Author Topic: Fun game, but I seem to have some issues with it..  (Read 35499 times)

0 Members and 1 Guest are viewing this topic.

Offline Sarabada

  • Newbie
  • *
  • Posts: 30
Re: Fun game, but I seem to have some issues with it..
« Reply #30 on: October 20, 2016, 11:54:26 AM »
Well, removing such a subjective trait is something I can understand - you're completely right in that different people find different things cute. I was going to try and re-add it just so the item would do something, but then I couldn't figure out what stats to alter because of exactly that. It seemed a bit silly to adjust Charisma for it when everything else based around 'sexy' already alters it, and that the only way to get it right now via those slippers are slippers that provide a penalty to charisma and refinement, so I couldn't quite figure out which stat would be related to 'cute' here, or what the thinking was. There's also the cat-girl trait that I don't think any character in the game has, or is able to get, which kind of annoys me when customers repeatedly try to look for girls with it. Famous I think is in the same boat, although I don't know if the trait is granted when I rank girls up high enough.

On a whim, I decided to try and add Hatsune Miku into the game, since Cherrywood had a Vocaloid pack, but it was missing a data file. After much mucking around, it seems the data file is easy enough to make and add the girl into the game (gave her the Famous trait as I wanted to test it) but it seems the biggest issue is that the images are tagged for the beta and not the alpha. I randomly picked a few images and created a very basic alpha-style tags file (with like only 4 images) and that worked enough for me to recruit her. Unfortunately, I'm too lazy to try and decipher all the tags to get them into an alpha-compatible format for the entire folder. I suppose since the tags follow a specific structure one could write a small app to decipher it according to the naming rules you set and generate a tags file appropriately, but that's more effort than I want to spend right now, so I'll scrap the idea and just wait for the beta.

Moving on, 'Hyperactive' could mean that you 'waste' energy, but it could also mean that you can generate that energy much quicker than other girls in order to maintain such hyperactivity - this is why I was confused. There's no particular 'set' logic for it, and there was no explanation.

This ties into the whole 'somewhat hidden' mechanics thing, and my comment about 'attention to detail.' In the above case, why not have that trait grant a bonus to fatigue reduction as well? And if smaller characters are more evasive, will they also have less HP for being physically less able to withstand attacks? If stunning is in the game, will they be more susceptible to stuns or knock-downs? Will Constitution and Intelligence provide extra HP and MP? Now, I understand there is a line you guys have to draw here with respect to what you include because it would get unwieldy and massively complicated, and if the answer to all of these is 'yes,' then I can come up with some more to find something you guys haven't considered, but the idea is that having one related mechanic hidden but not including another would be confusing. "If my small girl is more agile and can evade more easily, why can't she attack faster?"

Providing information about this mechanic, even via one short sentence somewhere, turns it from a hidden mechanic you have to figure out contextually into one that's explicit and clearly establishes an expectation for the player - "This girl's small frame makes her more able to evade." That's it, that's their expectation, they can't have any related to attack speed or whatever else because it's not listed. They could ask for such inclusion, but they would know that it won't be something they have to keep wondering about.

To your comment about Xela liking hidden mechanics which can only be understood by playing the game -- my argument is if this is still warranted if you can't figure it out by playing the game. Again, for Sakuya's 'Composure' effect, there was literally no way for me to tell what it did. Sure, I noticed her Joy was not rising to 100 for some reason, but it still held in the high 90's. Composure itself means being calm and in control, so does that mean she won't be 'panicked' in battle? Is causing panic even in the game? Maybe it means she doesn't get riled up against aggressive customers and can calm them down? I don't know, baseless guessing just frustrates me here.

For one hidden mechanic that did work well in my opinion, it was when some girls randomly were extremely hard to charm into high disposition / hiring range. Some actually seemed to need me to lower my charisma as much as I could in order to increase the disposition I got for talking to them. After some comparison of these girls, I managed to figure out it was probably a trait (and looking at the code confirmed this) - girls with the Nerd trait were difficult to woo, while ones with Impersonal were even more difficult and required dropping my charisma (not sure why, though.) At least here it makes sense that some girls are simply more resistant to your charms.

Either way, I am hoping you understand what I mean about adding more information from my girl list edits: utilizing color and text size can help specific information stand out and allow players to 'scan over' them without getting too lost in everything else, but everything else is there if they wish to take the time to read through it.

Finally, I forgot to mention this in my last post, but then that post was getting quite long already, but I did have another idea related to these traits. I'm not sure if I'll have the time to actually put in the alpha since beta sounds like it's coming pretty soon, and if it's already in the beta, then there's even less reason for it.

Basically, I'm slightly annoyed by the vast majority of random girls I'm going after only having 3 or 4 traits, and they're the very common ones. This makes them very similar to other girls and customers still demand the more uncommon traits anyway. Unfortunately, there's very few options for adjusting those traits. I had an idea that sending these girls to take classes would have a chance to add or improve a trait they have.

For example, sending a girl to an Etiquette Course would, at the conclusion of the course (when they normally get the stat bonus, not just when the course ends, to avoid players sending girls in for one day at the last day of the course) would have a chance to gain the "Elegant" trait. The chance would be 20% of their refinement stat at that time (so you would need like 500 refinement to guarantee getting Elegant, which I would say is warranted if you're already at 500 refinement and don't have it.) Refinement could also be the chance to provide 'Well-mannered,' although it would only be a chance for one of those traits at a time. If they already have Elegant, then we'll say 'Noble' is the 'second tier' trait, and the chance to grant this would be half what it would be for the 'first tier' - 10% of the girl's refinement. At the same time, 20% of her intelligence would be her chance of gaining 'Smart,' 10% for 'Nerd,' and 5% for 'Genius,' etc. Again, only one trait can be gained per course at maximum, so if Refinement already granted a trait, this would be skipped. On average though, you'd be looking at a 20-30% total chance to gain a trait after each course, unless you're sending girls to courses at beyond 300 of their respective stat.

Likewise, XXX Courses would have a chance of granting 'Nymphomanic,' 'Sensitive,' etc. Combat courses would take 20% of her attack to grant 'Aggressive' or 'Combat Training,' 20% of defense to grant 'Tough' or 'Athletic,' 20% of constitution to grant 'Manly' or 'Tough,' etc. And they don't have to be exclusive either, so Dance courses could also grant 'Athletic' or 'Sexy Air,' and so on. This way, taking courses provides additional tangible benefits, and also allows you to somewhat 'customize' the girls you hire. This would also still leave some of the more unique traits unique still (like...catgirl. Can't really train for this one.) However, it would make the majority of the 3-trait girls moderately more useful. Currently I'm just turning any 3 or 4-trait girl I get into service girls since service girls don't rely on any traits.

Of course, if implemented, each course should helpfully state which traits they have a chance of adding. The details about it can be left out, but it should be clear that sending a girl to a Beauty Course might grant 'Charismatic' and/or 'Egocentric.'
« Last Edit: October 20, 2016, 12:01:21 PM by Sarabada »

Offline DarkTl

  • Hero Member
  • *****
  • Posts: 4737
Re: Fun game, but I seem to have some issues with it..
« Reply #31 on: October 20, 2016, 12:54:46 PM »
In the above case, why not have that trait grant a bonus to fatigue reduction as well? And if smaller characters are more evasive, will they also have less HP for being physically less able to withstand attacks? If stunning is in the game, will they be more susceptible to stuns or knock-downs?
Everything is possible. No stunning in the game atm, but body traits already give different amount of health in the dev version. It's not hard at all to change stats, just one quick line for every trait in the trait file.

