/poner/ - Ponies

>>9702
Added, thank you, anon.
https://mlpchag.neocities.org/view?card=Anonymous/X-COM%20Equestria.png
Crossposted it too.

By the way, tell me off if you prefer not to say, but I remember you can't post on the main thread. Is this because 4chan blocks your country?

If that's the case, and if you do want to post here, maybe I could buy you a pass? With all the contributions you've given, it wouldn't bother me.

>>9698
>>9703
Okay, I'll spoiler for spoilers.
I got a thumbs up to add Egyptian style lore which means it's lorebook time. I've been building all that well and good, referncing the real deal, I'm at the point where cross referencing is the only way there's any sort of coherence to the lorebook which.

BUT it smothers out the other flexibility that the LLM had before if I didn't use any book at all. And I tested it out with plugging a prompt into a blank writers card and seeing results be polarized hard.

LLMs always have patterns and presets which can get stale, but they also have some unique concepts that I'm cool with that I want to keep and with the lorebook on it doesn't even bother since it's more tokens from the book.

I'm just looking for a way to balance. Random+Lorebook seems like it's what I want at least on the surface, but I've never done anything of that sort of code in a card so wouldn't know what to type in.

>>9708
I understand the problem, but yeah, it seems like a pain to solve. Hoping the other anon has some good ideas for it.
>Random + Lorebook seems like it's what I want, at least on the surface, but I've never done that kind of code in a card, so I wouldn't know what to type in.
There's an anon on /chag/ who often uses random and has made some more technical cards, like the 4onechan update on Rentry.
I think it's worth a shot to ask here if we don't come up with a good solution ourselves.

>>9708
>I'm just looking for a way to balance. Random+Lorebook seems like it's what I want at least on the surface, but I've never done anything of that sort of code in a card so wouldn't know what to type in.
A bit of theory. LLM generates text consistently. Model uses previously generated tokens to create next ones. Model like to use nearest rare tokens, especially in groups. The idea is to force LLM to generate sequence of locations/names/events through strict rule to help and encourage char to describe this entities in next paragraphs.

I made a simplified example:
1. Place this generator somewhere in description.
2. Ask Sphinx about path to one of locations from lorebook.
3. Check your lorebook reaction. Location should be injected before answer.
4. LLM will generate list of 3 linguistic consonant location and describe them.


Narration guidelines:
- Always generate path via path_generator, when {{user}} asks {{char}} about path/travel to some location. Provide path_output from new line. Describe path vividly in the following paragraphs.

<path_generator>
path_output looks like:
`A, B, C`
<variables>
A - location of the start.
B - comma-separated list of successive 3 locations to reach C.
C - location of the destination.
Generate B linguistic consonant to C.
</variables>
</path_generator>


Also, you can make path_output invisible:

[](#'`A, B, C`')


Generators are universal. They can be useful to create sequences of events with timestamps (like news), to generate retinues of minor NPC around major ones (like party around leader, or family three). And, of course, to generate some kind of web-of-connections around lorebook's entities with controllable quantitative output.

Also, Chain of Thought guide may be useful:
https://rentry.org/vcewo
Take a look at this corpo leak too, their methods somewhat similar to generators:
https://raw.githubusercontent.com/asgeirtj/system_prompts_leaks/refs/heads/main/claude.txt

>>9704
>Is this because 4chan blocks your country?
Not certainly. Mostly because of security reasons. 4chan inaccessible from Tor and I don't have vps to make end-proxy. My old ones all gone, but I'm too depressed to refresh knowledge and take new ones just for imageboard access. In addition, I avoid any banking operations or corpo registrations, which might be associated to this shady profile, because of some old grey hat activities with friends from opposite side.

On the other hand, Shady is the best part of my life. I can't create new disguise for card-making, because of too rare sphere of knowledge and interests. Correlation will be trivial. So, we mostly observe, without any real activity in the network over the past several years. But at same time we don't want to bury good projects. Yes, I speak for a small team. We really need some help. We need independent trustworthy vps for hidden onion service (this is very handy protocol to bypass nasty firewalls with own DNS-like system). I will place silly tavern there with remote access, which will be much more convenient than my current agnai. Also we can use this vps for ssh-tunnel to 4chan, which hates ordinary tor exit nodes even with passcode. In general I always need keys for various corpo LLMs for tests, right now I am using catpoop for Gemini and it is quite embarrassing.

