Why hasn't SS13 been translated to another engine, literally anything is better than BYOND

There was this one user who was working on a SS13 3d clone however he must've given up by now.

One guy is working on a Doomwad version.

You could do it in Java. Just like minecraft. Or just use something like C, like any proper programmer would do.

One of the things about SS13 is that a lot of the fast-paced or combat-oriented business is very dependent on the terrible interface, shit controls, latency and tile-based nature of the game. It brings a degree of unpredictability and tension that doesn't just feel like a roll of the dice, all while still falling in the precision of, uh, tiles.

If it were all rock-solid, you'd have a hard time getting an action in edgeways against superior equipment or superior reactions. Being robust is more of an art than a science.

I think it's because it's just kids who play it.

Yeah just do ss13 in your own engine with c and opengl like any real programmer would do, cant believe no one's done it yet

Children don't have the attention span to wait out a round they died in the first hour of just so they can git gud robust next round.

I thought that it was a retarded damage system where if you get like 60 dmg total you go into crit and then just get resp damage, or you either just go into crit because lack of oxygen or bleeding out

Just make lasers/projectiles slow and rubberband/semi-invsible

That's what happens when you put in full working organs and they're tied to HP, you'd die of lack of oxygen and bleeding out irl too.

Because porting a game to a different engine takes a lot of time, effort, and if you having people helping, money.

OpenMW, for example, has been going on for five years now.

Things should get more exciting once atmos and power are functional, both of which are receiving more attention now since gamemodes work okay at this stage, even if they're fairly unfinished compared to the originals until more equipment gets added. The UI indicator works, the station just isn't pressurized yet, so next up is hopefully setting off some plasma fires to welcome the new year.

I can't wait for Z& 3.0 to get a stable release, generally the updates are okay but the last patch led to intriguing netcode fuckery which in turn led to code janitor duty

Can the doom engine even handle shit like atmos, power, crafting, and all the UI elements for like computers and shit?

To an extent. I have a few possible approaches to crafting, I'm just still focused on the former two. It ties in with wrapping up the rest of the mobcode that handles burns and temperature/pressure.

Are you going to handle atmos in 2D? I assume that you are since the videos only show the maps on a single plane.

Doom maps are 2D by definition, most 3D stuff is done with clever tricks. Some engines have a few extra features to make stuff like multiple story structures easier to implement, but it's still hacky in practice.

It makes maps easier to port, though, which is nice. I'm pretty sure an utility could convert between the two formats, but it's a low priority.

Aight, are you going to implement atmos in the same way that it works in BYOND; as in that it's tile based instead of 'room' based?

I'm a huge atmos autist so this really piques my interest I mean a big enough autist that I remade the tile-based atmos in other engines, so this /really/ interests me

Simulating every single tile isn't going to be trivial since practical sector limits kick in before that. The renderer doesn't like staring across too many edges at once, so while it's possible to do a tile grid, it rapidly starts chugging and would probably require changes to the engine itself to have any chance of being efficient.

Sectors and tags are perfect for letting one set up something similar to rooms, though, and the plan is to use a ZAS-like system to at least keep track of flows and connections that way. Physical effects like velocity changes can then be done using actors or sector manipulation, or likely both, since source ports like to spread their functionality more than might be healthy.

It should also be mentioned that while I'm doing my best to dig through the original atmos code for a few different codebases to see how things work, I'm still not an innate atmos autist and any insight would be highly appreciated.

Anyone want to take a guess on a total budget, maybe on a four to six year time line. Shit at the very least we might be able to scam a little cash out of plebbit.

I can definitely see why you would be following the sector-based solution, especially when it comes to optimization.
I'm just not a fan of ZAS because it seems less 'accurate.' I want to see big walls of purples gas spreading around a room.
The optimization is a big problem though, which is why I've been looking at engines like UE4 and Godot, which allow for explicit multithreading so atmos calculations won't slow the rest of the game down.
My solution is this: There is a single 3d array which holds all the atmospherics info for every tile. The array shares an x and y coordinate with the map through some transform(because it's never going to be 1:1), the third dimension on the array is the moles of each gas, whether the tile is space/ a wall/ or a regular tile, as well as the total thermal energy (in joules). Then there is a single multithreaded process that loops through the array and balances the gasses and occasionally checks for tile updates to see if anything has changed.
Any object that has to do an atmospherics check just pulls from this array based on the same x/y transform.(Pressure and temperature can both be calculated from moles of gas and joules of energy, volume is assumed a constant since each tile is the same.)
The only other thing about atmos is that pipe networks are treated as a single container as long as they aren't divided by a pump or some other machine. Adding a pipe to an already existing network simply increases the total volume.

