HentHighSchool Development Forum

Please login or register.

Login with username, password and session length
Advanced search  

News:

collapse
* Recent Posts
Re: *NEW* Tagging tool by Goldo
[Yesterday at 11:07:04 pm]


Where do you get wallpapers? by GonzaloMad
[Yesterday at 06:33:37 pm]


Re: *NEW* Tagging tool by Chris12
[Yesterday at 03:16:45 pm]


Re: Town-only run by harmi
[Yesterday at 03:08:20 pm]


Re: Town-only run by Dante
[Yesterday at 03:29:46 am]


Pages: [1] 2   Go Down

Author Topic: My thoughts on the game so far  (Read 3392 times)

DrMabuse

  • Newbie
  • *
  • Offline Offline
  • Posts: 17
    • View Profile
My thoughts on the game so far
« on: January 06, 2019, 04:03:53 am »

I've played the game and i really like it. But there are some weaknesses/unbalances in my opinion:

1. The job refusal problem mentioned in another thread. Before reading that thread, i had absolutely no clue why total obidient girls refuse to work. And i don't think it's a feature that girls that learned sexual skills won't work in an ordinary job. Then there would be mechanics that are useless, like the last Perk in the "Whore-Tree" or the the requests from customers to have sex with the girls in a job.

2. In addition to the refusal, the continous "attacks" (especially late game) on the brothel make ordinary jobs nearly useless. When i figured it out right, whenever a girl is (or would be) attacked, the job is skipped. Since whores split there job in many encounters with customers at one night, they will work eventually. But since ordinary jobs are handled in one encounter, the whole shift for that girl is "lost".
In my game, my girls with ordinary jobs hardly worked because of this (and i had maximum security with a thread on low or very low). Conclusion: With the problems mentioned under 1. and the problems under 2., it's not worth to train girls in ordinary jobs, because they don't make money most time.

3. Missions: In my opinion, the rewards from the missions are most times too bad. I hardly got a mission that was worth the cash to send a girl away that long. The classes aren't much better either. You need them later on to get reputation with low level girls, but 3 days or something like that for 1 reputation? Way too long in my opinion. The missions should get better with advancing in chapters (if they already do, noticeable).

4. Please, add a skip button for the KOSMOS encounter. No one wants to read the same text a hundred times.

5. Playing evil is a pain in the ... . It takes way more micromanagement and ingame time to get similiar results as beeing the good guy. I think it should be a fun/taste choice if someone wants to be the good/bad guy, not a hidden difficulty level.

6. The "Fear-spell" for evil players doesn't seem to do anything (or maybe i misread it's effect?). I tried autocast at night and at morning, but fear decresed anyway over night. For sure, it never increased, and i checked a lot on that.

7. Collecting resources seems bugged. It is possible to get in the negative APs to collect them.

8. Hypnosis: Why does it get more expensive when i live in another part of the city? That doesn't make sense to me. I think the price should remain or depend on the girl level, not on the chapter level the player is in.

Hmmm, i had more, but i forget it. When i remember, i willl add it.

« Last Edit: January 06, 2019, 04:16:04 am by DrMabuse »
Logged

Neptune

  • Newbie
  • *
  • Offline Offline
  • Posts: 14
    • View Profile
Re: My thoughts on the game so far
« Reply #1 on: January 06, 2019, 08:40:46 pm »

I strongly agree with number 4.
Logged
Hey, if you see Triton anywhere tell her she needs to pay her goddamn rent.

Chris12

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 86
    • View Profile
Re: My thoughts on the game so far
« Reply #2 on: January 07, 2019, 11:53:21 am »

Regarding 1 and 2, I do not quite understand. If security threat is low or very low, your girls should not get beat up. At least that's how it worked under the old security system. Maybe the new experimental patch solves your problem?

Regarding 3, Missions, their scaling is rather weird right now. At first, Missions are as weak as you describe, but in the lategame, a single one can level up a girl from Lvl15 straight to Lvl20. Goldo is aware of this, but it's probably not that high up on the priority list. Personally, I'd prefer to see a buff for the farm, which does not seem to pay off at any point in the game.