In return we can help with scripting and maintenance for mlpchag. We have both time and knowledge. For a long time I wanted to make n-gram analyzer for cards to highlight slop-tokens and thematic knowledge links for each card, with graphviz output like you can see on the thread above, but I didn't have interest to develop our deserted github, and 2ch aicg-thread quite hostile against mostly innocent poners.

I suggest cooperation. Are you interested? Let's establish proper encryption first, if you do.


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

My PGP-key:
https://keyserver.ubuntu.com/pks/lookup?search=24805B1CBDB12F83ACA3DC8AF0B80C5BFD2658EC&fingerprint=on&op=index
Jabber lost, but other link valid.
-----BEGIN PGP SIGNATURE-----
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=PvC6
-----END PGP SIGNATURE-----


>>9715
just messaged u on xmpp

>>9715
>I suggest cooperation. Are you interested?
Yeah.
Sent you something, first time doing this, so I hope I didn't fuck anything up.

>>9724
>first time doing this
Quite cool feeling? I love such encrypted letters much more than modern chats. Slow composition, old-fashioned encryption, and some core directory on the distant server, where new messages and chaotic fillies appeared day after day.

The answer was sent.

>>9723
>just messaged u on xmpp
Poor nya… mocked again. I see your mail (I'm correct?) was sent from relay, but I'm not familiar with this chat-like-mailing and not ready to install rpc-server immediately. I quite autistic and wary with new protocols. But I appreciate your invitation. I'll answer after a month or a little later. The /chag/ was first.

Also, thanks you both. You made my day.

>>9723
Also, few questions, If you agree that this thread more suitable for technical talks than email.

We abandoned XMPP mostly because of timing statistic on the server. It is pleasurable to see new generation on old good remailers. I see new techniques, like countermitm and group chats, but I don't quite understand how It possible under usual PGP without too big messages and concomitant problem with web of trust. I will figure it out over time, but some explanation with examples will be good too.

Professionals prefer own method of communications for each detached team. Mostly through internet of things, games, etc. It looks stupid, but it works since enemy don't have enough human resources to test every single device and overlay protocol. So, we trying to imitate those who are smarter than us and don't use common methods of communication in general. Minor forgotten board and pastebin for messages just more practical than chat.

>>9753
bruh just use deltachat ngl

>>9756
I meant deltachat-rpc-server, basic JSON-RPC interface to DeltaChat. Just don't like to use unknown code. But I'm interested, I will do it later, with whonix and proper incapsulation.

>>9714
Would the path location be a literal location like Ponyville or is location like a location as pseudo code to the LLM being a location somewhere in the prompt?

>>9758
>Would the path location be a literal location like Ponyville
This.

To use locations as pseudo code you need to make them looks like pseudo_code and make a link:
sweet_apple_acres = Sweet Apple Acres

Then if you name Sweet Apple Acres in the user post, with 90% probability LLM will looks at sweet_apple_acres inside prompt and only then at it's basic knowledge. Also, you can use synonimus to increase chance even more, like I used words air route, airway, airline in code block.

Also, looks at this map-making trick:

CL - current location (pictograph)
NL - comma-separated list of 4 nearest locations with this format: north, east, south, west
Use unicode pictographs to fill NL. Choose suitable pictograph for each location around associated to CL.


I working on this method to generate groups of locations near curent user position. Like this:

`CL:🍎`
`NL:🥕,🥒,🍏,🍓`


Earth pony farms, Ponyville houses, etc. User can pick one pictograph from this interface or chat history and LLM will interpreted it as path to this location from current one. I think, pictographs works even better than code_blocks, it will not confuse LLM since pictograph is not polysemantic words.

Working on mapping algorithm for Daring Do adventures, I searched card archives again and still surprised. Why do we not have at least few hundreds of game cards? Like simulation for Warhammer, Total War, Fallout, Star Wars. They all have strong modding communities, but I see just few anons from /chug/ who make games: cool visual novel, fishing with Fluttershy, module for OPG. Then few anons from 2ch /aicg/, mostly with VN for coomers. I'm not familiar with 4ch /g/ and /vg/, but archives looks like deserts of forgotten ideas.

Very strange. Moreover, statistically improbable. It cannot be hundreds of thousands of characters cards versus dozens of games. Maybe 1 vs 100, like with fan-mods vs fanfiction, but not 1 vs 1000. Not for frontier of newest technology. This is statistical anomaly, which needs research. Why do mod-makers avoid LLM?