Will Constitution and Intelligence provide extra HP and MP?
No, there are already means to increase hp and mp permanently via items and leveling. If they also could be increased by increasing other stats, it will be too difficult to balance.

then I can come up with some more to find something you guys haven't considered
By all means. It's been a while since someone helped us with creating traits.

"If my small girl is more agile and can evade more easily, why can't she attack faster?"
She can, it's just not guaranteed. Attack speed is based on agility. While hitbox is something unchangeable, size itself cannot guarantee that she is faster than some trained ninja.

To your comment about Xela liking hidden mechanics which can only be understood by playing the game -- my argument is if this is still warranted if you can't figure it out by playing the game
Xela wants to create a wiki closer to the release. However, the more we document stuff, the less we create.

Either way, I am hoping you understand what I mean about adding more information from my girl list edits: utilizing color and text size can help specific information stand out and allow players to 'scan over' them without getting too lost in everything else, but everything else is there if they wish to take the time to read through it.
Yeap, this is an interesting approach.

Basically, I'm slightly annoyed by the vast majority of random girls I'm going after only having 3 or 4 traits, and they're the very common ones. This makes them very similar to other girls and customers still demand the more uncommon traits anyway. Unfortunately, there's very few options for adjusting those traits. I had an idea that sending these girls to take classes would have a chance to add or improve a trait they have.
Yeah, we didn't have time to make all traits obtainable for alpha. You can adjust their data files if you want, every trait there has a chance to be added when the random character is created.

Offline Sarabada

  • Newbie
  • *
  • Posts: 30
Re: Fun game, but I seem to have some issues with it..
« Reply #32 on: October 20, 2016, 02:54:23 PM »
Everything is possible. No stunning in the game atm, but body traits already give different amount of health in the dev version. It's not hard at all to change stats, just one quick line for every trait in the trait file.

Fair enough, for a game still in development, changes and additions are always in the works, but my meaning was less about 'modding' the traits to fit these ideas, and more about creating a standard that stays consistent across the entire player base, of which player expectation plays a huge part. Also, for the Hyperactive example in particular, would it be possible to add in an effect for it to restore energy / reduce fatigue more rapidly simply by adding a line to the trait itself? I was under the impression that such a change would require writing in some new ren'py code to handle the logic.

No, there are already means to increase hp and mp permanently via items and leveling. If they also could be increased by increasing other stats, it will be too difficult to balance.

Balance is, of course, a large consideration of such mechanics, and my suggestion was more a suggestion of commonly-understood mechanics in other RPG-type games that players would naturally understand. Having said that, you may understand my reasoning here when some aspects are integrated into the game, but others aren't -- normally not an issue, but becomes frustrating if which aspects that are and are not included aren't communicated to the player.

By all means. It's been a while since someone helped us with creating traits.

I'll probably need to wait until post-release to be able to make any suggestions, since I have no idea what is available and what is 'missing' currently in the development version. That said, I'll attempt to help out where I can, particularly when it comes to making the existing traits more sensible.

She can, it's just not guaranteed. Attack speed is based on agility. While hitbox is something unchangeable, size itself cannot guarantee that she is faster than some trained ninja.

Of course, a 'trained ninja' should definitely have very powerful bonuses related to agility and speed in general, but when you generally think of 'small,' you do equate it with a measure of speed which not only impacts evasion, but speedier movements that translate to faster attack speed than a 'lumbering giant.' Now, I'm glad to hear that the 'faster attack speed' option is planned, but it being not a very high chance is very disconcerting to me. I'll accept that just because a girl is small doesn't necessarily equate to being faster, just a higher chance (such as if she had the Dawdler trait, she probably wouldn't make much use of her higher potential agility.) However, if the girls that do end up with a higher attack speed, since this would not be the norm, end up not having any particular notification that they have this ability, this would cause a large amount of frustration to me -- if I am looking for agile warriors, how am I supposed to single out and find these particular girls without simply hiring everyone?

Xela wants to create a wiki closer to the release. However, the more we document stuff, the less we create.

I understand that; it's tough when you don't have many people working on it. I put a lot of effort into documentation and revealing information in a digestible manner for the mods that I made as well, and often that part consumes more time than I spent creating the systems in the first place. However, I would argue that player satisfaction and understanding would go a long way into making players more appreciative of these systems. Still, I would say a wiki would be low priority, and you could probably rely on players to help fill that in.

Yeah, we didn't have time to make all traits obtainable for alpha. You can adjust their data files if you want, every trait there has a chance to be added when the random character is created.

I probably will soon-ish. I kind of want greater variety in the traits these girls bring, but I've had a notepad open for a little bit documenting the traits I've 'discovered' in all the girls available in my game via talking to them right now and looking through the list, at least 8 of them have the same 'adventurer, clumsy, dawdler' traits and nothing else.

At this point I also have over a million gold, which is more than enough to move to the final Central Palace, and my only goal is finding 50 more girls to train to fill it with, so I'm okay with altering game balance now. I also find it pretty silly how a large palace still can't handle the amount of dirt generated by one stripper pulling in 130 customers, but can deal with 35 prostitutes going to town on all of them just fine. I'm more or less done and happy with the UI edits I've made as well, with much more information I can get at a glance now, so I'll turn my attention to beginning the various overhauls I mentioned in my OP. Priority 1 is probably getting my poor level 13 guard up to pace with the others, haha. Main thing right now is my inability to grasp what kinds of data I have available and when, so I was waiting for some confirmation on how to write into the devlog.

Offline DarkTl

  • Hero Member
  • *****
  • Posts: 4737
Re: Fun game, but I seem to have some issues with it..
« Reply #33 on: October 20, 2016, 04:16:06 PM »
Fair enough, for a game still in development, changes and additions are always in the works, but my meaning was less about 'modding' the traits to fit these ideas, and more about creating a standard that stays consistent across the entire player base, of which player expectation plays a huge part.
Absolutely standard traits do not feel unique. If, let's say, difference between body types is nothing but level of base constitution, then it's just boring.
Ideally every trait should be somehow unique, even if only a little. Otherwise they are not worth looking for, stats can be upgraded without gaining traits.

Also, for the Hyperactive example in particular, would it be possible to add in an effect for it to restore energy / reduce fatigue more rapidly simply by adding a line to the trait itself? I was under the impression that such a change would require writing in some new ren'py code to handle the logic.
That is what effects system for. Although, I'm not a big fan of self-restoring traits because they possibly can make many items useless.

Having said that, you may understand my reasoning here when some aspects are integrated into the game, but others aren't -- normally not an issue, but becomes frustrating if which aspects that are and are not included aren't communicated to the player.
Usually in jrpgs you have close to zero information about stats. While it might be an issue for some players, this genre is very popular nevertheless.