Regarding 4, go to Preferences and set "Skip All Messages". After that, you can press CTRL to fast forward through any text.
Logged

Dizarus

  • Newbie
  • *
  • Offline Offline
  • Posts: 3
    • View Profile
Re: My thoughts on the game so far
« Reply #3 on: January 07, 2019, 10:54:06 pm »

I strongly agree with number 4.
Press CTRL and it will be gone in a second.
Logged

Tomas1221

  • Newbie
  • *
  • Offline Offline
  • Posts: 34
    • View Profile
Re: My thoughts on the game so far
« Reply #4 on: January 08, 2019, 09:25:37 am »

4nr is already added so no problem. nr 1 is mess you mentioned like 5 diferent thing in one colum
only one thing i noticed is lack of usefull thing in the shop at end of game. and i want rest of the capters. sorry for broken english
Logged

Goldo

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 756
    • View Profile
Re: My thoughts on the game so far
« Reply #5 on: January 08, 2019, 08:28:50 pm »

Some answers below

I've played the game and i really like it. But there are some weaknesses/unbalances in my opinion:

1. The job refusal problem mentioned in another thread. Before reading that thread, i had absolutely no clue why total obidient girls refuse to work. And i don't think it's a feature that girls that learned sexual skills won't work in an ordinary job. Then there would be mechanics that are useless, like the last Perk in the "Whore-Tree" or the the requests from customers to have sex with the girls in a job.

People have mentioned this before but no one has provided me with a saved game that would allow me to investigate (or better yet, a theory as to why this is happening to some players some time). I cannot fix problems that I cannot reliably reproduce.

Quote
2. In addition to the refusal, the continous "attacks" (especially late game) on the brothel make ordinary jobs nearly useless. When i figured it out right, whenever a girl is (or would be) attacked, the job is skipped. Since whores split there job in many encounters with customers at one night, they will work eventually. But since ordinary jobs are handled in one encounter, the whole shift for that girl is "lost".
In my game, my girls with ordinary jobs hardly worked because of this (and i had maximum security with a thread on low or very low). Conclusion: With the problems mentioned under 1. and the problems under 2., it's not worth to train girls in ordinary jobs, because they don't make money most time.

Those issues are already addressed with the experimental patch, which is close to becoming stable enough to make it into the vanilla game.

Quote
3. Missions: In my opinion, the rewards from the missions are most times too bad. I hardly got a mission that was worth the cash to send a girl away that long. The classes aren't much better either. You need them later on to get reputation with low level girls, but 3 days or something like that for 1 reputation? Way too long in my opinion. The missions should get better with advancing in chapters (if they already do, noticeable).

Good point; it'd be great if someone had the time to crunch some numbers and tell me what they think the right values would look like (both at the beginning and towards end-game)

Quote
4. Please, add a skip button for the KOSMOS encounter. No one wants to read the same text a hundred times.

The skip button (for everything) is Ctrl. If it doesn't work, you might have to tinker with preferences, but I am 99% sure it is active by default for every message you've already seen. If that is still too slow, I would consider adding a skip option to this particular event, but try it first.

Quote
5. Playing evil is a pain in the ... . It takes way more micromanagement and ingame time to get similiar results as beeing the good guy. I think it should be a fun/taste choice if someone wants to be the good/bad guy, not a hidden difficulty level.

I agree. I want to make an evil play through a more valid option and am open to suggestions.

Quote
6. The "Fear-spell" for evil players doesn't seem to do anything (or maybe i misread it's effect?). I tried autocast at night and at morning, but fear decresed anyway over night. For sure, it never increased, and i checked a lot on that.

The fear notification is a little broken as you only see the latest adjustment (and fear naturally decreases very little every night). It doesn't mean other positive fear effects aren't happening. I will probably take out the notification to avoid misunderstandings.

Quote
7. Collecting resources seems bugged. It is possible to get in the negative APs to collect them.

It sounds like something that is fixed in the experimental patch. Please let me know if that isn't the case.

Quote
8. Hypnosis: Why does it get more expensive when i live in another part of the city? That doesn't make sense to me. I think the price should remain or depend on the girl level, not on the chapter level the player is in.

Good point. I will change the mechanic so that it works that way.

Quote
Hmmm, i had more, but i forget it. When i remember, i willl add it.

Sure! :)
Logged

Dragonblood

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1412
  • Code monkey
    • View Profile
Re: My thoughts on the game so far
« Reply #6 on: January 09, 2019, 08:25:30 am »

Some answers below

I've played the game and i really like it. But there are some weaknesses/unbalances in my opinion:

1. The job refusal problem mentioned in another thread. Before reading that thread, i had absolutely no clue why total obidient girls refuse to work. And i don't think it's a feature that girls that learned sexual skills won't work in an ordinary job. Then there would be mechanics that are useless, like the last Perk in the "Whore-Tree" or the the requests from customers to have sex with the girls in a job.

