Caves, caves, caves | Morroblivion

Well I started to replace the Morrowind caves with tilesets from some modified Skyrim sets. Initially my plan was to replaces the tileset and get people to add clutter around these in the hope that it would be enough to satisfy ourselves and gamers, however...

I was systematically doing each cave alphabetically and got through A, B, C and by the time I got to D, 40 caves later to be exact, I was about to blow my brains out with the repetitive nature of them. Admittedly I was doing it in a very drone like manner with one to one model replacement with no real thought on design.
Lots of straight tunnels that bend at 90 degrees, with 24 different set shape rooms. It doesn't take long before a gamer is pretty much staring at the same thing, cave after cave. There are 200 caves in all! Lets just say they are a major feature of Morrowind and a huge impact on the Skywind experience.

When I first started on caves, we had no custom environmental assets. Thanks to Ravanna and friends we have Bitter Coast, Sheogorad and West Gash regional assets already.
The grand plan is that all of the existing tilesets will be replaced with regional cave assets that are clustered together via the Creation Kit to get templates for hallways and rooms/caverns. The existing tilesets only act as a guide and will be deleted once replacement is done.

I've done some very quick proof of concepts for Bitter Coast, Sheogorad and West Gash. My initial conclusions is that with maybe a few customizations of existing rock assets we will make regional templates for caves in the creation kit, rather than trying to design whole tilesets in 3d programs. The reason is mainly due to 3d modeler specialists being able to spare the time, but also there is a good case for being able to create more variety in the long run.
Don't get me wrong there is still a desire to have 3d modelers helping with coming up with easier templates, but I'm not expecting too much of peoples time, as a lot the ground work has already been done.
So we are not aiming to make tilesets in the traditional manner. Rather with some basic assets we can make them in the creation kit and copy and paste as necessary to get quicker results. This will allow for much more variety and flexibility as well as putting the time on Creation Kit users rather than modelers, where we will be able to get more peoples time.

I'll pretty much be exclusively be working on caves until they are done. With 200 caves and I suspect I'll be doing a great number of them. It will definitely take me to the conclusion of rest of Skywind world building.
I think for me personally it will be really nice and keep me focused to stay on one thing. Although I was really keen to work on exterior world building, I feel like caves needs a champion, somebody to exclusively organize, direct and do creation kit building. Plus caves are super fun, so I'm all in.

The good thing about having somebody exclusively working on caves is that holistic planning and directions will really help. And I've been thinking about this really hard for a couple of weeks now and I've got some very solid ideas.

West Gash; so the idea is to turn the caves into open world deep ravines. I've done a proof of concept with the cave "Milk" and I'll declare it a success on my behalf. The cool thing about this is that not only do we make the "cave" systems more interesting but it also gives the exterior world more points of interest, as well as landscaping by proxy. Cave areas are relatively small by exterior world comparison, so it doesn't cover a lot of land area.
I've marked the locations on the exterior worlds where these will be with black rectangles.
The layouts will basically be the same, its just that they are open topped. Arches are used when cover seems approprate and there will still be closed caves to keep exploration different and interesting. Lets say about two thirds of each system will be split between ravine and cave.

Milk now as a ravine

Exploring

Bitter Coast; using the squarer shaped terraced rocks. So caves will have a stepped square shape in general.
I think if we create some rectangle blocks (3d modelers) then we can create the necessary hallways and rooms in the creation kit. Think lego. This method can also be used for the flooring (steps), so gives massive flexibility to slopes.

These screenshots are proof of concepts, don't take them as what will actually be used, as I whipped them up in like 10 minutes.

Sheogorad; I used the slate arch asset to show how basically just using one asset can achieve a slate look for hallways.
I think if we create some different shaped slates (individual) then again we can construct rooms, but probably needs more thought. The flooring will be stepped slates as well, so same as Bitter Coast method, gives unlimited scope for slopes.

There is a good article on level design by Bethesda, that touches on the faults of Oblivion caves http://blog.joelburgess.com/2013/04/skyrims-modular-level-design-gdc-2013.html

The essence of it is that basically human (actually all animals) are very good at picking up on repeating patterns. At the moment Morrowind is even worse.
The challenge for game designers is making sure the gamers feel like they have not seen something before. Oblivion failed to do this and it breaks immersion.
Bethesda really stepped up their game for Skyrim, they went into overkill with the cave model resources. Over a thousand basically, compared to around 130 for Morrowind. You can tell by the number of models simply not used in Skyrim or with only one or two references that they wanted to make sure level designers had all the tools they need to achieve non- repetitive design.
I then proceeded to check all the cave related systems in Skyrim and it confirmed that a lot of time and thought needs to go into each one to achieve a level of interest, realism and uniqueness that can only be done by human, manual placement of assets. The tilesets are only a small part of the design, where as in Morrowind they were the entire system of creating the caves.

Now it might be a little unfair to compare what we are doing, given that there were probably at least one full time modeler and several cave designers for the entire duration of Skyrim development. Say 3 full time employees over 6 years for caves alone.
However we do set high standards for ourselves, and I feel like I have a great strategy for them without needing too much valuable modeler time.