Of course, a 'trained ninja' should definitely have very powerful bonuses related to agility and speed in general, but when you generally think of 'small,' you do equate it with a measure of speed which not only impacts evasion, but speedier movements that translate to faster attack speed than a 'lumbering giant.' Now, I'm glad to hear that the 'faster attack speed' option is planned, but it being not a very high chance is very disconcerting to me.
You misunderstood. Of course there are games where lvl 1 fast character is always faster than lvl 100 slow character.
In our case, however, even slow character can be faster if the difference is 100 levels. But if the difference is small, then yeah, small character will have more agility because she does have bonuses to agility. Agility decided the speed, not pure traits, because unlike the main game battle engine only operates with raw numbers.

Main thing right now is my inability to grasp what kinds of data I have available and when, so I was waiting for some confirmation on how to write into the devlog.
Xela can help with that  :)
You better ask in the general discussion thread to make sure he notice the question.

Offline Sarabada

  • Newbie
  • *
  • Posts: 30
Re: Fun game, but I seem to have some issues with it..
« Reply #34 on: October 20, 2016, 05:52:34 PM »
Absolutely standard traits do not feel unique. If, let's say, difference between body types is nothing but level of base constitution, then it's just boring.
Ideally every trait should be somehow unique, even if only a little. Otherwise they are not worth looking for, stats can be upgraded without gaining traits.

I think there was a misunderstanding here, too. My intention was not to suggest creating 'boring' traits -- I'll agree if the only difference was something like Constitution, it would be kind of useless. My point was more that the traits themselves, however they are made fun or unique, should be designed in such a way that modifying it afterward is unnecessary, thus ensuring that more players experience the 'same' intended behavior. Making a trait unique should not be exclude it from following a logical standard, and deviations should be documented.

In the case of body types, we can establish that going from small to medium to large body types will, at the very least, impact agility and constitution, with the medium type being the middleground for both. Whether this 'baseline' set of expected stat impacts is documented (meaning shown to the player via descriptions or somehow) is up for debate, because although it would be technically hidden, it's standard across the three body types. However, if then the small body type girl additionally has, let's say, a bonus to Luck (completely arbitrary and random example here, not saying it makes any sense) while the larger body type girl has a bonus to Defense, and the medium body type girl has a bonus to Attack, these differences should probably be documented and explained because there's no way of extrapolating these effects as compared to the agility vs constitution situation. If you then additionally have a small body type variant that also increases attack speed on top, it'll be nearly impossible to figure out who has what ahead of time.

One thing you can do here is simply separate out these 'occasional' trait variants into separate-but-related traits, or some sort of trait hierarchy. In this case, the small body type would be the base for a girl, and that body type allows the possibility for the bonus evasion or the bonus attack speed traits to be added (which require the smaller body type trait as its parent.) Once the 'base' traits are taken care of, then you can move on to a pool of 'personality' traits, and then later on, to a large pool of 'miscellaneous' traits that would make each girl more unique.

Ultimately, I still feel very strongly about making information accessible, but I believe I've proven that if I have to I'll just expose the data myself, haha.

That is what effects system for. Although, I'm not a big fan of self-restoring traits because they possibly can make many items useless.

As far as I can tell, this 'effects system' still requires you to write in ren'py code, doesn't it? Unless you're talking about a newly revamped dev version that allows you to just 'hook into' a number of 'unused' effects by linking it from the traits file. This would also go into the whole modification thing.

Regarding self-restoring traits, they can be valuable, but if done well, they won't necessarily make items useless. In the example of the Hyperactive trait restoring stamina, it would be either a few points, or a percentage bonus every turn, but if you really needed a quick recovery, you would still use the potion to restore a large amount immediately. Additionally, you could also increase the upkeep cost incurred (the girl needs to eat more to maintain the hyperactivity?) so that you're technically paying continuously for the cost as compared to a one-time cost of a potion. At the same time, perhaps we can also say that, since it reduces maximum fatigue anyway, that although she can recover stamina more quickly, she generates fatigue slightly faster than normal as well, and will hit her limit more quickly.

Alternatively, you could also create a potion that temporarily raises a girl's maximum fatigue or whatnot past the usual limit for a number of days, or have the potion reduce the amount of fatigue that a girl will generate over a few days (say, a girl will only generate 80% of the fatigue she normally does.)

Coupled with your assertion of 'unique traits' from above, I would say having these kinds of effects would add to the uniqueness of many traits, especially as you can make more traits have effects, and have these traits lock you out from others -- in this case, Hyperactive I believe locks you out of having Dawdler. Dawdler currently just increases your fatigue by a small amount while reducing agility, strip and service, but let's say instead they generated fatigue at a slower rate all the time, although it would also take them longer to recover once they're tired out. Now you have two traits that compete with each other to provide a bonus to the same 'stat' in different ways, and both have merits.

Currently, there's no real reason to have Dawdler besides the customer trait matching bonus because it provides no other benefits, and actually comes with a bunch of penalties instead. Sure, it raises maximum fatigue by 10, but virtually every girl, except the ones that have Fragile, already start with the maximum of 300 fatigue, which as far as I can tell is hard-coded and cannot be increased, so even that one bonus is effectively wasted.

Usually in jrpgs you have close to zero information about stats. While it might be an issue for some players, this genre is very popular nevertheless.

You are right, in some of the more particularly in-depth RPG's I've played, there are usually some mechanics that aren't explained well, but while I still enjoy them, it just means I have to go look up guides and wikis that explain these things all that much more. As well, most of these mechanics are generally pretty consistent once you figure out the first 'step' to them.

One of the last 'complex' RPG's I played was 'The Last Remnant.' I remember being quite confused by that game for a while, too, but one example is your skills could apparently 'swing' two different ways: They could either become 'Nimble' or 'Mighty' as you leveled them up. However, this distinction was at least labeled in the game so even if I couldn't figure it out (though, honestly, nimble and mighty are pretty easy to tell they modify the skill's speed and power, respectively) I had a starting point to search. Hiding information from the player entirely just makes this second step needlessly frustrating, as now you have no idea and you have no starting point to look at. For TLR, you end up with guides like this one where it's a list of things people wish they had known early on - one major thing is that leveling up and grinding like your typical RPG was a bad idea! But you wouldn't know playing the game because it doesn't really tell you the consequences of doing so until you find yourself in an impossible battle because the enemies scale with your 'battle rank' while your own stats cannot handle it (your stats don't necessarily increase when your battle rank goes up.)

You misunderstood. Of course there are games where lvl 1 fast character is always faster than lvl 100 slow character.
In our case, however, even slow character can be faster if the difference is 100 levels. But if the difference is small, then yeah, small character will have more agility because she does have bonuses to agility. Agility decided the speed, not pure traits, because unlike the main game battle engine only operates with raw numbers.

I think we are just talking about slightly different aspects here. I agree with your scenario, and yeah, a stat bonus is a stat bonus when a fight is concerned, regardless of where it comes from. My point was about the 'additional' bonus effect of attack speed on top of the evasion bonus that they get innately from the body type, which it sounded like you said would only occasionally appear on girls with the smaller body type. My argument was that it should either come together as a package, or as two discrete bonuses that you can see, rather than as a hidden modifier.

