and then release that. It’ll probably be in 0.9, but I thought it was worth drawing attention to these elements now, as more ways to vary the conversation experience.

Conversation Scrolling

Added a scroll-bar to the main conversation window. I pondered for a while how best to do this. Firstly, I was going to have the scroll bar only appear to the left-hand side of the right face when the conversation gets long enough, but that would mean that the conversation window would then slightly resize itself (which didn’t look nice) and it could slow things down a little to have to measure the size of the conversation before printing it, and then altering the size of the printed conversation. I then tried having a scroll bar only appear on the right-hand side double-grey border when you needed to be able to scroll, but this was very inelegant and seemed disconnected from the actual conversation you were scrolling through. As such, my third try was to do what I’m now going to stick with, which was to add a dark version of the scroll bar always present on the right-hand side, and then have that scroll bar expand and light-up once the conversation is too long. See below:



(Though you’ll note just for testing purposes I asked the same question a lot, and they didn’t get bored – boredom still needs implementing!)

Next Week

Next week I’ll be finishing off the final parts of the elements above, and I’ll be working on the non-default responses. This means answers which are negative, e.g. you ask “What size is your army” and the reply is “We have no army”, and also special answers, like “What do you think about [nation]” which doesn’t easily fit into a sentence structure where you just switch out words, as they’d express positive/neutral/negative feelings very differently, so these need more complex answers. See you all then!

[UltimaRatioRegum]

All the remaining default conversation options, and all the remaining expansions, are now complete. I’ve also altered the expansion code such that certain expansions aren’t tied to certain words or sentences within a language and guaranteed to appear whenever that sentence or word is said, but instead they appear with a % chance for every instance of a particularly word or phrase someone in that dialect says, depending on their sentence complexity (as we discussed before, sentence complexity is now tied to individuals, not to entire cultures). Here’s a couple of examples, courtesy of our good playtesting friend, Orangejaw Moonblizzard, and some NPCs who may or may not have had their origins changed using admin commands for the sake of testing (as you’ll notice these replies could not be for the same nation!)…





Negative Replies

The big thing this week and weekend has been working on negative replies – so, for instance, if you ask “Are we near the desert?”, the default response is “We are near [desert] in [direction]”, or whatever, but obviously a valid option is “We are not near the desert” – and this obviously applies to loads of questions. What if the speaker’s nation has no army, or dislikes art, or have never travelled, or doesn’t know any other civilizations, or lives on a tiny island and knows nothing of the wider world, or doesn’t worship a religion, and so on? We therefore now have a body of negative replies for people to basically say “no”, “that’s irrelevant”, “I don’t know”, or “I don’t care”, in hundreds of thousands of interesting ways!

These negative replies effectively now split up into two categories, which we could usefully call “general” and “specific” negative replies. “General” negative replies include replies like “I don’t know”, “I don’t remember”, “I’d rather not answer”, “I’m not authorized to give you that information”, etc, which can apply to a huge range of NPCs in a huge range of situations. Since the player will run into these fairly often, I’ve made sure that there’s a lot of variation in these general negative responses – although in many cases, of course, there’s only so many ways that you can actually utter some of these things, but here are a few examples.



“Specific” negative replies refer to asking a question where the answer is still answering the question, rather than a general answer, but still a negative. For instance, if you ask someone what they think a particular policy in their nation should be, they might reply “I have no interest in politics”, or if you ask someone whether they know any distant cities you might want to visit, they might say “I know of no distant cities” – and so on and so forth. Each of these is often more specific and more varied than the above, so I’m trying to bias people towards using these wherever possible, although they are naturally dependent upon particular cultural/political/religious situations.



Crowd Disinterest

You’ll all recall the “conversation interest” idea that URR conversations will have implemented – that unless you ask relevant questions, NPCs will quickly lose interest in talking to you. This is to stop the player just going through every single question one after the other, and to encourage you towards asking sensible, logical and appropriate questions. However, I realized the other day that I can’t just limit this to a specific NPC getting bored; if you have a bunch of general questions you’re asking every soldier, for instance, then you could just go from one soldier over to the next soldier in the barracks and start questioning them, ignoring the questions you already asked Soldier 1, but assuming (quite fairly) that they will probably respond the same way, seeing as both Soldier 1 and Soldier 2 are just default soldiers.

