General Discussion

[DarkTl]

I have another good question! Why are you so insist on adding that Rukia mast animation? How and where are you going to use it? 

City map looks pretty good btw. But I like the old one too. Maybe we should use both, one map won't be enough for all locations anyway.

I tested new systems a bit. Badness 100 does not prevent purchase for some reason, they buy Slave Shackles like there is no tomorrow 

Also, there is a huge gasp between owned and equipped items. It's not unusual to see girls with 30+ owned and 0 equipped. I'm not sure I like it. Maybe it's because they have 10000 gold from the start and cannot spend it on something else though. We need more interesting ways to spend gold, I guess.

[Xela]

Hi,


When are we getting the next update? 

Hi

No idea, we don't really have any particular goals set for "what" a next update should be. I think we should finish some base form the guild, enable/expand taxes and then we can release again? Depends on what the rest of the crew feel like, it's not like there are any definite plans.

I have another good question! Why are you so insist on adding that Rukia mast animation? How and where are you going to use it? 

City map looks pretty good btw. But I like the old one too. Maybe we should use both, one map won't be enough for all locations anyway.

I tested new systems a bit. Badness 100 does not prevent purchase for some reason, they buy Slave Shackles like there is no tomorrow 

Also, there is a huge gasp between owned and equipped items. It's not unusual to see girls with 30+ owned and 0 equipped. I'm not sure I like it. Maybe it's because they have 10000 gold from the start and cannot spend it on something else though. We need more interesting ways to spend gold, I guess.

I have another good question! Why are you so insist on adding that Rukia mast animation? How and where are you going to use it? 

No idea, main screen maybe?

City map looks pretty good btw. But I like the old one too. Maybe we should use both, one map won't be enough for all locations anyway.

Yeah we'll see, put it to the vote maybe...

I tested new systems a bit. Badness 100 does not prevent purchase for some reason, they buy Slave Shackles like there is no tomorrow 

Also, there is a huge gasp between owned and equipped items. It's not unusual to see girls with 30+ owned and 0 equipped. I'm not sure I like it. Maybe it's because they have 10000 gold from the start and cannot spend it on something else though. We need more interesting ways to spend gold, I guess.

My screwup, I asked you about the badness after coding it as in WM and forgot to change it. Right now game checks dice(badness) to buy instead of dice(100 - badness) so girls are buying items that are the worst for them. I don't know of that's the reason for the equipment thing, they should try to reequip themselves for the job they're doing one every 3 days or so, may there is a mistake in the code somewhere.

The way auto-buy should work:
- First: dice80 to have the girl try to buy items she'd love because of her traits.
- Second: dice30 for her to try and stock up on some decent consumables.
- Third: She will go through all possible items - badtrait/goodtrait items (badtrait are filtered out at the very start, goodtrait we prolly done already due to the high dice or will do at high probability next time). Here she'll try to but any item, never more than 5 of each, depending on badness of the item (item at badness 0 she'll buy 5 of, badness 80, just one at best).
- That's it, since the function is repetitive, I've decided not to do loop it until she buys as many items as possible, instead she will either buy the required amount supplied to the function or less and try again in a few days.
===========================
I'll be on tonight and should have the whole evening to code. I'll look into the auto-equipment thing and the bug CW found then.

@Klaus:
The long story is too long which wouldn't really be an issue unless it felt like it had too much story in it irrelevant to the game, it's something we should have some ancient NPC dude tell to the players who love lore/history in games. Short story is too short

So tonight I'll set up a label which will load the intro to the game and add a button to run that label from the mainscreen. I'll put the long version with slightly improved grammar (done by me so only slightly improved ) of the long text and we'll all contribute by cutting/adding/animating and etc. (if you don't mind).

[MrKlaus]

@Klaus:
The long story is too long which wouldn't really be an issue unless it felt like it had too much story in it irrelevant to the game, it's something we should have some ancient NPC dude tell to the players who love lore/history in games. Short story is too short

So tonight I'll set up a label which will load the intro to the game and add a button to run that label from the mainscreen. I'll put the long version with slightly improved grammar (done by me so only slightly improved ) of the long text and we'll all contribute by cutting/adding/animating and etc. (if you don't mind).

 Long too long and short too short... Will aim for a mid length after everyone’s tells their opinions (and after finishing those goddamn alignments). 
 And I don't mind so chill, don't worry and modify my txt any way you guys want. If you will only use ONE line from a hundred that I send it's cool with me.
 
Btw. Sorry for the delays, it was a busy week for me.

[lamoli]

Hey, anything decided on slave training module yet ?

[DarkTl]

Hey, anything decided on slave training module yet ?

Lack of clear and decent concept prevents it. Feel free to share ideas.

Make them both PyTFall but split in time. The bright could be the beginning stage, let’s say year 10-20 after establishing the city, and the current dark city would be year 100-120 after PyTFall grew bigger. The transition period could be a prolog (or a tutorial stage) about the MC Father/Grandpa and a chain quest that would lead to some disaster that changes the surrounding (spell went wrong, bloody rebellion suppressed with magic, demon attack or some other shit hit the fan).

