CreatorsOk
vrengames
vrengames

patreon


Menu Cleanup and Getting Back to Writing

Lab Rats 2 is finally at a state where I'm able to start writing some nice, large chunks of content to add into it. Shifting back into content writing mode is taking some getting use to. When I'm working on the logical, software side of the game it is usually a good idea to think an idea through before trying to implement anything. Writing new content favours lots and lots of writing, then trimming out or changing anything that doesn't work. With a little hard work I'm hoping to get back up to LR1 levels of content productivity.

While I was writing content and testing some stuff I also noticed how impractical the current tooltips for serums and serum traits were. Those have all been updated to more concisely show all of the critical information. Serum traits in particular now list all of their positives (value, status effects, etc.) and negatives (production cost, research needed) in coloured text. Serum traits are also now more logically sorted and labeled. Traits that are exclusive with one another, like the production level traits, are now grouped with each other and sorted by tier.

Once a few key events for the head researcher are written I will be moving on to the addition of the first major returning character: Stephanie. She will be the first chance I get to take advantage of some of the earlier work I did with personalities. Stephanie can have her own unique "personality" type defined that overwrites certain responses, but calls on another personalities responses when no special response is listed. This will let her properly reference events from the first game, provide unique opinions on events, and otherwise be an interesting character.

Comments

Tried free version few minutes ago. User interface just beyond good and evil. I deleted the game before I could understand how to move around the city. Sorry.

Pat Patrick

Excellent writeup, you've captured a lot of the problems I'm expecting (but plan to overcome, of course!). I think v0,8 is going to end up being a litmus test for how the game feels when it has a solid amount of actual content in it instead of being a random number generator with boob sizes coded in.

Vren

Glad to hear it! Knowing that my code is public and going out to anyone who wants to look at it also encourages me to keep to programing good practices, so it's win-win!