Therefore, I need to implement some kind of “crowd disinterest” solution, and I think I’m going to do this on two levels. On the “local level”, NPCs within a building will see who you’re talking to and what you’re asking them about, and also within a map grid (within your line of sight, or nearby), and will take note of the questions. So if you question Soldier 1 about pointless stuff, and they tell you to go away, then you start asking Soldier 2 pointless stuff, the time it’ll take Soldier 2 to lose interest will be shorter than normal; Soldier 3’s will then be even shorter; and so forth. Then, at the global level, I think we need a system whereby information about the player slowly spreads through cultures/cities/religions/etc in the entire world so that people get some idea of whether they should respond to the player or not. Neither of these systems will be in 0.8, but they definitely need to be there.

Next Week

Remember those two new conversation features I mentioned a while back – replies and counter-questions – and also all those questions that have more complex replies, such as lists? Some combination of those will be coming next week – probably the complex replies, I would think. See you then!

[AgingMinotaur]

I think Elona has a system for public performance, where a player who makes too many unsuccessful attempts at swaying the crowd, will ultimately get lynched.

As always,
Minotauros

[Aleksanderus]

I think Elona has a system for public performance, where a player who makes too many unsuccessful attempts at swaying the crowd, will ultimately get lynched.

As always,
Minotauros

Getting killed by a crowd throwing rocks at you is always fun!

[UltimaRatioRegum]

Ha! I didn't know that, but that's neat. I'll have to look that up...

[UltimaRatioRegum]

Two exciting realisations about in-game conversations this past fortnight! In the process of continuing with sentence generation it became rapidly apparent that in order to actually ensure conversational flow, and to boost the range of experiences that the player can actually have in talking to the game’s NPCs, several new additions would have to be made. As readers will remember, until now there have been basically three kinds of question – the kinds of question that anyone can be asked from the start of the game (such as “What do you do?” or “What are your religious beliefs?”), the kind of question that only specific people can be asked from the start of the game (such as “What are you guarding?” which would apply only to guards, but can always be asked), and the kind of question that can only be asked after a particular trigger is met (such as “What do you think of the novels of [author]?”, which requires you to have come across the author). There is another “category” of question, as well as a new kind of reply that both the player and the NPC will be able to produce within a conversation. These do not especially increase the required volume of sentence generation implementation, as the creation of an overall framework for sentences (in progress) will be equally applicable to all possible sentences, but they will substantially enhance the variation of in-game conversations, and also their realism and believability. Read on!

Counter-Questions

Counter-questions are, as you might expect, questions that the person you’re talking to winds up asking you, which then lead to a variety of possible replies. So, for instance, they might ask you about YOUR religion, in which case you can be truthful or lie and so forth, and the same will then apply for a range of scenarios. Questions of this sort will be particularly relevant when trying to get past guards, for instance, or when trying to enter particularly restrictive or xenophobic or militaristic nations, and the like. I think this will add a very strong extra layer into the conversation system and the ability for the player to “bluff” their way through certain areas, whilst also (like the above points) boosting the realism of the system by changing it away from a rather simpler question-and-answer system. Here’s an example of what counter-questions might look like based on some of the conversations from last week:



Replies

A “reply” in URR is something that can only be said in response to another person’s statement. When a reply is possible, the “Replies” dropdown list will appear, highlighted, right to the special options and to the left of the dialects. Clicking this option will then offer this special sentence/response; if you select another option that isn’t in the replies list, the replies will remain and can be returned to later in the conversation, so this isn’t just a one-off chance. This means that as particularly long conversations continue, the number of possible replies will grow, but it will never grow above half a dozen or so at the most, I would think, and some of the replies will actually be the same as other default questions that simply take on newfound relevance after a previous question. Here’s the same conversation with some answers and two different models: which do you prefer? In the first version the reply and the next question are apart, and the other person might make another comment, whereas in the second version, the reply and the next question are combined…





If you don’t select the reply when prompted, various things will happen. If they ask you a question and you just go onto asking them another question, they’ll likely be offended. Alternatively, some replies will be akin to “follow-on” questions, so they mention something, and you can then pursue that with them as a result of that conversation. These options will stay in your reply options for that character forever, and can be addressed (or not) whenever.

