Please Release a new house every 2 weeks, I will give you my money

[[PVP] Boss Prime]

Here take it before one of these clones catches up and I'm gone like a ghost!

[Maya-Neko]

You need like 20+ devs to achieve something like this. If you're able to really pay that many people, then it's propably better to build your own dev studio at that point.

[AllergicToBrainRot]

No. Small dev team isn't going to pull that off.

[Riley⁧⁧⁧2142]

It would be totally feasible for them to come up with a procedural-generated system for this, HOWEVER, considering they have 0 PR skill, and keep the community in the dark for most things, I wouldn't ever count on new stuff like this which would fundamentally change game play and make the game better.

[Carnasid]

Originally posted by Riley⁧⁧2142:

It would be totally feasible for them to come up with a procedural-generated system for this, HOWEVER, considering they have 0 PR skill, and keep the community in the dark for most things, I wouldn't ever count on new stuff like this which would fundamentally change game play and make the game better.

!. They cover pretty much everything they are working on regularly on the official Discord.
2. The team is literally three people.
3. You clearly have no idea how difficult a "procedural-generated" anything is to make.
4. Fundamentally changing gameplay isn't how you keep people happy, it's how you get review bombed, and black listed by consumers.

[Kaisoku]

Originally posted by Riley⁧⁧2142:

Now I'll explain how easy it is to make something procedurally generated:

You simply need to create a data set of repeating assets/models, built around a certain aspect such as buildings etc. and then you need to ensure that when it pulls from said data set, it creates a similarity every time, but it also does it in a unique way. You cannot tell me it is hard to do, considering just about anyone can do it efficiently on a large or small scale with the right hardware and know-how.

That doesn't take into account how the ghosts need to know how to interact with the map. How they decide their pathing (such as how far into rooms they'll go), or things like how the "ghost room" is selected; from their varied descriptions of past issues, it sounds like the map actually has the rooms in a list form, and decides how to move to "the next room" via that list, which is how some bugs occurred where it jumped across the entire building.
How the ghost paths and creates waypoints around the building, what it can and can't see via line of sight, and then taking into account the 20 or so different ghost behaviors and how they'll interact with the building (some ghosts have roaming differences, others can teleport and then path back to their room, etc).
Something as simple as "which way a door swings" can make or break a map. They had to do a total revamp of maps a little while ago when they took away "hiding behind doors" because they realized there were entire areas that had no hiding spots.

I could maybe see it working as a sort of "chunk" format of procedural generation, maybe, as then you could ensure certain aspects like the door not swinging to create a hiding spot against furniture, or ghost pathing properly into rooms so that a certain number of hiding spots are guaranteed.
But that means you are probably looking at your "procedurally generated" house basically having maybe 3-4 chunk areas that are randomly pieced together.

In my past experience with things like that (Torchlight 2 or Diablo 2, I believe, although there are others as I recall, like Hellgate London I believe), you get used to those chunks pretty quick. It would be "kinda neat" for a while, but you'd start to see the limited number of patterns. There's only so much that can be done when you have so much dependent on the layout of the building and its furniture.

And as much as you make it sound like it would be a simple thing, it would be a lot of work to design this from this point, especially since they are still trying to hammer through the entire progression aspect of the game.
This might be something that could be worked on after the game is.. well, "finished"... as in, there's no new ghost behaviors to worry about, or equipment and progression aren't hanging over their head preventing people from viewing this as a complete game.

..
I get it. 2022 was a slog of a year, compared to previous updates. They warned it was going to be like this. They gave a roadmap at the beginning of the year, and they said it was going to be mostly engine rebuild/VR update and whatnot for the first half of the year.
We are just now getting into the Custom Difficulty part of the roadmap, afterwhich we should see the progression and equipment overhaul getting done. Completing the game, so to speak.

They've just hired a map creator so that they aren't stuck with nothing coming out for the foreseeable future. However, I would suspect that if you are wanting more "content" (read: maps to play on), then you may be looking at next year's roadmap for any serious gains there.

[Riley⁧⁧⁧2142]

Originally posted by Kaisoku:

Originally posted by Riley⁧⁧2142:

Now I'll explain how easy it is to make something procedurally generated:

You simply need to create a data set of repeating assets/models, built around a certain aspect such as buildings etc. and then you need to ensure that when it pulls from said data set, it creates a similarity every time, but it also does it in a unique way. You cannot tell me it is hard to do, considering just about anyone can do it efficiently on a large or small scale with the right hardware and know-how.