Vren

First of all hats off to you sir for getting this far and not going totally crazy. So much goes into the "making the damn thing work" aspect of game creation that the end users never get to see and these peeks behind the curtain are always awesome. TL/DR: Your stuff is good and writing content is very labor intensive. Writing good content even more-so. OK, so a short essay on what makes a good female character in a game meant for you to collect and bang female characters (honestly this works just as well for males, robots, furries, and sentient farts). One of the biggest problems to overcome is that our brains have been streamlined through thousands of years of natural selection so we tend to dump out anything common and repetitive (like individual leaves in a field of bushes) to allow us to dedicate our limited processing power on the uncommon (like that Tiger that wants to eat us). In game terms any NPC with shared traits as the multitude around them becomes background noise. This is especially true when it comes to visual traits, as that is our strongest input sense (and in VN games pretty much the only sense). This means anyone wanting to have a game with lots of intractable characters really needs to invest in visual differences for each intractable individual. the greater the importance that character has the more unique their design must be because are brains are already filing any clones we see as unimportant background noise. You see this all the time where a plot companion will generally have unique hair, physique, accessories, tools/weapons and wardrobe while quest givers are the same 3 types of uniforms over 4 different races with 5 hairstyles to chose from (Quick math aside: That would give you 60 'Unique" palette swaps, 120 if you add in both genders) and background bodies look like the only store on the planet happened to be just outside their cloning vats. So what does this have to do with sex games? Everything! Well, most things. For a character to be fuckable in any significant way they cant be considered part of the background furniture. They need to have as many unique things about them as possible. The greater the uniqueness the more our brains tell us this person is important and we should interact with them (using our virtual genitalia, naturally). A stumbling block with this is that the content creator will often limit themselves to models with their perceived notions of beauty and sexuality thus cutting the pool of possible permutations for characters in the game. There is a reason most sex games have a very tiny cast; Making unique models is time consuming, and any too-similar characters are going to be discarded by the user. If you are to attempt to make a small village of walking, talking sex dolls you are going to have to consider the unthinkable: have the potential for undesirables to be added to the pool. Older models, too fat, too short, too thin, too tall, the gender opposite of that which you want to bang. All these seem like a terrible idea for a game built for banging, but are actually really great for making super-hot conquerable characters. Those hotties stopped being common and now your brain wants them more. There is a longer rant about what makes the ideal image for a sexual partner but for now lets over-compress it into this: You want a game with an insane 1000 fully unique NPCs. OK here's what you need. - At least 5 skin tones (10 preferred) - Generally 3 eye colours (unless doing a lot of close-ups, then bump it to 6) - At least 10 different hair styles (most developers do 20+ and still need to be supplemented by the modding community. -At least 7 different hair colours (15 preferred, this includes punk and goth style unnatural tones. Bonus points if you can 'layer' your hairstyles to allow for tints and highlights). -At least 15 body morphs (Short,medium and tall variants on Fat,Thin,Average, Voluptious, and Athletic bodytypes 5x3=15) -At least 24 makeup variants (3 for each body colour plus "age" affect detailing for age variants) -At least 100 outfits (this seems rather high, but bear in mind that that is only 10% of our target. this number is more easier reached by collections of at least 10 different styles for each clothing article with at least 7 colour swaps per) -At least 250 faces minimum (faces are key identifiers. ideally each of the 1000 NPCs would have a distinct face. More realistically, permutations of sets of 6 assets per facial feature; 6 eye shapes, 6 lip varieties, 6 noses, 6 ears, 6 eyebrows, 6 head shapes, 36 assets total) -At least 10 different accessories (we don't count these as clothes). Add all that up and that's roughly 113 unique assets that can be used for creating unique NPCs. As far as a computer is concerned that's over 3,000,000 unique people. But as far as our brains are concerned its closer to 5000. Because we see faces first, shapes next and colours last. Minor variations wont cut it; A tall blonde haired, blue eyed woman with a silver watch and red pumps is the same to us a tall blonde with blue eyes wearing a gold necklace and wearing loafers. Any engine designed to pump put these npcs automatically will have to be fine tuned to offer as few similarities as possible, which may mean retiring certain assets after a set number have been committed to use. This, my friends, is a LOT of work in both designing the assets and programming the engine. So lets move away from the visual side of things and look more towards the interactivity going into each NPC. As stated before good model design will draw the user in, but if every interaction is the same then tall that work making design assets and a Skynet-esque AI to create genuinely unique characters was for nothing. AI scripting is hands-down the hardest thing to do. A personality is not just a list of option boxes that can be ticked. A true personality is closer to one of those artsy pictures of a thing made up of smaller pictures of other things. Moreover a 'real' personality requires a knowledge of history on the part of the NPC. Generally this is where a lot of games are weak, because that's a lot of system resources going into 1 of hundreds of NPCs who are constantly adding to their history collectives. The bypass on this is they only remember key plot events (in story games) or have a sliding 'like/dislike' gradient for the user (in sandbox games). In rare occasion a developer will also add a 'like/dislike' slider for NPCs to use on each other (The Sims franchise). This last one honestly gives a lot more depth to the open world type games. So how do we continue to make our 1000 NPCs from before unique? Write 1000 custom personalities? That's crazy. However lie with the visual permutations quirks can be added and combined to give a more unique feel. Each of these quirks should be allowed to build off each other. For example lets say you have a girl with a bare bones 5 quirk package (Intellectual, flirtatious, Likes [x,y,z],Hates[a,b,c],good memory) how that would play out is: Being intellectual all opposition checks (your skill vs hers for convincing) carry an intelligence check that must be also passed to get the full reward, likewise player characters with matching or higher Intelligence would get more dialog options with her. Because she is also flirty, it is easier to generate sluttiness points with her and she would get to wear outfits a bit above what her sluttiness cap would usually allow along with opening up a few more dialog options. The likes and dislikes are pretty much targets to hit/obsticals to avoid within the aforementioned dialog options. the last quirk means her like/dislike bar will not naturally slowly degrade back to neutral, also she may resent the player a bit if you show affection/generosity to another NPC that likes what she hates. So here's a quirk list that could be used (again think of combining these into randomly assigned groups of 5-7): -Charismatic: All opposition skill checks have a Carisma component, Worker makes friends easier, more dialog options for player with equal or higher Charisma. -Intelligent: All opposition skill checks have an Intelligence component, more dialog options for player with equal or higher Intelligence. -Focused: All opposition skill checks have a Focus component, Worker has a harder time making friends, more dialog options for player with equal or higher Focus. -Bubbly: Happiness decays slower, less Happiness lost from actions, more Happiness raising dialog options. -Flirty: Sluttiness decays slower, less Sluttiness lost from actions, more Sluttiness raising dialog options, riskier outfits allowed. -Dedicated: Obedience decays slower, less Obedience lost from actions, more Obedience raising dialog options, girl will not quit, can only be fired. -Likes(x): Generally 3-5 recurring items in the game such as colour (happiness buff if mandatory uniform contains > %50 that colour), food or beverage, Actions (flirtation,praise,small talk,work), hobbies (set of 6 - 12), sexual position, and if you want to be really meta; Quirks (this is more for how they like or dislike their coworkers) -Dislikes(x) like above, just opposite. -Gregarious: more dialog chances -Reclusive: less dialog chances -Sensitive: rewards/penalties are greater -Good memory: Opinion of player does not naturally reset to neutral. Holds grudges -Poor memory: Opinion of player degrades to neutral faster. -Jealous: reacts negatively to player if player shows affection/generosity to another NPC in the same 'room' -Virgin: Rewards from flirt attempts are halved until she is deflowered -Maneater: will actively pursue the player (can also be used to create the annoying clingy type that always hangs around). As a side note Hobbies could also dictate what regions of the map she often spend her free time. That all said, I'm very interested in seeing Vren's approach to the herculean task of making desirable random generated NPCs in a sandbox game.

Sadly we are forced to wait for that one. Fingers crossed we'll see Vren make mention some prototype belly tests in v0.9

I'd like to take a moment to salute you for making this game the way you have, and explaining the technical side of what you are doing. There is a value to your efforts that extends well beyond the game. Thank you.

blakeness

Looking great, so far! :) Just wondering, is there pregnancy in the game, yet?

acac


More Models and Creators