Theoretically, one can go one step further than "tile based".

By utilizing invisible mobs that dont collide with anything, you can produce a far more granule dynamic atmos system. Each mob acting as an abstract of a group of particles. You can assign different properties to them that are then projected over a raycasted aoe projection.

While inside of this aoe, different things can be controlled via scripts, or possibly decorate if you're a real madman. For instance, mobs that breathe who aren't on internals might slowly chip away the health of these particles, the lower the health of the particle, the lower the pressure of it's aoe. these gas mobs would be controlled to follow a waypoint that updates every other second or so and the mobs would be directed to the waypoint, who's position would be calculated so that unhealthy mobs are attracted to healthy mobs aoe. mobs would equalize their health in this aoe.

the primary issue with this is that you'd end up with pockets of vacuums in rooms. This can be stopped by combining it with ZAS. By comparing the floorspace to the healthiness of the mobs, it would determine average pressure and apply it to the "vaccuums" where breathing mobs will draw from all the particles a very tiny amount.

there are some other issues with it, but this is close to a best-case atmospherics system possible.

t. minor collaborator

the primary issue with this is you'd end up with rooms

to give an idea of how this can actually be abstracted much farther, rather than relying directly on aoe of the particle mobs, you could generate a dithered "cloud" for the room and depending on where the player sits in that cloud.

But that would require slow updates like Linda. Like once every 5-10 seconds. And it would require some serious autism.

pastebin.com/FNgyH23

...

NO
NOOOOOOOOOOOOOOOOOOOO

KickStarter.
Don't be a cunt and say anything about scamming, especially on here, and there will be crossovers. We know plebbit lurks here.

You'd need, ideally, $60,000 per person you want to bring on, assuming you want to bring them on full time. $250,000 would cover 4 people plus any equipment/software required, assuming they work online only.

Chuck in another $100,000 if you want them to have an office space to work in IE shitty rented out apartment, plus living costs.

And that's per year.

You could probably get $500,000 by shilling it hard on the reddit and a few servers, IE /tg/, Goon, Bay and Paradise. Also, shilling for GameDev grants wouldn't hurt.

Hard part would be finding a team of devs that are actually able to work on this. The obvious solution would be to hire devs from one of the codebases to work on it, as they're familiar with the systems, but then they have to transfer to a new language/engine, and they would have to leave the job they currently have, as you want them full-time.

Someone should post a thread on the reddit to get them to brainstorm ideas and make it happen.

...

never put those two words in the same sentence ever again

commission some furry pr0n in exchange for code - problem solved

I had a look on Reddit just to see if OP was crossposting.

Sure enough, around the same time as this one, a thread started there.
reddit.com/r/SS13/comments/5k3q4j/space_station_13_remakes_what_keeps_happening/

I read through the comments, and the OP worked on Aurora while one of the main /tg/ devs said they would be in if someone was paying for devs to do it.

So there you go, OP. If you're the same nigger as the Reddit OP, you should start a KickStarter on there and pay some /tg/ devs to do it.
Also, some of the devs/exdevs from Paradise would be helpful. Despite the furry players, the coders generally aren't.

Off the top of my head, a few of the Devs you'd want to hire are:
Necaladun, Incoming5643, Iamgoofball (He's really good, and also a cunt that keeps furries and people in line), KorPhareon, CrazyLemon64, Markolie and ZomgPonies.

I know Zomg was working on his own version of a standalone, based off the old SS13 standalone by (I think) Goons devs that ended this year and had the source released, so he might have something decent lying around.

Even if someone reposts it under a new link it won't be the same. RIP in peace.

tfw ligger abuse is over

Why hasn't anybody posted Griefly yet?