That doesn't take into account how the ghosts need to know how to interact with the map. How they decide their pathing (such as how far into rooms they'll go), or things like how the "ghost room" is selected; from their varied descriptions of past issues, it sounds like the map actually has the rooms in a list form, and decides how to move to "the next room" via that list, which is how some bugs occurred where it jumped across the entire building.
How the ghost paths and creates waypoints around the building, what it can and can't see via line of sight, and then taking into account the 20 or so different ghost behaviors and how they'll interact with the building (some ghosts have roaming differences, others can teleport and then path back to their room, etc).
Something as simple as "which way a door swings" can make or break a map. They had to do a total revamp of maps a little while ago when they took away "hiding behind doors" because they realized there were entire areas that had no hiding spots.

I could maybe see it working as a sort of "chunk" format of procedural generation, maybe, as then you could ensure certain aspects like the door not swinging to create a hiding spot against furniture, or ghost pathing properly into rooms so that a certain number of hiding spots are guaranteed.
But that means you are probably looking at your "procedurally generated" house basically having maybe 3-4 chunk areas that are randomly pieced together.

In my past experience with things like that (Torchlight 2 or Diablo 2, I believe, although there are others as I recall, like Hellgate London I believe), you get used to those chunks pretty quick. It would be "kinda neat" for a while, but you'd start to see the limited number of patterns. There's only so much that can be done when you have so much dependent on the layout of the building and its furniture.

And as much as you make it sound like it would be a simple thing, it would be a lot of work to design this from this point, especially since they are still trying to hammer through the entire progression aspect of the game.
This might be something that could be worked on after the game is.. well, "finished"... as in, there's no new ghost behaviors to worry about, or equipment and progression aren't hanging over their head preventing people from viewing this as a complete game.

..
I get it. 2022 was a slog of a year, compared to previous updates. They warned it was going to be like this. They gave a roadmap at the beginning of the year, and they said it was going to be mostly engine rebuild/VR update and whatnot for the first half of the year.
We are just now getting into the Custom Difficulty part of the roadmap, afterwhich we should see the progression and equipment overhaul getting done. Completing the game, so to speak.

They've just hired a map creator so that they aren't stuck with nothing coming out for the foreseeable future. However, I would suspect that if you are wanting more "content" (read: maps to play on), then you may be looking at next year's roadmap for any serious gains there.

I've been awake now for going on 23 odd hours so apologies if anything I said/say doesn't make sense! (Cursed be IT jobs!)

I definitely get what you mean in regards to doors, and how certain things would be generated, which is why they are normally a single-use interaction to prevent such "visual defects" and whatnot from manifesting.

Path-finding would have to be dynamic, and there would have to be specific scripted triggers tied directly to certain assets such as the doors or movable objects. A path-finding system hard coded to follow between these specific triggers is possible but not easy to perfect. It's been done for games before, but because of how complex these systems can be and the costs associated with it, it's simply not feasible if they do indeed only have 3 developers. I definitely have no clue how their back-end systems look, an it would probably be completely different compare to something I would throw together. I agree wholeheartedly that they probably don't have it in mind and quite possibly never considered it on a serious note for inclusion.

As for the aspect that a player would definitely get bored of generated maps, I 100% agree with that statement, as all things get boring after a while. I do however know that surprise can drive horror games to succeed longer term based on how complex they are and how the environment reacts to the player themselves.

Definitely not harping on the game at all, it's a great game, but they lack proper PR, which is something you have to be on top of with a small team, so that your priorities remain intact throughout EA, otherwise you end up with a game like DayZ that had been in EA for well over 7 years. I only bought Phas a few months ago after a friend of mine suggested it.

I do hope to see this new update by the end of the month, as I'd love to see the new and improved Asylum, and the custom settings will make for MUCH more enjoyable game play overall.

[Rhapsody]

Creating some sort of map generation tools for the game would be a massive undertaking, depriving much needed development time from pumping out actually needed feature and content updates. User-generated horror content would be completely subpar to handmade levels and effects.