People have mentioned this before but no one has provided me with a saved game that would allow me to investigate (or better yet, a theory as to why this is happening to some players some time). I cannot fix problems that I cannot reliably reproduce.

After playing around in a spreadsheet a little bit, I think it might be due to the formula in BKgirlclass.get_obedience_check_target()

Specifically, with a "maxed" Rank C girl for a good-aligned character (50 Obedience, 100 Love, 100 Mood, and fascinated by all sex acts), I'm getting a ~3% chance to work versus being guaranteed (i.e., >100% chance before being capped) to whore.


Even more specifically, I think the problem comes from negative fear / positive love values. For comparison, with all else being 0:

Obedience 0 -> 50 changes work chance 65.0% -> 95.5%
Obedience 0 -> 50 changes whore chance (indifferent) 30.0% -> 90.9%

Fear 0 -> 100 changes work chance 65.0% -> 99.4%
Fear 0 -> 100 changes whore chance (indifferent) 30.0% -> 98.8%

Love 0 -> 100 changes work chance 65.0% -> -1974.6%
Love 0 -> 100 changes whore chance (indifferent) 30.0% -> -4049.3%

The only thing available for a Rank C girl that seems to overcome that staggeringly massive penalty is the sex preference modifier, and even then just barely -- the girls will still only whore if they are at Fascinated or (for maxed or nearly-maxed Rank C girls) Very Interested.

Also, this is before any Obedience Target effects. Those would probably help somewhat, but unless they are able to get the coefficient down to zero or negatives, it won't fully solve the problem.

This problem largely disappears once the Obedience stat gets to 100 or over, but for Rank C girls, it's a very noticeable breakdown in the formula. It looks like it's basically just an issue of the "mod" value going negative and then being used as an exponent, which is creating a far larger variable than intended. So, maybe just checking if the "mod" value is negative before doing the exponential part of the formula, and using an alternative formula instead in that case? (Edit: Or adding in a static base value, like +100, to make sure it never goes negative? Would definitely need to crunch numbers to figure out what else would need to change to stay balanced.) I've got a head cold at the moment, so I'm really not wanting to dig into this much more tonight; even just this has taken me an hour or two.
« Last Edit: January 09, 2019, 08:29:38 am by Dragonblood »
Logged

Chris12

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 86
    • View Profile
Re: My thoughts on the game so far
« Reply #7 on: January 09, 2019, 10:54:23 am »

I took a look at get_obedience_check_target() and I think you missed the if train, probably due to your cold. I think the calculation in question is this:
mod = self.get_stat("obedience") + self.get_fear() + self.mood//4
which is always used unless you're training and she's in love with you.

Still, you could well be right that the problem is here. AFAIK, fear can go down to -100, if she trusts you a lot. Due to the exponential growth of 0.96 ** mod, with negative values for mod, this leads to quite staggering values for what must be intended as a slight decrease to the coeff. E.g. 90 trust and 20 obedience ->
target = (0.96 ** (-90+20)) * 70 = 17.4 * 70 = ~1200. Which is quite impossible to reach with rand(100).

Edit: Even a mod of just -10 already leads to 70*1.5 = 105, which is already impossible.

Good catch.
« Last Edit: January 09, 2019, 11:00:26 am by Chris12 »
Logged

Dragonblood

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1412
  • Code monkey
    • View Profile
Re: My thoughts on the game so far
« Reply #8 on: January 10, 2019, 10:35:07 am »

I took a look at get_obedience_check_target() and I think you missed the if train, probably due to your cold. I think the calculation in question is this:
mod = self.get_stat("obedience") + self.get_fear() + self.mood//4
which is always used unless you're training and she's in love with you.
...yes, and? That's the formula I was using in the spreadsheet I made all those calculations on...