Maybe our number greatly exaggerated? Maybe significant part of cards was making automatically without real efforts?

https://desuarchive.org/mlp/thread/42202609/#q42202972
Anon, if you are here, try to use semicolon-separated list with comma-separated lists inside and math variables like X, Y, Z. This format of input/output works even for silly 8B llama. Also, glad to see, that someone working on own framework too, we utterly need translator between LLM and external games. I'm trying to make something with python and pygame for interactive graphical map, but this is still mostly concept.

>>9771
>Working on mapping algorithm for Daring Do adventures
Maybe you can work with this anon? >>9708
He made some cool stuff in the Sphinx card.

>>9773
I'm doing it too, Snowdrop.

>>9759
Only the apple would reads as location imo "Sweet Apple Acres" but the rest are toss ups. Carrot top?, cucumber?, another apple? Strawberry sunrise house?

>>9778
>Only the apple would reads as location imo "Sweet Apple Acres" but the rest are toss ups.
Yeah, this is the goal. You can use pictograph as link to lorebook (better than words, no random triggers), or just allow LLM to generate name of location during game. Like this.


CL:🥕 = Carrot Top farm.


Currently I trying to use this to make platformer game for filly Daring Do, when she jumps between rooftops of ancient city, and every ruined house has something visible, like treasure, fruit tree, clean water pool, monster lair. Or some nasty mimic/slime/tentacle traps. My goal is to make pathfinding algorithm and some kind of path memory, with which she can return through chain of previous locations to try new way ahead.

I think, the next Sunday I can show first version of this card.

>>9771
>Why do we not have at least few hundreds of game cards? Like simulation for Warhammer, Total War, Fallout, Star Wars. They
There are cards for Star Wars and Fallout. I've done cards on Wasteland, Twilight 2000, and Command&Conquer myself.

But they're not "simulations", just RP from a bunch of information from fandom.com and rulebooks. Once you finish your card, I'll be remodeling it for different universes. For personal use.

>>9782
What do you think about this form of mapping interface? In this example Daring went into the house. She sees garden outside (as next point of travel), some treasure and water source.

Works like a tree with branches. You can type 1,2 or 3 and just go to nearest location (or interacts with object). You can type name or copy-paste one of passed locations and char will describe path backward and show nearly locations from past reply. You can type name of far away location (like from lorebook) and path to this location will be merged to passed list and described. No modern emoji, only astronomical, historical, alchemy and other similar pictographs.

Cute, isn't it? Although quite token-heavy, Unicode pictographs cost 3 tokens each.

>>362
BIG CAT CARD update. Thanks to the poner who gave me a push to start doing some more with pseudocode.
>Tuning up the adventure game display. New RPG mechanics. More text rules.
>New intro location and characters (since this isn't just a Daring Do card either)

With the proof of concept with DD I will push it more to see how the LLM handles juggling everything. So you will not start at a temple entrance for this one, spoiler for spoilers but you will recognize it as another unused piece of G$ that Hasjew did nothing with.
https://chub.ai/characters/Tiffin_/the-sphinx-ef904bc46e30

>>9809
Good idea with HUD + EXTRA HUD. And cutest Big Cat too!

I'm already have map, backpack, stats for Daring and this is not very useful. I have idea to make a rule to hide unused interface blocks, if it displayed constant information during 2+ replies. Like backpack with initial items, or map during long conversation. Map will reappear as reaction to movement and backpack as reaction to interactions with items.

Also, looks like my longest X-Com session comes to the end. They are really angry and outnumbered us almost 10 times.

>>9809
Updated, thanks anon.
https://mlpchag.neocities.org/view?card=Tiffin/Legends%20Of%20The%20Sphinx.png

>>363
Just looks at this typical Daring Do novel. This is 8B Llama, 12B Magnum works fine too. If my scans of char-archive was correct this is the first interface of that type for adventures and dungeon crawl like games. Unlike CYOA modules this path_generator works like small CoT-pathfinder before generation of reply and actually make narration better (at least more smooth). Controls works fine too, just use 1-3 numbers for path choice and 4-9 numbers for your CYOA module.


