Spiral Knights - Open the game's level editor up to the steam workshop

Open the game's level editor up to the steam workshop

32 replies [Last post]

Sat, 02/14/2015 - 17:59

Fehzor

Take the existing level editor, provide documentation for it, and build a means of loading levels onto the server and testing them. So I'd make a level, send it to the server using some sort of upload+test option in the menu, and the server would put me in it. I would get zero loot for doing this. That's all that needs to be coded- uploading a file to the server and running it as a level just once, maybe twice and deleting it off of the server. If it takes up too much traffic, limit the number of levels that can be uploaded and ran to only a handful at a time. If someone else wants to test, they'd have to wait.

Why is this a good idea? Because right now there isn't enough content floating around and the players want more constantly. You don't need every player to have an opportunity at creating levels, just a few that are willing to learn and really want to. Then, once or twice a month look on at the best levels. Take those, and put them in the game. Just like that- a truly infinite world begins to build itself.. and if that wasn't the intended nature of Spiral Knights then I don't know what is.

Sat, 02/14/2015 - 18:39

Chaos-Mist

Warning: Superiority Complex.

I just replied to another thread that you said this before, then I saw this. xD

Although, yh. +1

Sat, 02/14/2015 - 19:10

Fehzor

I never formally wrote this as a suggestion, and it comes up so often and I keep typing it.. so I thought that it would be best to just give it its own thread, since it really is one of the better ideas in my opinion.

Sun, 02/15/2015 - 19:49

Gbot-Vtwo

Yeah. +1

I want to make a level.

Sun, 02/15/2015 - 19:58

Vohtarak-Forum

I'm vohtarak, call me by that name

Or...

Put something like it in the game itself so its not restricted to steam only

Sun, 02/15/2015 - 20:37

Eltrooper

Sounds good to me

If this were put in place, people could test their levels and work out bugs. They would need to be able to save those levels offline of course, and perhaps allow other players to join their party to test them out. Also, you could simulate loot drops and utilize any weapon while in that level to test out gear for that level (but not have the loot added to your inventory in-game or those weapons added to your inventory). Have the main editor be from above and/or from any angle. While testing, the knight who is creating the new map could pause and use a system similar to GH design mode for small things they notice during the tests.
So, the level editor would have components provided by OOO and have limits like length and where things can go so that a knight can physically get through?
This could be quite a bit of fun, though only people who really want to will be able to successfully create a map worth adding to the game because of the time and effort needed to create a decent map. I suppose that is a good thing...
~Elt

Sun, 02/15/2015 - 20:51

Fehzor

@Oohnorak

I mean that would be the optimal thing to do from a design perspective, but I'd rather they take the more realistic approach-- coding the level editor into the game and making it usable would be incredibly challenging from a coding standpoint and isn't entirely necessary. Rather, simply opening up the level editor and accepting levels would be enough.. consider the current number of new levels added as permanent content we see per year- about one or maybe two. If the developers accepted one new level per month, that would be a HUGE increase.

As far as using steam, it's just the easiest way to go about it and since the community editing levels would likely be tiny it wouldn't really negatively effect anyone. Nor would it exclude non steam users- the map makers would use steam as a forum for communication, but would store and work with local copies of their maps. Ergo, a Kong player could test their maps via the upload.

@Eltrooper

We can't talk about features of the editor because they're already in place-- the level editor in question is the one that Three Rings developed and currently uses. It's free to us right now, but we can't test levels and there is zero documentation.

Mon, 02/16/2015 - 06:09

Eltrooper

Ah.

-Thanks for the clarification.
-In addition, levels that "win" the devs' approval and get added could have the stamp (similar to TF2) of "Community Level created by:[Insert Name Here] under the level's name.
-Monthly would be great if there were good-quality maps created monthly.
@Fehzor: Why exactly would Steam be the easiest way to make a level editor for SK?
~Elt

Mon, 02/16/2015 - 10:36

Fehzor

Well first, the level editor would be a stand alone package as it currently is and not hardwired to steam. The steam workshop only allows us to share map files and comment on then in a highly organized manner.