Put another way, if I have two "small" type girls with very similar stats, which one do I send into combat? The one with the attack speed bonus of course, except there's no way for me to tell.

Xela can help with that  :)
You better ask in the general discussion thread to make sure he notice the question.

Will do, I need to draft up the math first for whatever I plan to tackle first; Probably the dirt, fatigue and XP generation for strippers and service girls since they should be relatively simple few-line edits once I get a good formula down. I'll poke him in that thread if he doesn't respond here before then.

I should also mention that I hope you're not taking any criticisms personally here; I simply enjoy discussion, and am glad you are responding with such detail.

Offline DarkTl

  • Hero Member
  • *****
  • Posts: 4737
Re: Fun game, but I seem to have some issues with it..
« Reply #35 on: October 21, 2016, 02:04:28 AM »
One thing you can do here is simply separate out these 'occasional' trait variants into separate-but-related traits, or some sort of trait hierarchy. In this case, the small body type would be the base for a girl, and that body type allows the possibility for the bonus evasion or the bonus attack speed traits to be added (which require the smaller body type trait as its parent.) Once the 'base' traits are taken care of, then you can move on to a pool of 'personality' traits, and then later on, to a large pool of 'miscellaneous' traits that would make each girl more unique.
It's possible to make traits dependable, yeah. The problem here is ideas for the said miscellaneous traits: names, descriptions, stats. One thing is to make a few body traits and forget about them, and another is to come up with multiple lesser traits for every base trait which will feels unique and logical.

I can upload the current traits files and you can help with them, if you have enough ideas (I don't).

As far as I can tell, this 'effects system' still requires you to write in ren'py code, doesn't it?
More in python than in renpy. Consumable items in the dev version have this option too, this way trait or item can be as complex as we want, within reasonable limits.
But not difficult or time consuming at all.

Currently, there's no real reason to have Dawdler besides the customer trait matching bonus because it provides no other benefits, and actually comes with a bunch of penalties instead. Sure, it raises maximum fatigue by 10, but virtually every girl, except the ones that have Fragile, already start with the maximum of 300 fatigue, which as far as I can tell is hard-coded and cannot be increased, so even that one bonus is effectively wasted.
I changed its logic in dev version long time ago to make it a bit more usable.

However, this is a part of another problem. Not every trait should be definitely desired. There are multiple negative traits in WM, but most of them we cannot implement due to the game concept.
For example, they have that malformed trait which gives huge penalties to charisma. But the character does not look malformed at all, and it's close to impossible to find such pictures for most characters anyway.

Many games don't care much about it, you can even change there boobs size while the characters pictures stay the same.
But we decided to be tied by pictures. Sure, there could be slightly different pictures for the same character, but the general concept remains the same. Character cannot lose hands, become ugly or old. So in order to create negative traits we have no choice but to use "invisible" traits, such as Dawdler.

For TLR, you end up with guides like this one where it's a list of things people wish they had known early on - one major thing is that leveling up and grinding like your typical RPG was a bad idea! But you wouldn't know playing the game because it doesn't really tell you the consequences of doing so until you find yourself in an impossible battle because the enemies scale with your 'battle rank' while your own stats cannot handle it (your stats don't necessarily increase when your battle rank goes up.)
Yeah, that's why I dropped it as soon as I found out about it. Severely punishing for grinding and forcing players to speedrun is a very bad concept for an rpg.
It's a matter of being forgivable though, ingame mechanics should be forgivable enough to not make mistakes deadly if you don't provide clear explanations for them.
Or if the said mistakes are a part of game mechanics when you have to start anew over and over, like in those games where you can level up only after death, such as Rogue Legacy.

For instance, in Dark Souls a novice player cannot possibly know which stats to upgrade to make a successful build as fast as possible. But he can grind as much as he wants and make all stats 99 nevertheless, so his mistakes are not deadly at all, they are just inconvenient for him.

Put another way, if I have two "small" type girls with very similar stats, which one do I send into combat? The one with the attack speed bonus of course, except there's no way for me to tell.
Battle engine only operates with numbers. Battle speed based on agility. Agility is a stat which is visible in the character profile. I don't see a problem  :)

Meanwhile evasion is a special, rare stat which is not visible in character profile because Xela wants it to be hidden. It's fine though, we always can make a hint about hidden mechanics in descriptions, ie "helps to evade attacks".

I should also mention that I hope you're not taking any criticisms personally here; I simply enjoy discussion, and am glad you are responding with such detail.
We had lots of discussions like this three years ago. Then people lost interest as they realized we are not going to release a completely ready game in a few months  :D

Offline Sarabada

  • Newbie
  • *
  • Posts: 30
Re: Fun game, but I seem to have some issues with it..
« Reply #36 on: October 21, 2016, 03:07:09 AM »
It's possible to make traits dependable, yeah. The problem here is ideas for the said miscellaneous traits: names, descriptions, stats. One thing is to make a few body traits and forget about them, and another is to come up with multiple lesser traits for every base trait which will feels unique and logical.

I can upload the current traits files and you can help with them, if you have enough ideas (I don't).

I can take a look at them, certainly, but I can't make any promises or guarantee I'd be able to do anything meaningful with them. Still, it wouldn't hurt to take a look. For the 'misc' pool, I thought the majority of existing alpha traits would be reused, and you'd simply try to come up with some new ones. I'm also not sure what you mean by 'forgetting' the body traits. There is another problem here, though, that you touched on later, and I'll get to it in a bit.

More in python than in renpy. Consumable items in the dev version have this option too, this way trait or item can be as complex as we want, within reasonable limits.
But not difficult or time consuming at all.

I am assuming you are talking about how traits and consumables and such would have the ability to link to an 'effect' name in its definition, but you would still have to come up with the code for the actual effect logic. To me, this still sounds like I'd have to mod it in via Python code (as I have no expectation that anything I write would be incorporated into the base game.) I am trying not to have to mod too many things post-beta-release, partly because I kind of feel like I'd be 'cheating' if I made changes that weren't in the base game - I'm not sure if you understand my feeling here.

I changed its logic in dev version long time ago to make it a bit more usable.

However, this is a part of another problem. Not every trait should be definitely desired. There are multiple negative traits in WM, but most of them we cannot implement due to the game concept.
For example, they have that malformed trait which gives huge penalties to charisma. But the character does not look malformed at all, and it's close to impossible to find such pictures for most characters anyway.

Many games don't care much about it, you can even change there boobs size while the characters pictures stay the same.
But we decided to be tied by pictures. Sure, there could be slightly different pictures for the same character, but the general concept remains the same. Character cannot lose hands, become ugly or old. So in order to create negative traits we have no choice but to use "invisible" traits, such as Dawdler.

And this here is the other part of the problem I mentioned above. You are right that not every trait should be desired. However, I feel like if your MC is savvy enough to run a brothel, he would probably be adept enough to figure out how to make the most out of even the bad traits. It just wouldn't be as attractive as the 'good' traits, but as of the alpha, having one 'useless' trait is pretty painful considering most of the random girls only have 3 traits to begin with. In fact, I have one girl on my list here that I've listed as having "no traits?" because I talked to her 50+ times between the 'interest,' 'hangouts,' and 'about her' categories and absolutely nothing appeared besides the 'fallback' responses telling me she's a stripper. This 'problem' of one bad trait being immensely negative on the desirability of a girl could potentially go away if the girls are guaranteed to have at least a few more traits.