I'll leave it there, I'll explain ideas more when I have further interest in cave development. I'll just keep beavering away (try to imagine a grumpy mole).

Last edited by grumpycat on 30 Nov 2015 - 22:32.

love the idea for the ravines

love the idea for the ravines but oh lawd the immersion! What does it look like with less gaps?

I love the idea of the

I love the idea of the ravines as well, but doesn't this bring the risk to break the level design of some of these caves?

Great topic and article (I

Great topic and article (I read whole)!

Ravines are interiors or exteriors? I see that they are interiors, so maybe "roof" should not be too open and sky-visible. I just don't want invisible walls on top to avoid getting out. Remember about levitation. Are some signs of open roofs on surface? Maybe some unmarked cave-entries on roof?

----------------

200 caves sounds like a lot of work. So it's hard to avoid repeatability. I was doing some level design a few, when I was doing in Construction Set (MW) ages ago. I'm not sure but I think they are things which could help in your workflow.

So you need only pencil and paper! There is technique in logo designing, when you don't have any idea what to do. Just close eyes and draw... whatever. Don't care if it looks like a crap. This is always some step forward.

[External image]

Like in example of clouds when you looking at them you give some of them conceptual meanings (this one looks like a cow, this one like wolf etc.) You choose which you like best and you have shape of cave's wall or some shape of cavern. You can also take some lines, and turn them 360 degrees to give some shape, like lathe in 3ds (link in page 8) Which you can work later. You can create a lot random shapes which you can mix together. Which will help you keep some diversity. I don't know how your skills are good in 3ds or Blender, but these could be useful for creating space organic concepts in great quantity, quite fast. Just by playing FFD feature (or other modifers) and mixing them (3ds sutff) Then you can divide them on "categories" which you "invented".

I don't know, Bulk-shape, T-shape, Y-shape snake-shape whatever-shape. You can also can mixing them to create new ones. I'm using it on very begging in some concepts (mostly for organic looking). Random things later I add them meaning.

In that way you get great numbers, which you can mix with each other to get even bigger numbers.

You can also mix silhouette and simplify them to get some interesting shapes too. In horizontal and vertical section view. You can attach them to main/small chamber shapes, corridors shapes and so on. Just how you like.

So from this you get 3x3x2=19 possible shapes in easy way. This post is somehow close in my way of thinking in this topic (creating totally random things in shape) You can create some caves' list with main features and so on (how many big caves/small, vegetation/noveg inside, cracks to exteriors/ or not, epic/cool/boring, etc.). It is iteration, so it can take time to sort out things. For me it is always easier to expand ideas, and cut down worse, improve some of them than going one by one example. Of course this will be only guide than stiff plan to keep everything diverse but it's better than nothing.

Then you get big visual library to choose. But if is it not big enough, you always mixing, changing, cutting down them and so on. This part is quite crucial to decide what to reject, to improve, to accept. A lot of work on paper before will be anything done. I use it in concept art, but I think here, it would be useful too. There is also photobashing technique for tumbnails to choose and expand, but in this one it needs some experience.

And of course this more about overall direction (shape) with existing assets than creating new ones.

I hope you understand what I mean.

A suggestion regarding the

A suggestion regarding the ravines:

Interior lighting with the exterior sky, which would allow for some great lighting coloration where the upper parts have different colors than the lower playable space, like this:

[External image]

I made a quick proof of concept to see how it would look ingame:

In this concept, I also made use of the blowing ash effect to make it look like dust drifting on the dunes above is falling into the ravine, with what daylight does reach in reflecting off the dust particles.

Also, it might help make things less repetitive for some West Gash caves to be ravines, while others have covered ceilings. Something like this could be nice, sunlight let in through small holes in some of them:

[External image]

I love the idea for the

I love the idea for the ravines, and I look forward to wandering around looking up at the sky, but I have to ask: what will happen when a levitating player floats up out of the ravine? A loading screen? what if someone falls in? invisible walls?

Levitation is awesome, and It MUST be a part of skywind, but I can see what a pain it is for developers to work around it ;)

...I cant help but feel like there needs to be a huge discussion on how to handle it, and what is going to happen when players float into interior cells.

Other than levitation, the

I don't see levitation as too much of a problem, a barrier at the normal ceiling height would imply that the spell can only take you so high with the amount of magika it requires. Other than levitation however, there is a somewhat minor problem with the open-roofed ravines is in terms of consistency, considering they only exist on the interior. Perhaps using the fissure assets could give the appearance of narrow ravine tops on the exterior? I would still suggest making the ravine look a bit less common, with ceilinged West Gash caves as well for variety.

This wouldnt be easy, and

This wouldnt be easy, and would need quite a bit of coordination between the landscapers (me in this case) and the person doing the ravine, to at least try to line up the tops of the caves and exterior fissure thigys. Oh and to make the fissures fit, the tops of the caves would need to get narrower as we go up.

I also think it would be cool if steam would come out of the bottom of the cave, and out through the fissures, just something to think about.