Second, this needs to be a community thing, so that players and developers alike can see all of the levels and judge them by popularity and discuss any other factors on levels. Now, we could make our own infrastructure for this, but one already exists- the steam workshop. Using it would be highly advantageous over spending months on making an entirely new sharing system. It is also noteworthy that the game masters are the steam forum moderators for our game, and that their duties could be expanded to any sort of workshop feature.

And finally, using the steam workshop looks fantastic- it allows the developers to make themselves out as being open minded and integrated with their community, properties that can make our break whether intelligent and worthwhile people stay and play the game. If you get these people, you win. That's just how it works. This is particularly true in that the steam workshop isn't just for spiral knights, but for all games and so draws from a larger number of players.

Tue, 02/17/2015 - 12:46

Fangel

Word.

Level editor is always on my list of things to have added, and Fehzor here is great at giving suggestions for it.

One integrated into the game would be the best option, but we already have some steam-only features (steam trading, promo mask, voice chat), so it isn't a bad call. Plus the rating system and all that is already in store. Having that feature be advertised on the steam page would also be great for bringing in wonderful ideas and new people.

Tue, 02/17/2015 - 15:10

#10

Eltrooper

Alright

And if anyone wants to level edit who isn't a Steam knight, they can join Steam and merge their account to Steam.
~Elt

Tue, 02/17/2015 - 16:00

#11

Fangel

It is an option

And if anyone wants to level edit who isn't a Steam knight, they can join Steam and merge their account to Steam.
Alternatively, they can create a steam account and make an alt account to create levels for the workshop if they do not want to merge.

Tue, 02/17/2015 - 16:53

#12

Fehzor

I mean you could also make levels just for fun or practice without submitting them... in fact, the ONLY time Steam would be required would be when you go to share it with other people so as to garner support and have it added to the game.

I tried to explain this above, all clients would be able to test maps via probably the escape menu (which would have an upload level space added), and the software for the map maker would be made public via a link on the wiki, a space in the new tab, and of course would be added on steam.. so if you hate steam because it slows your PC and you're just too pro for it, you wouldn't have to use any of its functions for more than about 15 minutes while you put your level up for review.

And thanks for the support, Fangel.

Fri, 02/20/2015 - 16:05

#13

Cheshireccat

A Tweak to make this possible with no extra dev! OOO: take note!

There's an in-development MMO called Trove. It's a minecraft-block world and their levels they are relying on programatically generated content (like SK).
BUT! They are having the players design all of the weapons!

Here's how it works (and could work for SK):
1. Open up a minecraft editor. That gives you 3-D block building elements.
> Using these tools means OOO does not need to develop a level editor of their own.
> If we use a Minecraft Editor we will all be working in a standardized format.
2. In the block designer you are told you have a certain max size and you can not go beyond it. You are also told to use blocks of a certain size to build the weapons.
> The maps built in this editor are *not* going to be literally translated into SK maps. They will be an easily-viewed representation of a map.
> SK levels are already a grid. OOO just dictates the map dimensions and the block size to use, which will then match.
3. Certain color codes correspond to certain components, like where the weapon is gripped by the game sprite.
> Establish a visual legend for user creation. Blocks colored #FFFFFF = a wall. Blocks colored #FF0000 are fire traps. #FFA500 = monster spawn point. Etc.
> Instead of building a WHOLE MAP like people suggest above, let's build the maplets. The individual map puzzle pieces that are automatically stitched together. For OOO that's a MUCH more manageable way to evaluate user-created content.
4. Users submit their designs in a subreddit that is monitored daily by someone on the dev team who considers all of the submissions.
> Perhaps our SK community liason would step in here?
> The SK devs would then use the "minecraft-esque" custom map as a blueprint to build an actual SK maplet.
5. Accepted designs are given an attribution in the tooltip for the weapon.
> We already have click-beacon tips on maps that pop up a message when you click, like you see in an arena. Just drop one on the map that says, "This area mapped by Cheshireccat"

OOO could start by accepting maplets for the Danger Rooms and Arenas. Both are very simple, right?
If that works accept maplets for clockwork style levels. Those would be the next simplest.
If it works they could expand to Clockwork Keep and eventually to the harder ones like Wolver Dens and Compounds.