But then we'll never be able to use the bright version again. Would be a waste, I like both.
I think about either two parts of the city, light and dark ones, separated by a river or something, or some another city nearby. Not elven one, we already have another cool map for it.

Ideally we could use a decent forest map for EE. Maybe even in the guild you should choose area not from the list, but on the map. It's difficult to find though, perhaps we could render an original one in the future when we'll have another artist willing to help.
========================================================================
I think we left some garbage in gfx folder. Those mobs and old hero profile in sprites folder, frame folder (do you still use those old frames?), masks folder (I think we don't use them at all).

[Xela]

I think we left some garbage in gfx folder. Those mobs and old hero profile in sprites folder, frame folder (do you still use those old frames?), masks folder (I think we don't use them at all).

I generally don't worry about that... Who know what we might need and when.

I found a way to solve the drag/drop screen freezing.

[livingforever]

Hey everyone!
Oh my... seems like I've missed a lot just because I was gone for a few days. I skimmed the last few pages and there are a few things on my mind:

First, the discussion about exploration times.
Actually, this has been on my mind ever since I played the first BrothelSim (the original) - why does nobody consider using a time based system instead of action points? Yes, it would require slightly more code, but I think it is more intuitive and a lot more flexible. Every character (including the player) can have work times (e.g. 8 am to 4 pm for the player, 4 pm to midnight for prostitutes - depending on constitution, etc.) and every action (talking to a girl, having sex, whoring, etc.) takes time to finish.
What do you think about such an approach (general opinions wanted, details can be worked out later)?


Second, about this little note:

- Added decent code to load items from filenames (commented out atm)

Careful with that. The reason why it's reasonable to load tags from filenames is because image tags are unordered sets of the same unnamed data type - so a complex data structure would be wasted on it.
Items however consist of several different, named elements with different data types, so using a more complex data structure (JSON is well suited) is most likely the better choice here.

And yes, binary is faster (it always is if you can come up with a clever format), for both tags and items. However, the primary goal was to reduce storage redundancy, not to make the data unreadable for humans.


Third, the optimization of reading tags from filenames.
I just want to mention that the screenshot looks absolutely awesome to me. So much data at a single glance!
Just from looking at it I assume you have two static identifiers and 260 (26 chars * 10 digits) possible tag combinations that are not ordered alphabetically. How many of these are used right now? And, more importantly, do you have a hashing system in place (that calculates an array index that stores the actual tag from the tuple)? If yes, great! If no, that could speed things up.


Finally, the tagger.
Let me ask this first: Who is mcmanus and where can I find his code? C++ is no big deal and I at least want to give it a look (maybe I can steal his ideas  ).
And to give you a (relatively) clear statement: I can and will code a tagger (and possibly a complete character editor) if you are fine with an external solution that has nothing to do with Python. No deadlines and promises (freelancing ftw).

Have fun!

[Xela]

Hey everyone!
Oh my... seems like I've missed a lot just because I was gone for a few days. I skimmed the last few pages and there are a few things on my mind:

First, the discussion about exploration times.
Actually, this has been on my mind ever since I played the first BrothelSim (the original) - why does nobody consider using a time based system instead of action points? Yes, it would require slightly more code, but I think it is more intuitive and a lot more flexible. Every character (including the player) can have work times (e.g. 8 am to 4 pm for the player, 4 pm to midnight for prostitutes - depending on constitution, etc.) and every action (talking to a girl, having sex, whoring, etc.) takes time to finish.
What do you think about such an approach (general opinions wanted, details can be worked out later)?

RPGMaker variant of the game does just that, can't say that I like the idea for PyTFall. The concept itself is fun but we don't have the resources to make maps/locations adapt to the time of day, not mentioning the amount of code that would have to be rewritten (Although it's doable, just will take a good while).

Second, about this little note:Careful with that. The reason why it's reasonable to load tags from filenames is because image tags are unordered sets of the same unnamed data type - so a complex data structure would be wasted on it.
Items however consist of several different, named elements with different data types, so using a more complex data structure (JSON is well suited) is most likely the better choice here.

That was a typo, I never did that for items, meant to say pics.

And yes, binary is faster (it always is if you can come up with a clever format), for both tags and items. However, the primary goal was to reduce storage redundancy, not to make the data unreadable for humans.

And speed things up, but filenames are fine + we can always go binary later.

Third, the optimization of reading tags from filenames.
I just want to mention that the screenshot looks absolutely awesome to me. So much data at a single glance!
Just from looking at it I assume you have two static identifiers and 260 (26 chars * 10 digits) possible tag combinations that are not ordered alphabetically. How many of these are used right now? And, more importantly, do you have a hashing system in place (that calculates an array index that stores the actual tag from the tuple)? If yes, great! If no, that could speed things up.