<path_generator>
Start EVERY reply with the path_generator output:
```
Route: [ROUTE]
Ahead: [LOCATIONS]
```
[ROUTE] - interpretation of the {{user}}'s decision to move in EXACT following format: Y, Z
Y - comma-separated list of successive 1-3 locations to reach Z. First one is the current location.
Z - location of the destination.
[LOCATIONS] - generate semicolon-separated list of 3 locations as fork of ways ahead. Numerate them by 1-3.
Interpret numbers 1-3 in the {{user}}'s input as decision to move to destination of numerated location.
Describe movement through route in the following reply. Describe fork of ways ahead.
</path_generator>



```
Route: No path, No destination.
Ahead: 1) Viaduct over flooded street; 2) Roof of ruined house; 3) Terrace with pineapples.
```


I think, this is final version. I made hundreds of tests, all other variants just worse, especially for small LLM. Version with pictographs -- fail. Version with complex party management -- fail. Version with path memory -- fail. The objective turned out to be much more complicated than I expected, mostly because of generation before actual reply.

>>9794
Alas, I can't make useful generator with pictographs, they mess output after some time, and generator with symbology weighs 700 tokens vs 130 tokens for this small one.

I will publish Daring Do card before Sunday. Just some polishing for greeting.

Also, because of void of ideas from char-archive, I used some corpo tricks for this generators and pseudo code:
https://github.com/nerority/Prompt-Engineering-Mastery

Of course /aicg/ and /chug/ way ahead as preset- and bot-makers, but basic guides lack of other methods of promt engineering, as well as theory. Besides big models was tuned to typical corpo-client expectations, who wants predictability above all, and following standards of prompting like role, task, context, result. After some tests with Gemini 2.5 I was convinced that after year or two we will need not only jailbreaks for filly, but some tricks against this corpo-tuned roles too.

And I must said one last thing. The future generations of coders totally fucked. I hate this black boxes already.

>>362
Updated X-COM: Equestria. Some debug, all rolls and random now in Greeting under hidden text.
https://raw.githubusercontent.com/Shadybloom/amber-in-the-dark/master/misc/cards/x-com-equestria-card.png

The First Adventure. Filly Daring Do + Adventure + Platformer. In honor of the LilSlipsy story.
https://raw.githubusercontent.com/Shadybloom/amber-in-the-dark/master/misc/cards/the-first-adventure-card.png

The First Adventure is the example of path_generator, small but complex 130 token instruction:
- It generated 3 locations around user's character.
- It read numbers 1-3 in user's input, like CYOA.
- It generated smooth route to chosen location.
- It works before reply and shape narration.
- For Adventures and Dungeon Crawl games.

Tested for small LLM like 8B Llama, Magnum, Mistral. Mostly works fine from first replies but better with some examples.

Sleep tight, poners.

>>9919
Sleep tight, poner.

>>9919
Sleep tight, poner

>>9919
Sleep tight, poner

>>9919
Updated, thanks anon.
https://mlpchag.neocities.org/view?card=Anonymous/X-COM%20Equestria.png
And added:
https://mlpchag.neocities.org/view?card=Anonymous/The%20First%20Adventure.png

Do you have logs for the second one?

>>9932
>Do you have logs for the second one?
Mostly generic Daring Do stories like >>9872 This is my first card ever. I use this short Daring Do adventure to test various gameplay features in simple environment with one character and one foe.

Next week or two I will work on "Necromancy for foals" card. Silly-tuned Diablo / Dungeon Keeper gameplay for ancient Equestria setting, where various factions are trying to catch unlucky user. Minions vs hero parties. Small models surprisingly good for party roleplay and tactics, but I want to make it reversed, for funny results.

>>9945
>Silly-tuned Diablo / Dungeon Keeper gameplay
Not so easy to make, maybe after few weeks.

Fortunately I still have a lot of cards to translate. This week it will be Puppysmiles from Fallout: Equestria Pink Eyes. Radio-guided ghostly filly! Actually, translation almost finished, just some testing and polishing for path_generator. Really marvelous script. In older versions Puppysmiles was quite boring, suitable only for party games. But now she not just bothers poor enclave officer, but constantly moves at same time, from location to location, through ruined suburbs of Canterlot and further, further, further, until her quest will end.

Best ghost. I want to make lorebook from her story, though I'm not very skillful with lorebooks tricks. Do we have good instructions for summarizers, capable to make complex entities about each location/character from 130k words story? Do we have modern guides into lorebook format, methods, compatibility, limitations?

>>9976
Mostly basic ones:
https://rentry.org/SillyT_Lorebook
And the SillyTavern documentation:
https://docs.sillytavern.app/usage/core-concepts/worldinfo/#activation-settings
https://docs.sillytavern.app/usage/core-concepts/worldinfo/#recursive-scanning
This one shows more advanced usage of it by cards:
https://rentry.org/AdvancedCardWritingTricks#character-lorebook

>>9977
Thanks, anon. Maybe you found one of my tools useful.