--Cheshireccat

Wed, 02/18/2015 - 12:22

#14

Flawedknight

Sure! +1

I'd settle for something even very simple like the following

1) give the ability to set the environment (just like the guild hall) that determines monster status
2) give access to *all* existing room tiles (even from static levels)
3) have a simple algorithm that counts the number monster spawns and boxes to make sure the level isn't out of balance.

maybe monster family is determined by the highest number of specific tiles? Perhaps we can use something like a guild hall editor to reuse more code? (although I realize the walkable area is typical preloaded and you have to reload to change the room layout)

blah blah blah,
Flawedknight

Wed, 02/18/2015 - 20:56

#15

Fehzor

I'd absolutely love for them to make something like the guild hall editor but for everywhere and everyone. That would be awesome, if they're open to it.

Making the game compatible with Minecraft would be a bit.. odd though. Like, they'd have to really make their game around the idea of it being built by players to open up that sort of thing, and building that sort of thing right now might be a bit odd, considering how players would need to plan out waves of enemies.

The real problem though is that they'd have to go make an entire editor or whatnot and I'm not sure they're willing to do that. This on the other hand is not hard. Just the rewarding part of the hard.

Fri, 02/20/2015 - 16:16

#16

Cheshireccat

@Fehzor Sorry! You completely misunderstood

I must not have explained it well.

I am not suggesting they make the map compatible with Minecraft. I am also not suggesting that they design a new editor or APIs.
I am suggesting that we use a Minecraft editor to make easily-viewed blueprints in a standardized, sharable, comparable 3-D format. Minecraft editors have already been written and we would be repurposing them.
Devs will still build the accepted designs into actual maps, but using our block maps as the pattern.
Because Minecraft blocks and SK maps are both in a standard grid, this will make the matching relatively simple.

1) We players would take on the responsibility of design.
2) Someone at OOO would take on responsibility of content evaluation.
3) Another person at OOO would rebuild the maplets and release them into the clockworks.
Even if OOO released a custom map editor or APIs, they would still need to do 2) and 3). This saves them some front-loaded dev work in exchange for a little more manual work on the tail end.

I do see your point about the monster waves. Maybe we do the rest and SK figures out where/how the monsters trigger.
Either way I like your core idea--involving the player base in shaping the game.
--Cheshireccat

Fri, 02/20/2015 - 17:17

#17

Fehzor

Oh. But we could already do that, and just leave pictures in the suggestions forum. The point isn't to suggest things to them, it's to actually do the work of getting new levels so that we can have more all around.

Mon, 02/23/2015 - 09:19

#18

Cheshireccat

Well... why don't we?

I understand that there would be huge benefits to a full design mode experience. Other games *cough-TF2-cough* have very successful user design modes that have created actual distributed content for the game, but the odds of OOO giving us that power soon are slim.
So why don't we do what you suggested? Visualize some proposed maplets here in the forums in some picture format? It can't hurt, and I wouldn't mind taking a whack at designing a few maps myself.

Can you think of a better method to design a visual of these maps than the block editor? Is there another option that might be more user-friendly or clear?

--Cheshireccat

Mon, 02/23/2015 - 10:42

#19

Fehzor

The problem is that players have no credibility and very limited knowledge of how levels work, and establishing this credibility and knowledge as well as displaying a worthy idea as well as getting that information to a developer at the right time in the right situation would be key to successfully getting a level in the game.

The developers, if they really wanted our feedback, could solve this problem in a number of ways-- hold a contest every month, with the prize being your level in the game. They could tell us when they want levels designed, and provide a common guideline, which could use a block editor or MS Paint or even text.