In any case, the other big problem with these traits is that, at least in the alpha, most of the other stats aren't actually all that useful. Prostitution makes heavy use of Charisma and the sex skills, and traits for customer matching, but almost everything else is secondary or outright useless. Strippers are dependent on Charisma and Striptease, with some moderate trait matching bonuses for their traits as well, Service Girls are dependent on Service and Agility, and Guards only need their battle stats, with traits not even really mattering for the last two groups. I don't think any class even bothers with intelligence and character (though I believe character was related to how rebellious a girl would be, which isn't in this game?)

This basically means that you only have a few meaningful stats to put any bonuses and penalties to, which greatly limits the options and variety you can have. My idea for Dawdler and Hyperactive was trying to play on a different set of stats for that reason. This is also the reason why in my OP, I made some suggestions to try and make use of those other stats, which then means we can have traits penalize them.

For my suggestion for Dawdler specifically, it would still be a somewhat 'negative' trait because it still comes with all the stat penalties; its only benefit would be the slower fatigue generation as an attempt to explain that your MC has found a way to make her sluggishness work for him, even despite the fact that the girl would do a terrible job compared to anyone else.

Yeah, that's why I dropped it as soon as I found out about it. Severely punishing for grinding and forcing players to speedrun is a very bad concept for an rpg.
It's a matter of being forgivable though, ingame mechanics should be forgivable enough to not make mistakes deadly if you don't provide clear explanations for them.
Or if the said mistakes are a part of game mechanics when you have to start anew over and over, like in those games where you can level up only after death, such as Rogue Legacy.

For instance, in Dark Souls a novice player cannot possibly know which stats to upgrade to make a successful build as fast as possible. But he can grind as much as he wants and make all stats 99 nevertheless, so his mistakes are not deadly at all, they are just inconvenient for him.

I've never played Dark Souls nor Rogue Legacy, so I can't comment on them specifically; Do their stats not do what is typically expected of them? Or do they do a variety of things not normally associated with them? In the context of this game, I don't think it'd be outright punishing, but for a game about managing girls in a brothel, I get very annoyed when information I need in order to manage them properly is hidden away.

Battle engine only operates with numbers. Battle speed based on agility. Agility is a stat which is visible in the character profile. I don't see a problem  :)

Meanwhile evasion is a special, rare stat which is not visible in character profile because Xela wants it to be hidden. It's fine though, we always can make a hint about hidden mechanics in descriptions, ie "helps to evade attacks".

I assume you mean then that in our hypothetical scenario of a 'small type' girl, her 'hidden' attack speed bonus would appear as an unknown bonus to her agility. But then how do you differentiate it from gaining the effect of the extra evasion? (I am assuming there's an additional evasion bonus that has a chance to be granted on top of the small type's natural evasion bonus.) I still think detailing this bonus for the player would be best, and even just 'helps to evade attacks' would suffice - you could maybe insert the numerical bonus in smaller font under that, or something, but I suppose evasion would be similar to critical hits, and most players would understand that it's a very chance-based system. Although that then begs the question on whether Luck will play into it at all.

We had lots of discussions like this three years ago. Then people lost interest as they realized we are not going to release a completely ready game in a few months  :D

Well, I'm not surprised. These things take time to code, especially if it's complex and there aren't many people contributing to it. I still think the alpha is very playable, I just decided to mod in things I felt were overlooked, not expecting that the beta would come anytime soon either.

Offline DarkTl

  • Hero Member
  • *****
  • Posts: 4737
Re: Fun game, but I seem to have some issues with it..
« Reply #37 on: October 21, 2016, 04:22:00 AM »
I'm also not sure what you mean by 'forgetting' the body traits.
I mean, there is no need make any further adjustments. Body types are limited, like small, chubby, athletic, etc. After you created them, there is nothing more to add, even if there is some kind of exotic body type, you will never find enough pictures to make even a single pack for it.

While dependable small traits are limited by imagination only, something that I don't have much because I have to spend it on all game systems at once.

I am assuming you are talking about how traits and consumables and such would have the ability to link to an 'effect' name in its definition, but you would still have to come up with the code for the actual effect logic.
No, scripts for stats are very simple if you know python. We even already have a few effects which affect how fast disposition changes every time something tries to change it. It's almost the same thing as your proposition for fatigue.

You are right that not every trait should be desired. However, I feel like if your MC is savvy enough to run a brothel, he would probably be adept enough to figure out how to make the most out of even the bad traits.
Sure, for example some customers may even like Dawdler characters, so it makes sense to have one or two. Even then it can be a mostly negative trait in terms of pure stats, after all it's not a very good thing to be too lazy or too slow.

In any case, the other big problem with these traits is that, at least in the alpha, most of the other stats aren't actually all that useful.
Yeap, we added more stats checks to jobs since alpha.

For my suggestion for Dawdler specifically, it would still be a somewhat 'negative' trait because it still comes with all the stat penalties; its only benefit would be the slower fatigue generation as an attempt to explain that your MC has found a way to make her sluggishness work for him, even despite the fact that the girl would do a terrible job compared to anyone else.
Yeah, I suppose it could work.

I've never played Dark Souls nor Rogue Legacy, so I can't comment on them specifically; Do their stats not do what is typically expected of them? Or do they do a variety of things not normally associated with them?
Names won't tell you much. It's a different type of not knowing which stat to level.

For instance, endurance increases equip burden. But you can't know which level of equip burden you need unless you already know about all existing armors in the game, so it's very easy to underlevel or overlevel it. And you can increase stats only after leveling up, unlike our game, where you are free to increase anything as long as you have items/money.

So if you already know absolutely everything about stats and can freely upgrade them, there is not much to do in the game.

I assume you mean then that in our hypothetical scenario of a 'small type' girl, her 'hidden' attack speed bonus would appear as an unknown bonus to her agility. But then how do you differentiate it from gaining the effect of the extra evasion?
It's not unknown, you can clearly see which character has more current and max agility at profile screens, including all bonuses from traits.

The difference between visible stats like agility and hidden ones like evasion is the level of uniqueness. There are multiple ways to increase current and max agility, but to directly increase evasion chance, bypassing normal formula which calculates evasion based on the difference between attacker and defender agility, you need either a special trait or a rare equipable item.

Offline Sarabada

  • Newbie
  • *
  • Posts: 30
Re: Fun game, but I seem to have some issues with it..
« Reply #38 on: October 21, 2016, 01:14:00 PM »
I mean, there is no need make any further adjustments. Body types are limited, like small, chubby, athletic, etc. After you created them, there is nothing more to add, even if there is some kind of exotic body type, you will never find enough pictures to make even a single pack for it.

While dependable small traits are limited by imagination only, something that I don't have much because I have to spend it on all game systems at once.

Oh okay, I think I understand what you mean now; It sounds like the trait hierarchy and dependencies system is something you could support. I would probably say that 'types' such as Chubby and Athletic wouldn't be a 'base type' in this case, as you could have Athletic small and Athletic large girls, and as such, Athletic/Chubby would be in the sort of 'general' pool. In my eyes, the 'base types' should be things that are wholly exclusive to each other -- It's not possible to have a 'small large' body type, after all (setting aside the obvious quip of a girl who would be larger than medium but smaller than large.)