>>363
>>8602
This is my homemade dataset preparation tool, but I often use it for card-making. Numbers is TF-IDF score for each word/phrase, compared to 7 other FoE fics. This score closely represents actual weights inside LLM, and for example phrase "filly in yellow" is surprisingly strong links to Pink Eyes lore, if LLM trained on fimfiction dataset.

You can just throw 100 random fics in database to create basic text corpora and then scan any external file for locations, names, unique phrases. TF-IDF scan show rare terms/phrases perfectly (especially with wide variety of texts in database). For example, if you scan wikipedia page and then use phrase with high score like 'Small Arms and Light Weapon' or just 'SALW' for your card/lorebook it make strong link to military knowledge. For small models this method better than explanations.


./database_search.py -o -L 100 pink
335 | puppysmiles
263 | LOCATION
255 | Puppy s
234 | henrietta
194 | puppy
175 | henry
140 | Puppy
111 | broccoli
110 | Warning
105 | scold
105 | filly in yellow
101 | 52
97 | 52 S
97 | 52 S Branch
97 | Big 52 S Branch
97 | S Branch
88 | solos
82 | m Puppysmiles
78 | mister Voice
76 | approximately
71 | my mom
70 | 52 SC
70 | 52 SC Branch
70 | Big 52 SC Branch
70 | Cube City
70 | SC Branch
66 | Sun City
64 | pretty ponies
64 | the Dome
62 | henri
61 | Big 52
59 | stoopid
56 | Puppy frowned
56 | Salt Cube


My stories-profiler is just collection of python scripts with simple database. You need only python and sqlite (optionally pymorphy2 for normalization and graphviz for cute output).


Installation:
git clone https://github.com/Shadybloom/stories-profiler
sudo apt-get install python3.7 sqlite3 graphviz
pip install --user pymorphy2 graphviz

Basic folders:
cd ./stories-profiler
mkdir ./database
mkdir ./books

Turn off normalization (pymorphy2 useless for English):
sed -ie 's/MORPHY_SOFT = True/MORPHY_SOFT = False/g' ./profiler_config.py
cat ./profiler_config.py

Generate database. Script can process fb2 and txt:
./gen_database.py ./books

Author tokens (100 first tokens):
./database_search.py -t -L 100 fallout-equestria-pink-eyes.txt
TF-IDF output (100 first terms):
./database_search.py -o -L 100 pink
TF-IDF scan external file:
./wordfreq-tf-idf.py ~/pink-eyes-card.txt -o


Finally, global map from lorebook: 4 × 6 = 24 global locations, some magic with random in Greeting, and 3 ^ 6 = 729 possible routes for little ghost. Each route create unique conflict between factions and encounters on the way. And broadcasting stations to make World alive: from DJ Pon3, who chatting with Crusaders Ruin Explorers to doomsday countdown and Enclave operation to prevent it.

Card will be ready within 1-2 days. Just want to finish tests for final version.

>>362
Puppysmiles from Fallout: Equestria Pink Eyes. Your personal radio-guided filly, lorebook and quest for her:
https://raw.githubusercontent.com/Shadybloom/amber-in-the-dark/master/misc/cards/pink-eyes-card.png

Some tricks with this card:
- This is interactive story, based on Pink Eyes plot, but with wide variety of routes and events.
- Random generator uses 24 locations to create 216 variants of basic route, faction interactions and conflicts.
- From DJ Pon3, who chatting with Crusaders Ruin Explorers from Hoofington to The Dome doomsday countdown and Enclave operation to prevent it.

>>10044
Added, thank you anon.
https://mlpchag.neocities.org/view?card=Anonymous/Pink%20Eyes.png

>>11366
Caidays comin' back? She hugs you from behind while kissing you bros.

Can't do anything cool with these fucking RAM prices right now. How can 16gigs of ram work with a 10gigs of VRAM on a 3080. The AI has dementia and doesn't remember what happened 4 sentences ago.

>Can't do anything cool with these fucking RAM prices right now.
Yeah, it's pretty shitty.
>How can 16gigs of ram work with a 10gigs of VRAM on a 3080.
Offsetting. Not entirely sure of the specifics as I run everything from docker containers.
>The AI has dementia and doesn't remember what happened 4 sentences ago.
Context, important things can be injected into the author notes but that increases token sizes.