Our options are more limited- yes, we could establish our own format but we'd have nothing to go by and it would turn into a handful of people speculating on what a good level design should look like. We aren't developers and we don't have the necessary credibility to judge one another, and that's about where it falls apart.

The best we could hope to do would be to use the existing level editor that is available to us, which has no documentation and no means of playing the level created. If we had someone with experience look at the source code for it and the way levels are saved, we could also write an interpreter for them. It wouldn't be able to run the levels, but it certainly could double check them for whatever they need, and see if they're plausible and comparable using statistics. How many enemies on average does it have? How much loot would you expect? Is it possible to hit all of the switches? Once you have the level loaded into a program designed these things would be plausible to calculate.

The genius of doing this would be that it would attract attention and open the door for players without developer help, and allow us to bring what I suggested in the OP to a player ran reality. The problem with it is that it would also be time and labor intensive, with absolutely no results guaranteed... and again, the developers could help us out A TON by providing documentation for us if they wanted to.

If you'd like to try and have the right background (and I have a strong feeling that you do) I'd be more than willing to go in on it with you, and I'm sure more people would be more than willing to join our cause once we have documentation + an executable level editor in place.

Tue, 02/24/2015 - 21:30

#20

Skepticraven

↓

@Fehzor

Finally noticed you made a thread on this so I could comment about this idea.

"Take the existing level editor, provide documentation for it, and build a means of loading levels onto the server and testing them."

Let's break this down a bit.
1. Have you ever worked with any level editing slightly less straight-forward than something like JRPG maker?
I have, specifically this one and this one. Sure we could theoretically reuse all the models that OOO already put into this game... but that still leaves items such as z alignment, pathing [both character and projectile], procedural generation, monster spawns, triggered event locations, button+switch-linking, loot balancing, difficulty balancing, and a whole ton of other mechanics.

Making a level editor user friendly is a whole different monster, especially a procedural generated one to those inexperienced in such methods.
Let's face it. OOO seems to be running fairly alone with their Java-based game [compared to unreal, source, or unity engine games] and realistically there are probably 5 or fewer people that know how to build levels.

2. Documentation. Not really a difficult part, but a rather boring and extremely time consuming part. Refer back to the size of the team.

3. Loading levels onto the server and testing them. Big difficulty here.
a. Server load will become an issue. Logistically, uploading and testing from their server is just a bad idea. All other games that support map editing have a "testing" version of the game on a local engine to load custom maps.

b. "I would get zero loot for doing this." No, they should be separate servers... like the existing test servers where drops would function as normal but the accounts are in no way related.

4. "I'd absolutely love for them to make something like the guild hall editor but for everywhere and everyone."
One major problem with this... it covers almost nothing of the mechanics that need to be addressed in #1. Currently it is a pick+place on the 2D grid. Attributes could be applied to different objects to inherently apply different pathing, spawn locations, and such. Triggered locations, button+switch-linking, and procedural generation would definitely require this "current state of an editor" need a big face-lift.

=====

Could players do it? Given a good set of players, sure.
Will players do it? Judging by the number of players that edit the wiki... This game simply doesn't have enough of a following with the technical know-how to have it function without SIGNIFICANT hand-holding from OOO staff. A degree of hand-holding that would bring all other development to a grinding halt for a longer duration than the gunner update.

@Cheshireccat

SK is a 2D game, not 3D. Using a 3D editor would seem fairly... overkill. I linked to JASS that blizzard uses in my comments to Fehzor... which would be almost exactly the type of "editor" that would need to be used. It has a fairly high learning curve [since you need a dictionary of functions to do anything], an extremely inefficient GUI built around it, and inherently supports the ability to insert malicious code into maps for execution.

Tue, 02/24/2015 - 22:24

#21

Fehzor

Thing is, I didn't say to make it user friendly because I have used the sort of level editor in question and don't think that everyone should be using it. Like you said, it wouldn't be feasible to have any large number of knights making and testing levels all the time.. but if say, five people, ten people, fifteen started using it, then I think it could work out pretty well. Sending a file to a server, loading that file, and then deleting it after a little while shouldn't be too hard. I mean, the existing server sends how much data to how many players every time we play a level?