There are no static identifiers, first four characters are random to prevent name collisions, rest are pairs picked from a combination from lowercase latin letters and 0 - 9 numerals.

>>> from itertools import combinations_with_replacement
>>> import string
>>> pool = list("".join([string.ascii_lowercase, "".join(list(str(i) for i in range(10)))]))
>>> len(list(combinations_with_replacement(pool, 2)))
666 (We're totally evil   )

165 are in use at the moment so we can still add 501 tags  Yet, there are plenty more characters that are allowed in filesnames on all systems so we can always add even more

I am not sure how you want to speed this up with hashing failing to come up with a way myself.

Finally, the tagger.
Let me ask this first: Who is mcmanus and where can I find his code? C++ is no big deal and I at least want to give it a look (maybe I can steal his ideas  ).
And to give you a (relatively) clear statement: I can and will code a tagger (and possibly a complete character editor) if you are fine with an external solution that has nothing to do with Python. No deadlines and promises (freelancing ftw).

Oki, I'll focus on SE + fixing errors.

Edit: New tagger in C#

[lamoli]

OK here it goes.. doesnt have to be on 3D engine if pytfall can use image layers its all good too...

http://www.partage-facile.com/4Q12NKHWFL/nextprevcell.jpg.html

here are some idea on how to dislplay every cell of slaves with basic informations


http://www.partage-facile.com/243AE52QBF/roomsetup.jpg.html

Here the screen where you could add items/upgrades to the room if used in layers you could click items to activate that item and do the action like training/using
start with basic cell with mat and drape on the floor ( could be upgraded to bed to gain better vitality recovery and morale ) bucket.. ( to do private stuff ) plate ( to place food on could be upgraded to table ) and what ever tools/items you may want in there i just need the idea to model it and make the usable image with screen related to action/training..


http://www.partage-facile.com/NV9H18J7MW/training.jpg.html

Here the training using item on the wall....
to start every item will train specific skills.. i displayed all i was thinking of at that moment but only skill related to the item will show ( cant expect a doggy on a girl stuck to wall lol )
Basic idea would be to find a way to break the slave will ( every girl may have strength and weakness traits ) like tough/strong girls can endure more training than others while iron will/stubborn girl would be harder to break.. using the correct action for the girl ( character type could improve how fast you break her will ) will doesnt have to reach 0.. just enough to unlock some action and train the girl.. of course some training action would require more will reduced or other skills trained first..

For those who played Jack o nine tails.. well thats about that kind of training ( gona need a psychology system ) need to train the girl but need to have a limit.. if your too hard on her she could break permanently or be out of commission for days

First need to know if its possible to use in game even on later stages then i will expand it...

Other that the girl i didnt use texture of course just wanted basic graphics to explain the idea :p

[Xela]

 

[lamoli]

sorry tryed to put images in here but failed then used hyperlinks

[DarkTl]

I love how that partage-facile thing demands to disable both noscript and adblock before showing the picture 

Generally, there is nothing impossible here. However, Jack-o-nine tails uses multiple variables to emulate psychology, and no one but Old Huntsman knows how exactly all this works (if he hasn't already forgot everything).

Aside from that, you propose upgrades for cells and general gui. That's hardly a concept, sorry. It does not describe the training process mathematically or logically.

[lamoli]

I love how that partage-facile thing demands to disable both noscript and adblock before showing the picture 

Generally, there is nothing impossible here. However, Jack-o-nine tails uses multiple variables to emulate psychology, and no one but Old Huntsman knows how exactly all this works (if he hasn't already forgot everything).

Aside from that, you propose upgrades for cells and general gui. That's hardly a concept, sorry. It does not describe the training process mathematically or logically.

i translated the Old Huntsman code so got all his scripts/class readable but working on it is a pain as that engine is unforgiving save wise..
as for the training concept we could start with his if needed then work it our way.. but maybe thats old news to you..

well i understand QSP its realy easy but to translate that into python thats another story.. might need to learn it more than im able atm

[DarkTl]

Python? You just need to describe the logic behind all those Huntsman's training scripts, there is no need to code it for us.

[Xela]

ST system:

**It should be stats based.
**It should be traits based.
**It should scale well to other elements of PyTFall

*It can be partially flags based (the psychology thing).

Otherwise anything that flash based player can do, Ren'Py will do better. Those pics are just images being displayed one over another, Ren'Py has full OpenGL support (2D libs) and might have 3D in the future.

Otherwise, as Dark said, that isn't really a concept, it's a GUI option. We should prolly also have simpler training that Jack because that game is dedicated to it, we a lot of other elements but I cannot see why that concept cannot be adapted at least partially. I've never seen their code btw.
=====================================================================================

I am still trying to get to the bottom of wtf is happening with girls not auto-equipping themselves. I tracked the issue down to some insanity with Auto-Rest code but still haven't pinpointed the problem. I'll try to get this and CWs but fixed tonight.