I’m going to implement this because as I experimented, it became clear that conversational flow in real life is obviously far more of a back-and-forth than a question-and-answer, and the system needed to reflect this. The reply will allow the player to get more depth about a specific question asked instead of moving onto another question, and should be especially relevant if the NPC being talked to winds up revealing something the player wasn’t previously aware of, and wants more information on.

In the process, though, I found myself wondering about the exact flow of the overall. Right now the system is that you select a topic, you select a question, then the questions window closes and you see it spoken and a reply given. This is largely from the inevitable constraints on the size of the screen, fitting in questions as well as images and topic lists, and so forth. I became briefly concerned that this might actually break up the flow of conversation, especially if you want to ask two questions from the same topic list one after the other. To help cancel this out, if you return immediately to a conversation topic you’ll be at the same part of it you previously scrolled into, and I also added in a system that will keep any letters typed in for the search function whilst you’re still in one topic. So let’s say you ask about Artwork X from Nation Y, and you found the question by typing in Nation X’s name into the “Art” menu, and that question is asked, and then you want to ask about another artwork from Nation X, going immediately back into the art list will keep the letters there, but changing to another menu will reset the letter tracker. I think this is quite a good compromise. Equally, of course, we should remember that the conversation system in URR is specifically designed to be a little slower and more thoughtful than the systems one gets in more RPGs where the player is encouraged to just exhaust conversation options. I think the current system (with these recent changes) will now work very smoothly, actually, and I’m happy with the balance between ease-of-access and possibly encouraging the player to think about other relevant options as a conversation progresses.

Next few weeks:

In the coming weeks I’ll be trying (see below) to continue the implementation of counter-questions and work on the implementation of replies, as well as adding all the “Hmm…” stuff and punctuation options into conversations properly, rather than having them only present as a placeholder to test how things look. The other big thing to handle now is that all the [religion] or [date] or [place] modifiers work; a few have been implemented, but again, just quickly and crudely to get a feel for the system. Another task for the coming weeks is to go to all of these, make sure they work, and to make sure NPCs already spawn with the requisite information for personal details – “I was born in [place] in [year]” – as well as civilisation details – “I am from [nation] and we worship [religion]”, and so on. There’s a pretty big selection of these,so it’ll take a bit.

Why no blog post last week?

Last week was the first time in several years I missed a blog update – this, unfortunately, is simply because I am basically working twelve hours, every day including weekends right now, and I’m totally swamped with my work at the moment. Believe me, folks, I am *incredibly* desperate to get 0.8 finally finished and out, but there’s just so much on my plate at the moment with strict fixed deadlines which has to be prioritised, and it’s hard to carve out any time (even at weekends) to do much coding. Just rest assured I am working as hard as I possibly can to get 0.8 released, and I’m really sorry it’s taking so long and I feel tremendously guilty about dragging it out, but things are really tough right now (due also to other life issues beyond what I’ve described here) and I’m deeply stressed out, but doing my best. There’s only so many hours in the day and I just cannot prioritise coding at this month/two-month moment in my life, but I’m trying to grab an hour here and there to just inch forward on conversation implementation. In the coming month, therefore, some of most of the blog entries are going to be other entries on general games topics of discussion/interest I’ve had written for a while and I think you folks will be interested in, and I’ll upload those instead if (as is likely) there isn’t really any URR update to put out. I hope you’ll enjoy these entries; people have always enjoyed these before, so I hope these will keep you all occupied until I’m out of this rough patch. I’m really sorry, I hope you all understand, and I promise 0.8 will be worth it.

[UltimaRatioRegum]

After several months where I just haven’t been able to find even a couple of hours to code, I am extremely pleased to report that I’ve managed to get through this incredible glut of academic work and emerge into the wonderful fields on the other side. As such, this coming week I am now officially resuming development on 0.8 in my spare time – which is still not extensive, but at least I have some of it now, with more to come as the start of next calendar year comes to pass – with the objective of getting 0.8 polished off and finally released in the next few months.

So: where were we, and what remains to be done?