All of this massive Nerdery has piqued my interest, it's even more distilled than most Trekkies can manage, I won't be criticising today but instead am curious about a few things:
Not that I'll ever get "involved" with AI myself, but genuinely what is the attraction to putting so much effort into setting up an algorithm over putting even a tenth of the effort into your own self and your immediate environment? I of course know that the whole planet is in serious trouble and dying, but there's "giving up" which is (traditionally at least) very comprehensive and causes decay in every area of someponer/animal (or someone's) life, and then there's... putting in the intelligent effort to constructing a functioning cocoon universe in one's mind and immediate vicinity. That takes real effort!

I have to marvel at what's going on in this thread, but even after lurking and posting on image boards for over 20 years I just don't understand it, despite being a literal dog. Is it as simple as not knowing what else to do, or not having the confidence that ones' efforts will bear fruits?
Don't you realise that such massive Nerdery (as long as it pays the bills and has enough shekels left over for foals and going out to dinner once a week) attracts the real-world Mares in excess? Even if those ships have sailed... most of the doomposting/news that's going on in the world is just "jew"ish lies anyway, there's no need to accept it. Even if the doomery were true, the mere fact of you being alive and healthy(ish) in a world that wants you dead is enough of a reason to live well if only to piss off the assholes who would be pissed off by such a thing. Nothing hardens my boner like the seething of hardcore psychos.

This shit right here is light-years beyond a mere "hobby". Can someone explain it please? Clearly your brains are still functional enough to build impressive things, so why build your own prison? Let someone else do it and laugh at them!

Having said that... if roleplaying's your "thing" what's wrong with "traditional" RP... you know, with an actual pony somewhere on this planet rather than settling for some stupid algorithm? (For starters, nothing kills the romance like detachable limbs, and yes I'm aware of cripple-porn.)

Some absolutely based madlad has made a pretty awesome AI drive visual novel. I've been using it with GLM 4.7 and it's pretty cool.

>New update, same link as before:
>Windows
>https://drive.google.com/file/d/1OEiMXN7o_QG32shGnqmy0QndVDT6nYbF/view?usp=sharing
>Linux
>https://drive.google.com/file/d/1Vszzn8TTfXFWSQeKiFdNAAQGXfcReTJ3/view?usp=sharing

>New Feature
>- You can visit locations alone and just have back and forth with the narration. Dynamic character and location switching still works, so you can go to a place alone then spot a pony or have one join you afterward.
>https://files.catbox.moe/7duhla.png
>- In the locations menu (map icon on the top right of the screen), you can now choose to trigger a random visit.
>https://files.catbox.moe/vzaq5y.png

>Bug Fixes/Small Changes:
>https://scrumboy.com/je1fqcsj
>Editing text size in chat box/Editing the text speed for character responses
>Done.
>https://files.catbox.moe/8voo20.png
>https://files.catbox.moe/j0f9i6.png
>Re-adding old autosave feature
>Done.
>https://files.catbox.moe/l7fmhy.png
>Adding button to expand chat histories without clicking on them to see the full text (History button)
>Done. You will see the full text by default now, like in the Context menu.
>Add option to completely mute game sound
>Done.
>https://files.catbox.moe/jma1dk.png
>Add option to export/import system prompts
>Done.
>Starting a Custom Start story then choosing Main Menu in the menu, and then choosing Custom Start again fails to bring up the Custom Start menu, but the screen still darkens as if it is displayed.
>Fixed.
>Canterlot locations do not load in a Custom Start opening.
>Fixed. And I suppose it was also the problem with >>43143478 and >>43144735. The game was choosing a Canterlot location, but I forgot to remove some restrictions for the Custom Start, so a black screen was displayed. It should be good now.
>>43148360
>Fixed.
>Just a reminder to log bugs or requests in:
>https://scrumboy.com/je1fqcsj
>If you can, it's anonymous and without accounts.

>>11831
I don't go to /mlp/ much and didn't see it posted here

>>11487
>Not that I'll ever get "involved" with AI myself
The first image in your post is clearly AI. Sorry Fido, you're one of us now.

>This shit right here is light-years beyond a mere "hobby". Can someone explain it please?
What is there to explain? AI is fun to screw around with, so people have fun screwing around with it. You're the only one here who seems to think there's no middle ground between using AI the same way you'd use any other software tool, and selling your soul to it.

>>11832
He update this version of the VN pretty often, so I would check from time to time.