Because idiots think that they should recreate current SS13 in full form immediately. What they should be doing is first having a limited WORKING AND PLAYABLE game where you can walk around a station, pick shit up, some interaction, etc. After that, you start adding more advanced interactions (inventories, equipping, etc). Then once you can run around a station, equipping shit, and hitting others, only then do you start working on systems like electricity (which would be first priority), atmos (second or third), and so forth. Look at any attempt at a SS13 remake: they always start adding shit like electricity and atmos before they have anything playable. Guess what? SS13 at release had almost nothing feature-wise.

People think the difficulties are shit like atmos, which is easy, I can do it myself in a week (granted that's because I studied engineering which gave me the physical knowledge for it, but I doubt I'm the only engineer that is interested in gamedev). Here's the real problem: how do you handle replicating all those physics objects in 3D in a 32 player game? How do you handle wires/cables for electricity being put down all over the place in 3D? Most that go over to 3D also make it first person, which is perfect for the SS13 style of game but it makes things much worse in terms of development.

Hacking a door in 2D:

Hacking a door in 3D:

Going first person makes everything a hundred times harder, needless to say the second case needs much more effort from the developers. You can always make it a "minigame" to open the door but then you still have to make the damn thing. This is just a simple door example, everything explodes in complexity when going 3D and especially first person.

inb4 nuh uh atmos is the really hard part its magic
Notice how all the 3D (and 2D) first person SS13 remakes of note got atmos working, and working better than Byond SS13. Atmos isn't the problem.

kys.
It's fun because it's a hodgepodge of ideas in a shitty 2d environment. That's the whole appeal or it, it looks simple, but has a million things going on under the surface, that all require you to actually learn quite a bit of stuff before you can do anything terribly fun.

3D is just shitty, as then you can't use simple sprites, have much higher hardware requirements and it's harder for the average joe to code a funny item for.

It wouldn't have to be full 3d, you know. You could do it doom-style, with sprites for the objects/items; that way, it would be just as easy to add the 'art' for new items as it always has been.

Why not just do it the same way it's done in current SS13? I don't see why you couldn't have a system shock 1-like interface with popup menus and such for door hacking and computer usage.

Wasn't there that one user who was trying to remake the game in DOOM?

Read the thread

Look at the video of it right here in this thread:

He just copied over SS13 stuff. If your remake worked like that, adding more stuff would be just as easy

SS13 is a mafia/werewolf social game (yes, it's social, incredibly so, even if mostly the manipulation side of it) which gives you insane freedom of choice due to depth (you have so many ways of disposing of someone, for example), said depth also helps to make sure people that aren't antagonists still have plenty to do. That's what makes up the core of SS13, everything else is secondary. It's a very deep mafia game with even deeper surrounding mechanics, but that's what it is. This can be applied to first person, third person, top down, text-based, whatever the fuck you want to apply it to.

Apart from the insane explosion of complexity for the developer, first person only has one thing that hinders this kinda gameplay, and that's the voice chat, which can make some stuff hard (e.g. voice changers to sound like someone else are nigh impossible). If anything first person offers more immersion into the roleplay (not just visual, but also with positional 3D audio and all), but like I said above the core concept can be applied to any perspective. Harder to model for? Sure. Higher requirements? Sure, but it could still run on a toaster no problem. Harder to code for? Hardly, Byond code is fucking atrocious, a remake could be done in C# and even give the option for something like Python scripting.

Well I was talking about 3D in the full 3D sense because I was referring to the complexity most devs making it 3D faced (Centration, SSDelta, SS14-1, and more). Nothing stops you from having a 3D world with 2D sprites for a lot of stuff, like others pointed out someone is doing just that as a wad, which is a lot simpler.

What I said initially still goes though: if you're gonna take on a project as big as a 3D SS13 (or any sort of SS13 really), then you better not set off thinking your 1.0 is going to have the same features as a 2D game that has evolved said features over almost 10 years. At release SS13 had fucking nothing, so your SS13 Remake in UE4 should probably at least try to have movement and inventory before you start adding atmos, R&D, and engineering.