The State of 0.8

0.8 is, by far, the biggest release I have ever done and the biggest release I ever intend to do. Had I known that I’d run into this unexpected delay towards the end of the release, I would definitely not have tried to develop the speech systems alongside everything else and I would released everything except speech as 0.8, then released speech and conversations as 0.9, but as it stands, there are several major parts of 0.8:

NPCs

The current version available here on the site, 0.7, has a massively detailed world devoid of people. This has now been addressed, and there are millions of people – all procedurally generated, with the important ones stored and tracked, and the less important ones spawning and despawning to give a sense of the crowds the player moves through – that can be met in the in-game world. This, obviously, is one of the biggest changes to and developments in the game, and is the last major “system” that needs to be in place, and – via the conversation system – marks, finally, the beginning of actual gameplay! Although a lot of coding was required to keep track of NPCs at various resolutions, near the player, far from the player, inside buildings, outside buildings, inside or outside buildings that hadn’t been spawned yet, and so on and so forth, we finally got there. All NPC systems are finished, and only a few tiny bugs remain (found by my kind playtesters) to be fixed prior to a release of 0.8.





Clothing and Face Generation

Alongside these NPCs we have clothing and face generation. Clothing generation can offer many millions of variations of clothing styles, which the game then varies in four levels – “lower”, “middle”, “upper”, “ruler” – and distributes appropriately to NPCs. You can then view the clothing of anyone you run into, and the player of course now also starts with a set of clothing appropriate to the civilization they start in. Face generation, meanwhile, creates a set of genetic and cultural variations which can produce high hundreds of millions of faces, I believe, allowing you to make judgements about both the geographic and national/religious origins of people you encounter. Both of these systems are fully integrated and finished.



Buildings, Districts, etc

An entirely new, massively, and massively important class of buildings have been added: castles. The aesthetics of these vary according to their civilisation of origin (as do all buildings) whilst what is found inside varies according to the ideological and religious orientations of that nation; you’ll find altars, banners, barracks, torture chambers, dungeons, libraries, studies, and a million other things in there (alongside a relevant set of NPCs to inhabit them and move around within them). In districts, meanwhile, I’ve gone over and changed a bunch of the generation algorithms, fixed some unusual edge cases which could sometimes appear, made sure NPCs can always path to where they need to path to, and basically ensured that districts do function correctly in all the ways they have to. All castles and district changes are in place with the exception of a few tiny bugs to do with castle generation that need to be resolved before release.



Speech and Conversations

This is the final one, and the biggest one, since it marks the first meaningful element of gameplay being introduced into the game – which is to say, talking to people, questioning them, acquiring information from, presenting oneself in certain ways, deciding which topics to pursue, and all the other conversation mechanics and dynamics I’ve talked about in blog posts before this one. Speech generation is partly complete, and conversation mechanics are partly complete.



What next?

Therefore, there are basically three objectives left to cover before I can release 0.8, each of which is probably approximately of around equal size:

1) Finish Speech Generation

The primary objective is, obviously, finishing the generation of all sentences and potential responses. Now, you’ll remember from some of the final blog entries before I went on the coding break that I was introducing some more complex speech systems, but I’ve decided it’s more important to get 0.8 than it is to implement all the speech dynamics I have in mind. As such, I’m toning back some of these ambitions for the 0.8 release, and whatever kinds of sentences and comments don’t make it into 0.8 will make it into the (small and compact) 0.9 that should appear quite soon afterwards. Nevertheless, there are lots of sentence archetypes that need finishing, lots of sentence components, lots of variations for ideologies/ geographies/ cultures/ religions/ etc that need to be implemented, and then – most importantly – I need to make it all actually generates correctly! There’s so much content in there that some weird grammar mistakes are bound to crop up, and I’m sure some spelling mistakes have slipped in too, all of which need to be fixed.

2) Finish Conversation Mechanics

Right now actual “conversations” don’t really happen, since the AI you’re talking to just produces a response without, for instance, asking you something back. As above, since I’m not putting in every possible sentence form in this release, I’m also not putting in ever possible conversation mechanic into this release, but I still want things to flow quite well and to be quite interesting to deal with. Even though I hope to get 0.9 out fairly speedily, this is still going to be a version a lot of people are going to play for a lot of time, and I want to make sure there’s enough there to reward the wait, even if 20% of the mechanics I want will just have to wait until a later version when I can take the time to really perfect them.

3) Finish all Tweaks, Fix Bugs, Optimisations, etc

Not much to say here – I just have a list of things that need fixing, and I’m going to fix them. Now, some of these I’m going to leave for 0.9, which is going to be a far shorter release anyway, though it’ll only be the bugs that aren’t game-breaking, or are perhaps entirely aesthetic, that I’ll leave until then. Anything critical, or anything that is definitely going to be noticed, needs to be fixed now.