However, has the way the traits are chosen and assigned changed since alpha? I'm looking at the alpha code now, trying to get a handle on it so I know how to change the parameters properly in the rchars files to get a wider variety of traits. At the least, if we go with the hierarchy, I imagine the random girl generation code would have to run multiple 'passes' when generating traits as it has to select traits in each 'tier' of traits.

No, scripts for stats are very simple if you know python. We even already have a few effects which affect how fast disposition changes every time something tries to change it. It's almost the same thing as your proposition for fatigue.

The "if you know python" statement is largely what I'm referring to -- It may be simple, but it still requires you to actually write code to make work if it's not already present in the game. For example, if you have 'generic' code that allows you to just specify a stat name and a modifier in the traits file itself, then you could presumably add traits that add or subtract intelligence, or charisma, or joy, or defense every X turns easily, but if I then wanted to add a trait where girls that are 'lucky' will randomly find 200 gold every week, I'd have to write in that code myself because that's probably not something that would already exist.

Yeap, we added more stats checks to jobs since alpha.

Awesome, I'm glad to hear that. Do you have a list available of what stats each class needs now to make the most of them? I mean, in the alpha Service Girls check agility too, but I wouldn't have known that if I didn't check the code because nothing else tells me that; Even the service course, which I assumed was tailored for Service Girls, didn't indicate it would raise agility. Having a list of all the stats each class relies on would be helpful for creating trait effects that would help and hurt each one.

Names won't tell you much. It's a different type of not knowing which stat to level.

For instance, endurance increases equip burden. But you can't know which level of equip burden you need unless you already know about all existing armors in the game, so it's very easy to underlevel or overlevel it. And you can increase stats only after leveling up, unlike our game, where you are free to increase anything as long as you have items/money.

So if you already know absolutely everything about stats and can freely upgrade them, there is not much to do in the game.

Yeah that sounds like something in which I would just try to max everything out on because that's how I approach these things, haha. That's also why earlier in this thread when I was talking about how I send 20+ girls to the courses to level up their stats, all of them were in school for close to 6 months before they even started working at my brothel; They'd come in and immediately be rank 3's with over 100 charisma and 80+ in every sex stat, Strippers would start working with 100+ Charisma, Striptease and maxed agility, Guards would be capped on battle skills even without gear, and Service Girls would have 100+ Charisma, Service and maxed agility.

Still, I assume it's made clear at some point that endurance increases equipment burden, you just can't know what the optimal level is. In our case, you can't know the 'breakpoints' for various stats (meaning, for the alpha, prostitutes need at least 30 charisma and sex stats to avoid the joy penalty while working) but you still know that it provides that bonus. My concern was closer to if you had two Longswords, otherwise identical in most respects, but one just had 20 more attack than the other for no discernible reason. That would drive me insane in not knowing why there was such a discrepancy. If the second sword maybe was called "Mighty Longsword" then I can work out that "Mighty" is acting as an attack bonus modifier in that case.

It's not unknown, you can clearly see which character has more current and max agility at profile screens, including all bonuses from traits.

The difference between visible stats like agility and hidden ones like evasion is the level of uniqueness. There are multiple ways to increase current and max agility, but to directly increase evasion chance, bypassing normal formula which calculates evasion based on the difference between attacker and defender agility, you need either a special trait or a rare equipable item.

This concern was largely due to my impression that two girls would simply have "Small body" but one had a 'hidden' agility bonus which would inflate her stats, but not tell me why. (Recall my annoyance in why my upkeep calculations didn't match the in-game values.)

However, if we go with the 'tiered' trait approach as detailed above, this should theoretically be a non-issue, as the 'agility bonus' would be a separate, distinct, and clearly visible trait, even if it doesn't tell me exactly how much of a bonus she gets. If you were to go with the 'single' traits, you'd need to create many variations of each 'base' type to include all possible permutations of these extra bonuses, which would probably be more work than simply letting the game randomly put these together.

Honestly, I think the player not knowing the exact breakpoints would be enough. You could tell the player exactly how much of a stat they'll be gaining and it'll save some people some math to work out what that bonus is, but unless they look at the code, they won't know that they'll need precisely X of a stat for Y situation. They'd only know that trait A gives a bonus to one stat, while trait B gives a bigger bonus to that stat, making it more valuable, or trait C which gives an even bigger bonus to that stat, but comes with a penalty to another stat.
« Last Edit: October 21, 2016, 01:18:54 PM by Sarabada »

Offline DarkTl

  • Hero Member
  • *****
  • Posts: 4737
Re: Fun game, but I seem to have some issues with it..
« Reply #39 on: October 21, 2016, 02:40:00 PM »
It's not possible to have a 'small large' body type, after all (setting aside the obvious quip of a girl who would be larger than medium but smaller than large.)

However, has the way the traits are chosen and assigned changed since alpha? I'm looking at the alpha code now, trying to get a handle on it so I know how to change the parameters properly in the rchars files to get a wider variety of traits. At the least, if we go with the hierarchy, I imagine the random girl generation code would have to run multiple 'passes' when generating traits as it has to select traits in each 'tier' of traits.
Of course, all body types are mutually exclusive. I suppose in theory it's possible to find a small yet chubby character, but it's too much trouble for only a hypothetical rare case.

We have a more strict system now. Every character must have a body trait, a boobs trait, a personality trait, an elemental trait, a race trait and a class trait. If they are not set in the character file, the game will set default ones.

I attached traits files to this post. They are still a work in progress though, especially class traits.

if I then wanted to add a trait where girls that are 'lucky' will randomly find 200 gold every week, I'd have to write in that code myself because that's probably not something that would already exist.
If only every code could be as simple as that lucky trait  :)
There is no difference between adding a bonus to stat and some gold. Because gold is a stat too, technically.

Do you have a list available of what stats each class needs now to make the most of them? I mean, in the alpha Service Girls check agility too, but I wouldn't have known that if I didn't check the code because nothing else tells me that
Starting from the beta, characters will have auto equipment function. Meaning that unless character has very high disposition or is a slave, she won't allow you to manage her inventory, using autoequipment system instead and trying to use the best possible items for the current job. Well, most of the time at least, otherwise it would be too easy.

By looking at their choices you can figure out needed stats for any job. I think such indirect ways are the best when you need to demonstrate something hidden.

That's also why earlier in this thread when I was talking about how I send 20+ girls to the courses to level up their stats, all of them were in school for close to 6 months before they even started working at my brothel; They'd come in and immediately be rank 3's with over 100 charisma and 80+ in every sex stat, Strippers would start working with 100+ Charisma, Striptease and maxed agility, Guards would be capped on battle skills even without gear, and Service Girls would have 100+ Charisma, Service and maxed agility.
Starting from the beta stats maxes are limited by levels. You cannot infinitely increase stats without leveling up, and you can't level up without working. Depending on class traits, some stats maxes will increase faster than others, like warriors will get more attack and defense.