I do really like the idea of using the testing server instead of the main one for this though. I'll have to add that to the OP.

--A few other interesting tidbits--

"One major problem with this.."

Really? You only found one major problem with that?

But it could be done regardless; imagine if they color coded everything and included more abstract objects. The red gremlin would open the red door when killed, as would the red party button, red normal button, you name it... and all of those things would also spawn the red set of enemies, or trigger the red spikes. To change the height and install ramps, just use height marker objects. To define a boundary around the level, use a series of boundary objects.

The problem isn't it being impossible, just... incredibly time consuming and borderline impossible. Which is why I say that it would be wonderful if they went through with something like the above, but that enabling the level editor would be a pretty good option imo.

Finally noticed you made a thread on this so I could comment about this idea.

Yeah dog, if I interject my ideas everywhere as part of my evil plans I don't have to take any valid criticism.

It's great, you should try it. Love ya.

Wed, 02/25/2015 - 10:17

#22

Cheshireccat

@All

2-D vs 3-D_______________
Yes, it is primarily a 2-D game. But there are elements of 3-D.
My block-editor thought process was that walls and obstructions could be raised above the walkable map area--just to make it easier to "feel" how the players would move around the map.
Fehzor mentioned this 3-D element too: ever fire a bullet at a slope? :P When Fehzor posted this my first thought were some unique ideas to take advantage of that mechanic.

Level Design______________
I reiterate my previous point: we should not be designing whole levels. That is too complex for us amateurs.
It's a programatically-generated game; let's just do maplets (tiles? puzzle pieces? Next time a GM is in Haven, someone ask OOO what they call them!)
Accepted maplets would be seriously helpful! Imagine how many people would start playing the arcade again if they knew there would be new maplets constantly added to the game?

You're right; monsters are HARD.
You need to figure out monster group compositions. Treasure proportions. Knight-triggered spawn points. Let's take away monsters from this proposal and leave that to the professionals.
The only monsters players should be able to place are turrets and shankle/wisps. Turrets are integral to map design (imagine that S shaped maplet with the turrets and the spikes--without the turrets it's just a weirdly shaped map).
We might want to make suggestions "I think there should be bats spawning in the NW corner, because on this maplet design bats will really challenge the players" but beyond that let's skip monsters entirely.

Components_____________
So what does that leave us? For the basic core elements:
Floors
Floors Slope up
Floors Slope down
Walls
Button gates
Multi-button gates
Monster gates
Party-button gates
One-way gates
Switch gates
Fire/Spike/Freeze/Poison/Shock traps on timers
Fire/Spike/Freeze/Poison/Shock traps on switches
Thorns (aka spike traps that don't go down)
Switches
Blocks
Bushes (aka gravestones, aka blocks-with-treasure-inside)
Concrete Blocks
Crystal Blocks
Vanishing Blocks
Exploding Blocks
Vortex Blocks
Pots
Oil Pots
Fire Pots
Turrets
Wisp/Shankle
So: is that it? 26 tiles that comprise the majority of the game? Did I miss any?

Sadness_________________
It's really too bad that this is unlikely. You're right that it's a hugely complex dev project and only 5-10 players would upload serious maplets (and not just paths that spell out "DICKSBUTTSLOL"). But I bet those 5-10 players would submit over a hundred maplets in the first three months.
It would be SO MUCH FUN!!! :P

Wed, 02/25/2015 - 11:14

#23

Skepticraven

↓

@Cheshireccat

Buttons and gate location could be shown. They still need secondary work to link them to events and such.
Same thing with Wisp/Shankles - they need more information that just location [direction of movement, "2 direction, or 90deg turns"]. These are likely object states that can be set. Think spiralspy objects with a couple more parameters that need to be set.

Monsters arent hard. Nor are linking all these items together.
The hard part is making the combination of all these things available for users to generate content. Do they have a custom scripting language for handling events? Is it nice and easy to read? We don't know. That is part of the game code that we never see, because it is run on the server side.