Therefore:

It has been very hard on me to have to stop programming for a period and cease, however temporarily, working on a project that I care so much about and means so much to me. I know some people will have drifted away, I know the blog has lost some readership in the last few months, and – chances are – some people will have seen this as proof that a big project like this is doomed to fail! But worry not: we’re back, and the ramp back up to the speed I was developing last summer will take a little while, but it’s going to happen. My schedule and my life more generally will be freeing up gradually over the coming months, to some extent Jan/Feb/March, and even more so April/May/June/July, and that spare time will be going into URR. Next summer (much like last summer) in particular will give me ample time to code, and I’ll certainly be taking advantage of it. I’m starting now to keep the code open, to check in on what needs doing, remind myself where I left off, and to start slowly but surely developing the speech content first and foremost, as that’s certainly the easiest thing to get back into the swing with. URRpdates in the coming weeks might be smaller than normal, but they will still have something to show. URR is back, and the final push towards the first gameplay release is now – finally – on.

[UltimaRatioRegum]

As you may have seen last week, the development on 0.8 is finally ramping back up again. It has been a long road here, and a far longer one than I would have liked, but I now finally find myself with enough time on my hands to actually re-open that famously lengthy Python file and finish off this gigantic release. Last week I put out something of an overview of where the game stands at the moment, and what needs to be done: which is to say, basically, finishing off the content for speech generation, finishing off the most basic version of the conversation system I would actually feel comfortable releasing, and then dealing with whatever bugs and minor issues remain that cannot wait unil 0.9 (which is to say, crash bugs, or other kinds of serious errors).

The easiest thing to resume work on is the content of conversations. All the data structures and whatnot are present for me to put this stuff in; I just have to actually write it. That’s what I’ve been tasking myself with this week, and I realised that this wasn’t really finished, and that there were quite a few new kinds of variation I could add – with only the tiniest alterations to code – that would bring a lot of extra variety to the whole thing. Therefore, for the last few days in my spare time I’ve resumed fleshing out, and adding more detail to, the conversation system. This basically means ensuring that for all the possible political ideologies, and the various religious orientations that might come up, and so forth, people will have something appropriate and reasonable to say which will (in some cases) hint back towards their backgrounds and beliefs. The central part of this was filling out all the possible “expansions” – where an NPC can say a little something extra about a particular topic, as informed by their background. I thought I had done so previously, but I now recognise this wasn’t the case and there was a lot of valuable (and easy to add) extra content I could include. These are now all complete!



I also went to the list of alternatives for certain words in sentences – where one is selected at random for each word for each civilization – and finished this list off as well. As with the above, there was definitely more to be added here, and doing something relatively coding-simple, but content-heavy, has also helped with getting me back into the swing of things.



Lastly, I also finished off another segment of the game’s conversation system, which is ensuring that multiple phrases with the same meaning, but slightly different contexts or grammatical structures, are consistent. Which is to say, if is a civilization might use the phrase “put to death” in place of “executed”, then “execution” should also be “putting to death” – see what I mean? In this case the game randomly selects the way of expressing the “base phrase”, in this case “executed”, and then cycles through all the possible derivatives of that phrase and ensures that they are all combined with the original base phrase, such as “executed”, “executing”, and so forth. In some cases this is trivial, and in some cases more complex. For example, the word “produce” might be rendered for different civilizations as “produce”, “create”, “cultivate”; to render them in the past tense, it simply needs to add a “d” onto the end. However, if the words were also “make”, “yield”, and so forth, these have different rules to be transformed into “made” and “yielded”; I’ve therefore implemented a set of fairly simple but quite comprehensive grammatical rules to ensure that different tenses and uses can be handled, and that the game will also use the right one for a particular civilization when it has someone from that culture say a particular word.

So, that’s now all there is to say this week. However, I’ve had several comments that have stressed the importance of keeping updates rolling now that development has resumed, even if things start out slow and quite minimal. As such, I would expect some of these updates to be quite brief (like this one), but I think it’s better for me to put out short updates than to save them for longer updates and do other pieces in the interim: it’s important to me that it’s clear to everyone that development is moving again, and to gradually regain the momentum I had until around last September or so. In the coming seven days I’ll be looking at the set of conversation variables that have to take account of potentially-changing external factors, and be generated in particular sentences, which are currently as follows…



These need to be always able to draw on the relevant bit of information being mentioned, and to “output” with an appropriate grammatical structure. So, for instance, [god] is the name and title of one’s deity or set of deities; [greatestbattle] is the name of the most important battle of that nation; and so on and so forth. This is probably the next big task in speech generation, I think, after which the conversation mechanics will be returned to. This week I’m hoping to implement a basic framework for these elements, and start to get characters drawing upon their backgrounds and life information to fill them out correctly.