You still can get 1000 charisma for a warrior, it just will require much more work than just sending her to beauty course 100 times  :)

Honestly, I think the player not knowing the exact breakpoints would be enough. You could tell the player exactly how much of a stat they'll be gaining and it'll save some people some math to work out what that bonus is, but unless they look at the code, they won't know that they'll need precisely X of a stat for Y situation. They'd only know that trait A gives a bonus to one stat, while trait B gives a bigger bonus to that stat, making it more valuable, or trait C which gives an even bigger bonus to that stat, but comes with a penalty to another stat.
It's not so simple. In the dev version traits modificators work like mod [1,2], meaning +1 to some stat every two levels. At this point we cannot tell directly the bonus because it will be different every time. Especially if some another trait gives [-1 ,3] to the same stat.
At best we can say that a trait gives some bonus to a stat.

Offline Sarabada

  • Newbie
  • *
  • Posts: 30
Re: Fun game, but I seem to have some issues with it..
« Reply #40 on: October 21, 2016, 05:00:51 PM »
Of course, all body types are mutually exclusive. I suppose in theory it's possible to find a small yet chubby character, but it's too much trouble for only a hypothetical rare case.

I suppose if you're talking about finding images to specifically portray the various combinations, then yeah that may be difficult to do, although realistically, that combination is possible. I suppose if you wanted the images to actually reflect the body type to a moderate degree of accuracy, then some concessions will have to be made regarding the flexibility of how 'random' these permutations can be.

We have a more strict system now. Every character must have a body trait, a boobs trait, a personality trait, an elemental trait, a race trait and a class trait. If they are not set in the character file, the game will set default ones.

This sounds great to me; It seemed strange to me in the alpha that some girls would have absolutely no boob trait at all, as I figured it would at least go with the 'average' size as default. Still, it sounds like the groundwork is there for multiple 'layers' of trait picking already, and we can slot in the 'misc' pool and others after all that.

One other hope is that there will be enough 'random' girl types to comfortably showcase all available traits without having to resort to unique characters using them -- so far in my current game, I'm having an extremely hard time finding anyone else with the 'small boobs' trait besides the one Touhou girl, which makes it very annoying when customers decide they want to look for small boobs 7 times in one day.

I attached traits files to this post. They are still a work in progress though, especially class traits.

Thanks, I'll take a look over the weekend and see what I can figure out or come up with. I expect this weekend is going to be a busy one again, though, so no promises yet.

Starting from the beta, characters will have auto equipment function. Meaning that unless character has very high disposition or is a slave, she won't allow you to manage her inventory, using autoequipment system instead and trying to use the best possible items for the current job. Well, most of the time at least, otherwise it would be too easy.

By looking at their choices you can figure out needed stats for any job. I think such indirect ways are the best when you need to demonstrate something hidden.

Oh, okay, this sounds interesting. Some amount of inference will be needed then (especially as some gear usually has unrelated stats, too, but I assume in this case it should still be clear which stats the girls are prioritizing based on which stats have the most bonuses stacked, and the ones that have the least or even negative penalties applied probably have little to do with their job.) This also makes it an even worse decision to try and make girls do stuff that they aren't supposed to, unless they swap them automatically to match the job they'll be doing? Do these come from the girls' own inventories, or the MC's?

On the topic of dropping hints, I feel they could also casually drop hints for the player through their dialogue, if the MC interacts with them. For example, a service girl could, in order to hint at agility, comment that her shoes make it difficult to maneuver around all the guests she needs to serve. My thinking is that such a line would only be available if her current agility is more than, say, 10% away from her current maximum agility cap, and would also serve as a reminder to the MC that the girls may need new equipment to get the most out of them.

Starting from the beta stats maxes are limited by levels. You cannot infinitely increase stats without leveling up, and you can't level up without working. Depending on class traits, some stats maxes will increase faster than others, like warriors will get more attack and defense.

You still can get 1000 charisma for a warrior, it just will require much more work than just sending her to beauty course 100 times  :)

Damn. There goes my 'high-class' brothel. But that said, I think it's a good change, and also makes it more consistent with how the MC has to level up often to raise his stat caps. There are some problems this brings up though, and so I do have a couple questions about such a system:

  • If you are training fresh prostitutes, their low rank will mean they cannot acquire work in a higher level brothel, unless this system has changed going from alpha to beta. If not, they can't level up because they have no work, unless you feel the player must effectively be forced to buy and maintain one of the cheaper brothels just for the sake of level up these 'new' workers.
  • I sincerely hope experience gain scaling is fixed for Guards, Strippers, and Service Girls... otherwise you have what was shown in my previous screenshots - a girl who's guarded continuously for 2 years is only level 13, while prostitutes hired nearly a year after her are over level 100.
  • Are courses going to be useful still, or are you guys trying to discourage their use, since stats would be capped?
  • Is fatigue going to remain at the current static 300 cap, or will it also start off low and increase with levels?

It's not so simple. In the dev version traits modificators work like mod [1,2], meaning +1 to some stat every two levels. At this point we cannot tell directly the bonus because it will be different every time. Especially if some another trait gives [-1 ,3] to the same stat.
At best we can say that a trait gives some bonus to a stat.

Well, displaying the numbers itself is not the problem here, but rather your example is trying to provide the total bonus, which is unnecessary; The player can work that out on their own as long as the numbers are given. Though, again, at the very least, and indication of some bonus being given is better than nothing at all.

I just whipped up another quick edit of the girl profile screen in the alpha to try and test how adding extra information to a trait would look (whether it'd be too cluttered or not, etc.) I say 'quick,' but it was actually really annoying figuring out how to make the XML description field give me line breaks. On the other hand, I managed to get it working, as well as confirmed that it would parse ren'py text style commands just fine. I've attached a screenshot showcasing a very rough mockup of what it would look like to include trait information.

Offline DarkTl

  • Hero Member
  • *****
  • Posts: 4737
Re: Fun game, but I seem to have some issues with it..
« Reply #41 on: October 22, 2016, 04:02:59 AM »
One other hope is that there will be enough 'random' girl types to comfortably showcase all available traits without having to resort to unique characters using them -- so far in my current game, I'm having an extremely hard time finding anyone else with the 'small boobs' trait besides the one Touhou girl, which makes it very annoying when customers decide they want to look for small boobs 7 times in one day.
Yeah, not sure what to do about it. Building a random pack for this trait is difficult, not enough "flat" pictures.

This also makes it an even worse decision to try and make girls do stuff that they aren't supposed to, unless they swap them automatically to match the job they'll be doing? Do these come from the girls' own inventories, or the MC's?
MC can give them items and take them back, but characters may ignore them. They only use their inventories. They swap them automatically to match the job.

I sincerely hope experience gain scaling is fixed for Guards, Strippers, and Service Girls... otherwise you have what was shown in my previous screenshots - a girl who's guarded continuously for 2 years is only level 13, while prostitutes hired nearly a year after her are over level 100.
Realistically, fighting drunk customers once per week is not a good source of exp for a warrior. There will be exploration module, when you send fighters to explore dangerous areas. It will be a much better source of exp. For other classes, some balancing will be needed, yeah.

Are courses going to be useful still, or are you guys trying to discourage their use, since stats would be capped?
You still should increase stats somehow. Either via items or courses.