Landscapers could just make

Landscapers could just make ravine part in exterior around main cave entrance.
Would be nice to have more rough red stone. Such polished forms can look alien in WG.
This is a simplest variant. Also this can effectively hide entrance of cave, making it natural bandits/contrabandist's home.

SOME PICS

[External image]
[External image]

They will be exterior world

They will be exterior world eventually, it just that its easier to design them as an interior first. So I've marked the locations as seen below. This one is actually two caves Milk and Odibdaal.

OK, i did a mock up to give you the idea. Please imagine it properly landscaped, this will not be used in any capacity, i just plonked some shit as people seem to have a hard time imagining stuff

On the edge peering down

So there will be a number of ways to get down to them. Normal cave entrances that have tunnels down. But actually also I was thinking about some tomb raider style platform jumping (yep no shit). I even got a water cavern jump down to get one, similar to that one in Skyrim where you fall for ages. Paths back up I'd like to make well hidden, so that players are like, hmmm how do I get down there!?

I love the idea of the ravines as well, but doesn't this bring the risk to break the level design of some of these caves?

Surprisingly not really. Don't get me wrong some thought needs to go into some of them, but from a layout perspective they are exactly the same. NPCs, platforms, loot, all the same, the paths to get to them, all the same. They really are just the caves but with a twist.

Interesting stuff Lukkar. I'd like to see if you wanted to help with making some of the templates in the Creation Kit?

Played around with Bitter

Played around with Bitter Coast caves concept today and excited by the results for the effort. I've cut up a couple of Ravannas limestone rocks. The reason is that you really want to be able to see through them when designing, otherwise areas become closed off really quickly.

The methodology if anybody is going to be interested in doing some of the Bitter Coast caves, is that I used the Skyrim mines tileset as the base for planning. In the end its more of a guide, but it is what is used for the flooring. Then used walls and ceiling assets to construct around it. The good thing about this is that its forgiving if you have tiny gaps between assets as the tileset seals it off.

Cavern

Hallways

Only 2 models used, one rock cut into 4 for walls) and one cut into 2 for the ceilings. There will be another 6 limestone rocks cut into 4, so another 24 assets to give variety.

These will be ready for next release, so once again if anybody is keen, let me know.

I've seen a similar approach

I've seen a similar approach back in Oblivion, David Brasher (legendary mod maker) used exterior rock models to construct interior caves without using any interior tilling kits. It just takes a lot of time rotating, scaling and positioning all those rocks since you can't snap them like kit pieces

[External image]

scumpunx was working on some

scumpunx was working on some stuff to use for caves.
can revisit those again some time.
I cant output much cave stuff atm unless you'd like to use Molag Amur as a test bed for new cave methods for a while.
What is the minimal tile set needed. flattened cylinder straight, 45 degrees/90 degrees,up/down,
Lukkar can probably hash out enough ideas, its just cylinders its pretty easy.

Anyone wanna start the ball rolling with molag amur styles? ideas? lava hollows?

Anyone wanna start the ball

Anyone wanna start the ball rolling with molag amur styles? ideas? lava hollows?

Did a quick discussion with Erik about lava hollows caves, so that should be on the cards. The minimum for any cave tileset is; straight, up, down, 90 degrees, dead end, 4 way, "door" frame x 2 (left and right), and a twist piece for left to right.

But also having a 3 way would make things better (that came out more sexual than intended).

Skyrim then had variants of bends left and right, ramps (longer up downs) and longer corridors. Also pieces with hollows in the roofs. However if we just get the bare minimum would be enough.

Scumpunx made some stalagmites, which look great.

Plus he also made all the moss and vine assets, which I intend to use extensively for Bitter Coast caves.

Thought I'd post my good deed

Thought I'd post my good deed for the day. Added viney roots to the Bitter Coast cave assets, so now we have very quick method of having overgrown creepy caves. I used speedtree and definitely need to play around with it more to perfect the methodology to get best results, but at least you can see the concept.

Alrighty then, due to

Alrighty then, due to creative differences I will be maintaining a separate mod for the caves of Skywind.

Its named "Skywind - Caves.esp" and I will be maintaining the link in this thread through out the duration of this project.

You can enjoy the first two caves "Milk" for West Gash and "Aharunartus" for Bitter Coast.

You will need the latest master "0.9.4.0"

Download here:

https://drive.google.com/file/d/0B1vwmz6hbHGPOGw1aHlVMzAtd28/view?usp=sharing

Ahh, may just be me, but it

Ahh, may just be me, but it seems like an awful waste of effort not to have these caves in the main game, because otherwise I don't think they'll get touched plus conflict issues may arise. Whatever creative differences abound, we should try to resolve them, rather than branching into multiple paths.

I second that. Not touching

I second that. Not touching these at all would be a waste and doing work twice is always bad, too. Even putting aside that we then have another plugin that we need to account for in terms of compatibility and that people will have to decide whether to run it or not. I think we should try to find a compromise on whatever these problems may be.

I liked the ravines idea, if

I liked the ravines idea, if that's the difference in question. If it can be made to work, then I would say, go for it.