See you next week!

[huw2k8]

In my opinion the most promising roguelike currently in development, the depth it aspires to is incredible and I hope with all my heart this game is built to mark's vision because if so we are all in for one of the greatest games of all time, no doubt!

[UltimaRatioRegum]

In my opinion the most promising roguelike currently in development, the depth it aspires to is incredible and I hope with all my heart this game is built to mark's vision because if so we are all in for one of the greatest games of all time, no doubt!

Hey - thanks for the very nice comment! I'm so glad you think so (and I hope so too). Do let me know what you think of 0.8 once I finally get it out there!

--------

This week I’ve finished off the generator for greetings, farewells, compliments, insults, threats, and giving thanks; each of these can produce easily over tens of thousands of variations, and then when you factor in elements outside the sentence generator itself – the name of a god, the title of a ruler, and so forth – we readily push well up into the millions. In this entry I’ll talk a little about how these generate, give some examples, and look at the kinds of roles I want these elements to play within the conversation system.

In working on the speech generation, it quickly became clear that having a set of phrases distinctive to each nation will be an easy and quick way to potentially identify the national origin of a character you’re talking to (and an easy way to fake your own, if you know all the common phrases…) and so I’ve tried to break these down into appropriate groupings. In the end I decided that there were six major categories I could vary from culture to culture: greetings, farewells, insults, compliments, threats, and thanks. Each of these will be generated for each culture, and will also vary each individual time anyone says one, so one person from Culture X might say “So long for the time being, and may our great military leadership lead us ever forwards”, whilst another from the same culture might say “Goodbye for now, and let us hope our grand military leaders lead us forever onwards” – and so forth. This has required another large table of syllables, of course, but since these are very regular and common sayings I thought it was extremely important to make sure these varied even within cultures and between individuals, rather than (as with most phrases) having them vary only, or primarily, between cultures and religions and backgrounds, and so forth.

Greetings

For greetings I wanted to make sure that these would be sentences that wouldn’t be too lengthy and therefore potentially annoying to see repeatedly, but should also contain at least a little bit of detail in them (this applies to most conversation elements, but I think greetings are particularly relevant in this regard). I went through several iterations of how these might be generated until I was able to settle on one that hit these two requirements (brevity and detail) reasonably well. At this point, therefore, greetings tend to be of the structure “[Greetings] [from] [X]”. The first element will vary between cultures and between individuals, such as “Greetings”, “Good [time of day]”, “I greet you”, “My greetings”, and so forth. These can sometimes be pushed to the back, so you might get “From X, Greetings” or “Greetings from X” – some variations are grammatically correct in both variations, whilst some are only correct in one variation, and this is all coded in. The second element will vary in the same kind of way – “from”, “on behalf of”, etc – and so will the third, which is inevitably the most varied element. This third element will look at who the individual is and the kinds of beliefs they have, and then generate or select an appropriate greeting as a result. In most cases they will explicitly mention their nation of origin (e.g. “Greetings from the brave soldiers of [nation]”), though in some rare case they will mention their religious belief in their greeting instead of a national or cultural origin (“My greetings on behalf of the zealous defenders of [god]”). As you can see from those two examples, in the first case it seems reasonable to assume the speaker is proud of their army – perhaps a standing army, or perhaps an imperialist nation? – whilst the latter is clearly strongly religious, so perhaps they come from a theocratic or religious zealous nation? Here are a bunch of examples – see you can take some guesses about the political / religious / cultural / geographical / etc backgrounds of the speakers…