For the comparison with the other map editors I've used... WC3's JASS language most closely resembles the missing part of the puzzle in this game.
It is an event-based language which fits exactly. With this you could even create new custom AI for monsters.

Anyways... the list could be much more simplified.

1. Terrain/Pathing
2. Objects [pots, keys, boxes, monsters]
3. Trigger Areas
4. Linking [events that describe how the 3 above interact]

A visual editor could easily handle 1-3, but 4 doesn't exist visually.
In fact, the GH editor pretty much already handles #2.
I would argue that linking and balancing are the most time consuming aspects of map generation. [Back in SC1, I was one of the original authors of "Sunken TD", and also contributed to a WC3 map called "Island Defense" - it's not difficult work... it's just time consuming work.]

@Fehzor

"Really? You only found one major problem with that?"

I pretty much inferred that using the GH editor would be just as effective as using MS paint to draw a picture or a map. I sure think that is a major problem and anything else in comparison is less important. As I just described with my response to Cheshireccat above, it needs the event triggering language.

Sun, 03/01/2015 - 15:05

#24

Blandaxt

Love the idea

Love the idea. Though having it ingame would get more players interested in my opinion and more player would know it exist.

Sun, 03/01/2015 - 16:11

#25

Fehzor

Yeah that's what a lot of people have been saying. My response is in post number 15-

I'd absolutely love for them to make something like the guild hall editor but for everywhere and everyone. That would be awesome, if they're open to it.

The real problem though is that they'd have to go make an entire editor or whatnot and I'm not sure they're willing to do that. This on the other hand is not hard. Just the rewarding part of the hard.

Tue, 11/24/2015 - 05:02

#26

Jenovasforumchar

an already existing editor?

'...take the already existing editor and...'

pardon me, there is already an Editor here??

Where can I download it? I would like to mess around a bit with it and may it just to post the result as png to explain a certain idea or explain the Setting of a hardcore Arena how it is meant to be (refering to a certain Topic a few days ago) <3

Edit:
Necro, just realized, lol :P

Tue, 11/24/2015 - 09:09

#27

Falminar

Fehzor? What Fehzor? Where did Fehzor post in this thread?

Is Six-From-Syrup Fehzor? What?

Tue, 11/24/2015 - 13:09

#28

Jenovasforumchar

@Vana

obviously she is another char from the same account. I also did wonder why she took another in front of her Butterfly though.

But nevermind this: Where can I download that already existing editor? My fingers are tingeling :3

Wed, 11/25/2015 - 02:00

#29

Thunder-The-Bright

algorithms are the solutions to any problem.

here's spiral spy, it's been here for a while.

Wed, 11/25/2015 - 16:34

#30

Fehzor

NEW! Name, same great Fehzor.

The editor is in one of Three Rings free software packages, I believe that Spiral Spy (the user) got it working a while ago so contacting them would be a first step. That said, the infrastructure for actually creating a level is missing.. and that consists of the following:

1. Easy access to the editor i.e. ability to just download it
2. Instructions on how to use the editor
3. Ability to test levels
4. Willingness (by the devs) to use user created levels
5. Some way to share and consider those levels

This is in essence the steam workshop, and without it you will not be getting far.

Wed, 11/25/2015 - 17:29

#31

Jenovasforumchar

Spiral Spy

... huh, Spiral Spy is a level editor?

From what I have seen from Spiral Spy yet, it only allows you to customize a knight and preview how certain loadouts would look like.

I will take a look into it tomorrow - I didn't expect that it can actually create a map :o

Wed, 11/25/2015 - 17:57

#32

Fehzor

No, SpiralSpy is the model viewer as well as a name of the player propagating that level editor. Spiral Spy found the model viewer used by OOO and by Spiral Spy to make Spiral Spy in the extra packages they made freely available... but also found the level editor they used to make levels in a similar location, and wrote an article about it which may or may not still be up.

Search

Open the game's level editor up to the steam workshop