Edit: Even a mod of just -10 already leads to 70*1.5 = 105, which is already impossible.
Keep in mind that the girl's sex act preferences factor into the coefficient value. 70 would only be for if the girl is Indifferent for all the sex acts she has enabled. If she is Fascinated for all enabled sex acts, that's adding on a whopping -75 modifier to that coefficient value, bringing it down to -5 overall (which is also why the girls are still perfectly willing to whore in the related bug reports -- they've been trained extensively, so that huge modifier to the coefficient is causing the result's sign to get inverted, leading to a greater-than-100% chance to whore).
Logged

Chris12

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 86
    • View Profile
Re: My thoughts on the game so far
« Reply #9 on: January 11, 2019, 06:43:34 am »

Sorry, my bad, I worded it badly. What I meant to say is that Fear/Trust is the "Culprit", not Love. Other than that, I mostly agree with you, anyway. ^^

A few more thoughts regarding this:

*) Maybe a simple linear progression would also work here? The only advantage to the exponential part I see is that it never goes into the negative. But with coeff being able to do so anyway, all you get is the usual trouble with exponential growth. (It grows explosively, the human mind is bad at estimating it, ...)

*) On the other hand, this may even be somewhat intentional. Even with a linear function, if coeff is 70, negative values for mod would still lead to high target values and probably 100% refusal at some point. Maybe the biggest problem here is that the "culprit" is not easy to identify. It would probably help if the message was not only "... refused to work", but something like "... refused to work because is not afraid of you at all." or something like that. To be honest, I was not aware that I had to offset trust with obedience before reading your post.

It would mean that you cannot just play Mr. Nice Guy from the start, and have to be a little strict at first. Adjusting this a bit might also be a way to buff playing evil, or nerf playing good.

*) The sex act preference is another good point. Kinda weird that whoring gets easier than working. If it is as you say, you're even guaranteed a 0% refusal chance. Actually, there's the same thing in the Job formula: preference_modifier[self.get_preference(act)]. But I never noticed anything like that in the GUI. Probably just returns 0 at this point?
Logged

Dragonblood

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1412
  • Code monkey
    • View Profile
Re: My thoughts on the game so far
« Reply #10 on: January 11, 2019, 07:57:34 am »

What I meant to say is that Fear/Trust is the "Culprit", not Love.
Ah, I see. You're right. I was getting trust and love mixed-up in my mind. I used the right formula but the wrong term. :)

The only advantage to the exponential part I see is that it never goes into the negative.

The main advantage to the diminishing exponential is that it makes it so that there is a drastic change early on, and very little change later. You wouldn't really expect there to be much of a difference in the rate of refusing to work between a Rank A and Rank S girl, for example -- at that point, they should be pretty well-trained to obedience if they are anywhere even approaching their stat cap. A fresh-off-the-street girl, on the other hand, could be expected to be much more inclined to just outright refuse to work.

On the other hand, this may even be somewhat intentional.

Unless Goldo weighs in saying otherwise, I'm going to assume not, on the basis that I've made similar mistakes in the past myself. ;)
(It's really easy to forget that a value in a complex calculation can potentially end up negative or zero when it typically will be positive.)

The sex act preference is another good point. Kinda weird that whoring gets easier than working.

Yeah, I personally think I'd rather have it be just a slightly lower-than-when-working coefficient rather than an auto-pass at Fascinated.
Logged

Goldo

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 756
    • View Profile
Re: My thoughts on the game so far
« Reply #11 on: January 12, 2019, 02:56:45 pm »

Thank you guys, this is great insight.

Here are the changes I'm considering:

Whoring
I think I'll set a floor at 30 for the 'coeff' value when whoring as per Dragonblood's suggestion. This means it won't ever be much lower than the work obedience target.

Trust effect on obedience
I want to keep the idea that being too 'soft' with your girls at the beginning when they are not obedient makes them unruly, but some adjustment is needed of course.

Idea #1: Set a hard ceiling on the obedience target (maybe 70?) for working, so that even unruly girls don't completely stop working.

Idea #2: Have dynamic positive and negative boundaries for love and fear according to girl rank (say, from 25*rank to -25*rank). This could avoid a number of other balance issues at the beginning, but it could also break some things, maybe.

Idea #3:
Quote from: Dragonblood
(Edit: Or adding in a static base value, like +100, to make sure it never goes negative?
This is also a good way to deal with it, but I would then need to change the formula to offset this, and uh... The math involved isn't clear to me.


Logged

Goldo

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 756
    • View Profile
Re: My thoughts on the game so far
« Reply #12 on: January 12, 2019, 02:58:06 pm »

Thank you guys, this is great insight.

Here are the changes I'm considering:

Whoring
I think I'll set a floor at 30 for the 'coeff' value when whoring as per Dragonblood's suggestion. This means it won't ever be much lower than the work obedience target.