Farewells

So, farewells once more needed to be something that could vary sufficiently much and sufficiently often for them to not get boring when somebody might be talking to the player multiple times, or the player might be talking to numerous people in the same civilization or who worship the same religion. This varies by being broken down into polite, neutral, and sharp farewells. A polite farewell would be of the form “[1]” + “[2]” + “[3]” + “fond_farewell” + “specific_farewell” + “!”. So, this might be something like “I’m afraid I” + “have to” + “take my leave.” + ” Now I bid you goodbye, and” + “may you find enlightenment in study” (for a monastic nation). Or, alternatively, “Alas, I” + “must” + “depart.” + “I say farewell, and” + “may all of your hunts bring trophies and glory” (for a venatic nation). A neutral farewell does not contain the first section, and does not contain the “Now I bid you goodbye”, so whereas a polite farewell might be “I’m afraid I have to take my leave. Now I bid you goodbye, and may you find enlightenment in study!”, a neutral farewell would be “I have to take my leave. May you find enlightenment in study.”, whilst a sharp farewell uses the same earlier components and a different end component – rather than a culture or religion-specific end point, you would get something like “I have to take my leave. May all be well” – a generic, general departure which is sharp and not especially friendly. Again, here are some examples, which should give you clues about the speakers…



Compliments

Next up, a pair of related elements – compliments and insults. Each of these will come up less often, but I still naturally wanted these to be very distinctive for each nation, and each example will be worded differently on each generation. “I wager you are as wise and far-sighted as a hawk”, or “I believe you are as clever and sage as the hawk”, or “I know you to be as smart and far-seeing as a great hawk”, and so on and so forth; it’s clearly the same greeting, but each person says it in a different way and will say it in a different way each time, too, to ensure that kind of variety is maintained. These again generate according to ideologies and religions and so forth, and I think they yield a very pleasing level of variation. Examples:



Insults

Insults were slightly tougher. Whereas compliments can work quite well if saying pleasant generic things, insults have to be relevant to a range of reasons why they might be insulting the player – refusing a trade, stealing from them, challenging them in combat, being a worshipper of a forbidden religion, or whatever. You wouldn’t want a character to kill somebody in an arena, and then someone from a pacifist nation praises them for their pacifistic tendencies. I’ve mixed things up to therefore create broader, and more sweeping sentences that should be applicable to a range of possible situations, whilst still allowing the character to say a logical compliment. Originally my plan was to make insults and threats fairly interchangeable, but just alter the first few words. So an insult would start “May you”, or “You should be”, “You ought to be”, or whatever, whereas a threat would be “I will see you”, “You will be”, “I will have you”, and the like. In the end I decided not to go with this model and to introduce variation between the two and thereby more overall variety into the conversation system (the kind of decision I’ve usually made!) and I split these into two. It was much easier to build appropriate threat generation than insult generation, actually, but the insults have come out really nicely and have a rather unusual sense to them; they’re quite distinctive, and run through quite a range of different ideas and concepts. Examples once more:



Threats

Threats, then, are similar to insults, and the variation is best illustrated simply by showing some examples:



Thanks

Last, but not least: we have “thanks”. These start with words like “As”, “Being”, “Speaking as”, and so forth, and then something to do with their background, and then a form of thanks. For instance, someone from a conscript nation might say “Being a conscript proud to serve [herhis] [homeland], [thanks]”, or someone from a zealotry background might say “Being as one with the great light of [god], [thanks]”, and so forth. These are shorter and snappier than some of the other generated sentences in this set of six, but they work very well, and again get the point across snappily and effectively, whilst being relevant in a lot of situations. Whereas insults and threats were tricky because they were dependent on what the player had done to merit the insult/threat, thanks are dependent on the speaker who is relating what particular traits or characteristics they find especially valuable. Examples:



Next Week

By next week I should have these actually implemented into the game, rather than Python’s output log, and these will come up during conversations. I’m still deciding how exactly things will work with regard to when you say greetings – perhaps they will be automatic, or perhaps they will come up as a default option, or something like that – because I want these to be present to make conversations appropriate and smooth and realistic, but also not add unnecessary work greeting every time. Right now I think the best solution is for the game to automatically give you the greeting options when you open a conversation, rather than having you enter the greeting options manually, but I’ll try a few options and decide on which runs the best. Aside from that, I’ll be continuing work onto speech generation, and slowly moving towards the conversation system – the former is easier to get back into, so I’ve been working there so far, but I’ll now be slowly transitioning into handling the conversation system elements. See you all next week!

[UltimaRatioRegum]

Meanwhile, here's this week's blog entry, "2016 in Review", with a summation of URR development and everything else I've found myself up to in the world of games in the last year:

http://www.ultimaratioregum.co.uk/game/2016/12/31/2016-in-review/