Is fatigue going to remain at the current static 300 cap, or will it also start off low and increase with levels?[/li][/list]
No, increasing it infinitely is dangerous for the balance. On  the other hand, there are means to get permanent bonuses to it.
Also, different body traits give different starting fatigue (we renamed it to vitality btw).

Offline Sarabada

  • Newbie
  • *
  • Posts: 30
Re: Fun game, but I seem to have some issues with it..
« Reply #42 on: October 22, 2016, 12:28:39 PM »
Yeah, not sure what to do about it. Building a random pack for this trait is difficult, not enough "flat" pictures.

Perhaps not for the existing girls, as I suppose most of the elf-type characters are typically portrayed as fairly busty, but the 'flat' archetype should be popular enough that there should be another character you could pick to use, although this would involve creating another 'class' of character. But without it, you'd be unable to really appease customers who decide they want that trait unless you mod in a bunch of unique girls with that trait.

MC can give them items and take them back, but characters may ignore them. They only use their inventories. They swap them automatically to match the job.

All right, that sounds good to me. I assume disposition plays a large role in whether or not the girls will equip your items, but the auto swapping to match the job feature sounds great.

Realistically, fighting drunk customers once per week is not a good source of exp for a warrior. There will be exploration module, when you send fighters to explore dangerous areas. It will be a much better source of exp. For other classes, some balancing will be needed, yeah.

I agree, of course, and I wanted to make it so that the 'aggressive customer' event isn't totally eliminated at max security rating in the alpha largely because such a system wasn't in place in the alpha, so they had no other options. However, I still stand by the idea that even with this system in place, 1000 security rating should not be 100% chance to avert such events. My idea was to have max security rating cap out the reduction at something like 90 or 95%, then the security presence rating plus number of guards would bring an additional 4-9% reduction, so that even at max security rating, 100% presence and 5 guards, there would still be a small 1~5%ish chance of these events occurring, necessitating you actually placing more than one guard on duty at 1000 security rating.

Once I figure out how to work out the code properly I'll write that into the alpha (along with their random 'customer fining' and 'item confiscation' events) and see how well it works.

You still should increase stats somehow. Either via items or courses.

Hmm, okay. Sounds like you'd have to pull active staff out to run courses more often since you can't simply have them sit in a course for 6 months right away. What do you guys think about my idea of allowing these courses to have a chance of granting additional traits at their completion?

No, increasing it infinitely is dangerous for the balance. On  the other hand, there are means to get permanent bonuses to it.

Also, different body traits give different starting fatigue (we renamed it to vitality btw).

Oh, I didn't expect it to increase infinitely, but I was thinking more that 300 wouldn't be a 'hard' cap, but a 'soft' one. In other words, most girls would start off lower, but eventually cap out at 300, except the girls that have traits with fatigue bonuses which would allow them to go past 300 to whatever their bonus is, so you'd have those particular girls at maybe 350 or 400 fatigue max, making those traits actually very valuable. At the least, I'm glad to hear their starting fatigue/vitality will be different, which sounds like at least part of this suggestion is already implemented.

On that note, I took a quick look at the db files and I think I understand what your meaning was; It seems like in addition to a trait providing the static minimum and maximum bonuses, they also provide a 'per X level' bonus. This would be extra information to convey, but is still very doable in a similar manner as I did in my previous screenshot, though.

Offline Sarabada

  • Newbie
  • *
  • Posts: 30
Re: Fun game, but I seem to have some issues with it..
« Reply #43 on: October 26, 2016, 07:28:51 PM »
Welp, another busy weekend, but getting back to poking at this stuff now.

I'm still piecing together exactly what the new traits you sent me really mean, but I do have a couple ideas to throw in. I just haven't quite worked out what exactly they would do yet because I am still not quite clear on what the final list of available stats will be, nor what stats are relevant to each job, so some ideas I have may overlap with existing traits, which would end up being redundant. For example, there could be a "Spoiled" trait, but that could potentially be very similar to "Noble" depending on what stats I can reasonably affect.

If there's a way you can send me that information, I can try to work out more specifics for you.

Going back to the alpha, I just finished an initial pass over all the random girl generation JSON files. I tweaked all of their traits, as well as added any from the full list of traits that worked with their archetype/description, and on a few quick tests, it seems to definitely be aiding random girl diversity immensely. The average number of traits a random girl had previously (stock alpha) was around 4. Once the updated JSON files were loaded, the average was around 9 (with many girls having a guaranteed boob trait as well.) I may have gotten slightly overzealous though, as some of the girls had upwards of 15 traits, which is probably excessive, so I'll be going through them and reducing the chances for some of their traits. Still, simply having more of them available, even at a 10% chance, allows for greater variation, as they will usually have different traits to each other.

To enhance the diversity between girls, I also went and added more specific override ranges for each girl's starting stats (everything from primary stats, to sex stats and battle stats, and even luck.) I tweaked them so that certain girls would be a better 'fit' for specific roles, hopefully consistent with their description/archetype. For example, the Sea Elves are said to be cultivated into a race of maids, so on top of a 100% chance of 'Professional Maid,' they also start with much higher Service (40+) but at the cost of having very abysmal combat stats (Attack, Defense and Magic are set to start at between 5-10.)

Meanwhile, the Soldier class has no magic (stat is 0-5) and is pretty awful for most brothel work initially, but has very high combat ability (45+ Attack and Defense, and 25 Agility) and starts with increased reputation and fame. The adventurer is more balanced, but most stats are more average as a result, while the Priestess/Sage have a higher Magic stat with low Attack and Defense stats. The Angel class is great for combat, and has high charisma to boot, but their lack of knowledge of the human realm means they start off barely any better than Soldiers are for brothel work.

I did, however, run into some issues here, which means I'll have to revisit the JSON files again anyway; Due to the random girl generation code for some reason modifying the randomly chosen value from each stat's specified ranges, the values that end up in the game are usually much higher than I expected/wanted them to be. In one case, though, I think it was with a Slime girl who starts with very poor stats almost across the board and then picked up more of the negative traits on top of that, the girl wound up with some negative stats. I'm trying to reproduce it now so I can figure out how to avoid that.

In doing so, I also did verify that using devlog.info() was enough to output data into the log for me, so that's good to know. I ran into multiple errors the first couple times I tried to load the game after the JSON edits due to various typos in them. The first error spoke of expecting comma delimiters, which, while not specific enough to pinpoint which file caused it, was enough for me to run a regex search through all the files and fix the half dozen or so missing commas across all of the files. Beyond that, I ran into errors where the code's natural error reporting failed to tell me anything meaningful (something about dict not having an index, but the error report code was unable to fill in the variables.)

I threw in some code to output the values being used at each step of the function to devlog.info() and immediately discovered two more typos with the traits (one was "Silly]" which I must have accidentally added that bracket to in one of the files.) Fixed those and off we go to try and rebalance these numbers now..

Offline Xela

  • Global Moderator
  • *****
  • Posts: 6893
  • "It's like hunting cows"
Re: Fun game, but I seem to have some issues with it..
« Reply #44 on: October 26, 2016, 09:48:09 PM »
Good luck, it seems like an unreasonable amount of effort put into very old version of the game but as long as you're having fun.
Like what we're doing?