Also here's the latest 3D SS13 remake attempt. This one is still alive apparently, kinda, but will likely die soon like the rest. Also it's Russian and apparently has plenty of stolen assets, so that doesn't bode well.

I think the reason everyone tries to make a remake is that just porting the current SS13 to a better engine would be incredibly boring, SPACE STATION 13 IN 3D ON UNREAL ENGINE 4 sounds a lot more exciting.

One thing particularly interesting about the specific Doom source port I'm working with is that it has previously managed to pull together working Mafia-type gamemodes before, even if they work somewhat simply compared to the beast that is SS13 + extensions. There's a WAD called WhoDunIt which is pretty much a stripped-down Traitor, and while not particularly deep mechanics-wise, it has a lot of neat stuff and makes for a good party game. Outside of that, you can find a decent assortment of mods that show off the extended multiplayer features of the engine in general, although a lot of them suffer from being abandoned/picked up by new teams, like AOW2 which is generally a fairly faithful recreation of C&C: Renegade, with alterations made where necessary.

Which brings me to limitations - every engine is going to have strengths and weaknesses, and it's not trivial to adapt things especially when actual BYONDcoders don't even have the fucking luxury of being able to port things from one codebase from another without having to deal with inconsistent bullshit. I think this is one reason a lot of the remakes on modern engines tend to focus on some subset of features at the cost of shoving other things aside, one example being TechCompliant and its focus on the engineering aspects. That, and it's impossible to point out an 'official' SS13 codebase outside of either the Exadv1 original or r4407 or whatever the build was, and that's without taking into consideration versions like modern /tg/, Bay, Goon, CM, and whatnot. The two former ones are my personal nominations, but again, good luck gathering consensus.

Either way, it's all going to burn. That's okay, though, it's part of the spessman spirit.

Do you have a page for this, or something? Are you looking for people to give you a hand, or are you doing it all by yourself? It looks pretty impressive, and DOOM seems like a worthy successor to BYOND in terms of changing engine, funnily enough.

What would you call a particular weakness of DOOM as an engine, apart from maybe not being true 3d?

I've been procrastinating with setting up a dev blog, but at this point I've been working on the current iteration of the project for about a year, so it's likely a wise decision to set something up sooner rather than later.

Right now I'm the only actual developer outside of some sprite contributions and testing here and there, but the plan was always to open up to collaboration at some point where the basics are in place - I respect the open-source nature of SS13, it's not that, but conversely I've seen endless port attempts that simply focused on gathering the biggest team they could, writing boilerplate, and then generally grinding to a complete halt because nobody actually had any vision as to how to proceed from there. A lot of time has been going into experimenting with different systems to slowly piece together an internal roadmap and get a general idea of the order in which things should be worked on.

Probably the largest disadvantage is that you're working with a BSP map format, which for the non-technically inclined essentially means that doing per-tile things is somewhat of a problem, aside from major engine changes. This issue is something that tends to polarize players - while a Security or Medbay player might not be too worried about construction being less fleshed out than the original, it's absolutely crippling for an Engineering regular. While some projects have messed around with building systems in Doom, either using game objects or sector arrays, there are a few quirks to the engine which makes both run sluggishly after a certain level of complexity. If you're curious, go look up Sectorcraft or Survivalism. Apart from that, the text handling also leaves a lot to be desired - native string handling is a complete mess, and even though the Zandronum devs are working on a better chat system, stuff like comms and message modes need some terrifying hacks to work at this point. There /are/ more sane ways, but they need executable patches, and right now I'm generally trying to stay with the native engine features until they can't go any further.

Some modders might disagree, but I honestly feel like there's a fair bit of overlap in philosophy between the two, at least from an accessability and extendability perspective. Doom WADs do have their share of cluttered code, and it's not always pretty to look at, but on the other hand it's an engine that explicitly rewards tinkering. It's like BYOND in that the scripting language is a little rough around the edges, the game object definitions can be somewhat messy if you don't think ahead, but it's still possible to make stuff that's elegantly written and works well.

just don't fucking open up brutal doom and look at the code, you'll be scared away from the engine altogether because what the fuck did he mean by this