Trust effect on obedience
I want to keep the idea that being too 'soft' with your girls at the beginning when they are not obedient makes them unruly, but some adjustment is needed of course.

Idea #1: Set a hard ceiling on the obedience target (maybe 70?) for working, so that even unruly girls don't completely stop working.

Idea #2: Have dynamic positive and negative boundaries for love and fear according to girl rank (say, from 25*rank to -25*rank). This could avoid a number of other balance issues at the beginning, but it could also break some things, maybe.

Idea #3:
Quote from: Dragonblood
(Edit: Or adding in a static base value, like +100, to make sure it never goes negative?
This is also a good way to deal with it, but I would then need to change the formula to offset this, and uh... The math involved isn't clear to me.  :'(
Logged

Dragonblood

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1412
  • Code monkey
    • View Profile
Re: My thoughts on the game so far
« Reply #13 on: January 13, 2019, 11:26:48 pm »

Well, the good news is that I'm feeling mostly better, so I can probably help with the formula tweaking now. Just some questions first:

Idea #1: Set a hard ceiling on the obedience target (maybe 70?) for working, so that even unruly girls don't completely stop working.
Just to make sure we're on the same page, when you say "obedience target," you are referring to the returned value from BKgirlclass.get_obedience_check_target() ?

Idea #2: Have dynamic positive and negative boundaries for love and fear according to girl rank (say, from 25*rank to -25*rank). This could avoid a number of other balance issues at the beginning, but it could also break some things, maybe.
Could you clarify where you were thinking of having this applied to? I'm assuming that you mean to the love / fear values in the formula for the exponent?


I'll try to work out formula changes for each of the 3 ideas proposed later on, and give you some example return values for different scenarios to help compare them.
Logged

Dragonblood

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1412
  • Code monkey
    • View Profile
Re: My thoughts on the game so far
« Reply #14 on: January 14, 2019, 09:36:00 am »

Okay, going with my assumptions outlined in the above post so I can get some initial data crunched. Correct me if any of them were mistaken. :)


So, idea #1 (ceiling of 70 for obedience target): It helps, but not much IMO. There's still a lot of sharp discontinuities where the will-work chance suddenly plummets by 20+% over a very narrow stat range, it's just that the minimum is a 30% chance to work instead of it being literally impossible for them to work.

Idea #2 (rank-based limits on love/fear effect): It's... weird. It helps (quite significantly at times) in some stat ranges, has no effect in others (i.e., still literally impossible for girls to work), and lowers chances in other ranges.

Idea #3 (add a flat bonus so that exponent is never negative): This one is turning out to be quite hard to balance. I'll keep looking into it, though, and see if I can come up with anything workable.


After looking at the other results, though, I came up with one more possible formula change:
Increase the exponent base to 0.98
Add a flat +50 to the exponent value
Halve the love/fear effect in the exponent value.

This actually lowered the chance to work slightly across the board, and significantly, it added a very small (but still possible) chance for girls of any stat combination to not work:
* At 100 obedience / 100 fear / 100 mood, a 0.32% chance to not work @ Fascinated or 0.74% @ Indifferent;
* At 250/100/100, a 0.015% chance @ Fascinated or 0.036% @ Indifferent;
* At 300/100/100, a 0.0056% chance @ Fascinated or 0.0131% @ Indifferent

There are still stat combinations with no chance of the girl working, but not if she has a high mood and/or obedience. Even just a high Rank C obedience (50/-100/-100) still gives a 33.62% chance to work even with a coef as high as 110, as does 0/-100/100; being Fascinated brings that up to a 81.9% chance.

A "perfect" Rank C girl for a good player (50/-100/100) has a 75.83% chance to work (versus 0% chance currently, 30% chance with Idea #1, 60.36% with Idea #2, or 91.25% with Idea #3).

Personally, I'm leaning towards this being the best solution. It makes it so that there's still some variation even at endgame (you'll very rarely have girls refuse to work; good-aligned players should see an average of about once per month per 11 girls if they are around 200 obedience and are at Fascinated, if my math is correct), but early game is still feasible for all play styles.

If you want more details about the return values for any of these, just let me know.
« Last Edit: January 14, 2019, 09:42:46 am by Dragonblood »
Logged
Pages: [1] 2   Go Up
 


SimplePortal 2.3.3 © 2008-2010, SimplePortal