As for the OP, that's clearly sarcasm to me. But I'm not sure if we need more of these kind of topics, sarcasm or no. 8)

[Maya-Neko]

Creating a procedural algorithm in itself isn't that hard. Making it look natural and especially making it look like real buildings on the other hand is definitely on another scale though, unless it deliberately made to look ridiculous like for example in Hello Neighbor.

[Riley⁧⁧⁧2142]

Originally posted by Maya-Neko:

Creating a procedural algorithm in itself isn't that hard. Making it look natural and especially making it look like real buildings on the other hand is definitely on another scale though, unless it deliberately made to look ridiculous like for example in Hello Neighbor.

It would be cool if it was done in that fashion, as it would create some really wacky and fun scenarios.

[Kaisoku]

Honestly, a "Winchester Mansion" style of procedurally built building, that is intentionally built to be confusing and maze-like, even in how the hallways and stairs go, could make an amazing map.
A weird hedge-maze or abandoned mine type map (thinking of Labyrinthine) would work as well.

However, it would have to have be built with lots of corners to lose line of sight, and "rooms" would have to frequently have some kind of hiding spot to get to, to make it survivable.

As a couple "wacky, use at your own risk" style maps, this could work.

However, yeah, could take quite a lot of new code style of work, which should all happen after the rest of the game is done, honestly.
Don't want to take away from proper "finishing the game" dev time, and wouldn't want to have to re-code it after some final mechanic ends up breaking things.

Still.. I'd love a couple maps like that!
It is correct that the surprise factor would keep the game a bit more fresh in play.

[GamingWithSilvertail]

Originally posted by Carnasid:

Originally posted by Riley⁧⁧2142:

It would be totally feasible for them to come up with a procedural-generated system for this, HOWEVER, considering they have 0 PR skill, and keep the community in the dark for most things, I wouldn't ever count on new stuff like this which would fundamentally change game play and make the game better.

!. They cover pretty much everything they are working on regularly on the official Discord.
2. The team is literally three people.
3. You clearly have no idea how difficult a "procedural-generated" anything is to make.
4. Fundamentally changing gameplay isn't how you keep people happy, it's how you get review bombed, and black listed by consumers.

Both Unity and Unreal Engine 4 have the foundation of procedural generated stuff.

So I can slightly understand where he would be coming from. Its not as 'complicated' as you might think it would be

Also, It doesnt really have to touch the existing stuff.

Originally posted by Maya-Neko:

Creating a procedural algorithm in itself isn't that hard. Making it look natural and especially making it look like real buildings on the other hand is definitely on another scale though, unless it deliberately made to look ridiculous like for example in Hello Neighbor.

This is why developers whom implement this feature, generally have catagories for what works together and what wont.

[Kaisoku]

Having a house built procedurally isn't necessarily the problem. It is extra work that takes away from other stuff, and would be better to finish the game first, but still it's not "that" hard.

Having that nice looking house play nice with the ghost mechanics is the problem.

Going by the amount of work they've had to do over and over to make the houses work properly with the ghosts, I'd be surprised if it was as "easy" as folks are implying to just plop a procedurally built house down and expect it all to work right.

[Tyler Zeta]

Originally posted by Fear:

No. Small dev team isn't going to pull that off.

Even big dev teams would struggle with that since they would need to work overtime.

[Tyler Zeta]

Originally posted by Riley⁧⁧2142:

It would be totally feasible for them to come up with a procedural-generated system for this, HOWEVER, considering they have 0 PR skill, and keep the community in the dark for most things, I wouldn't ever count on new stuff like this which would fundamentally change game play and make the game better.

They have a discord server where they talk about the progress of the game and a trello with the games road map.

[Tyler Zeta]

Originally posted by Carnasid:

Originally posted by Riley⁧⁧2142:

It would be totally feasible for them to come up with a procedural-generated system for this, HOWEVER, considering they have 0 PR skill, and keep the community in the dark for most things, I wouldn't ever count on new stuff like this which would fundamentally change game play and make the game better.

!. They cover pretty much everything they are working on regularly on the official Discord.
2. The team is literally three people.
3. You clearly have no idea how difficult a "procedural-generated" anything is to make.
4. Fundamentally changing gameplay isn't how you keep people happy, it's how you get review bombed, and black listed by consumers.

So many people kept coming here with feedback/gameplay suggestions while totally ignoring that the devs have a roadmap and a discord.