next »

[amasinton]

(*** I planned to post this a week ago, but felt it was tone-deaf to do so – with so many more important things demanding our attention.  Black Lives Matter.  The protests have now gained a steeled determinism and are exerting unrelenting pressure.  Real, positive change is perhaps even now finally happening. **)

Content, Content, Content (Part II) - Art
Last week I wrote about some of the shader and code developments from the past couple of months.  I’m focusing on a demo because I feel that, at last, that might be a realistic goal.  There’s a lot of art to produce, though.

Every time I go about reconstructing something from the past, the act of reconstruction opens up my eyes, gives me new questions to explore, and, sometimes, answers.  It has been satisfying discovering the history of this site in the game.  I’m finding things I did not know were there.  This is exactly what I hoped would happen.  It’s what Tolkien wrote about in his most important, and shortest, story: Leaf by Niggle.

Vegetation and Landscapes
The final, fullest extent of the parish church is complete.  Now, I’ve been working backward through its previous phases.  I created the Romanesque church and then the Pre-Conquest building.  I already had the Roman version of the site.  Well, most of it.  I added a stone-lined well based on the one found at Heslington East, where I did a little work once upon a time.  The really interesting thing about the Heslington Roman well is that one of the stones that line it is a re-used finial from a high-status building (maybe a mausoleum) which had once stood nearby.  The fragment would have been visible to users of the well which suggests it was re-used deliberately, meaningfully.  The well itself eventually met the end of its service when it was capped by a fill which included the careful deposition of a large stone, a complete jar, an adult cow, an red deer antler, a young red deer, and a dog.  (Ritualistic “death” of the well?)  Anyway, I always liked that sequence of events and the glimpses at humanity it offers.  So, I thought it would be fun for the player to play through that, too.


That is a section to die for! (I love the stratigraphy in this and I’m not even much of a dirt archaeologist.)  (Image: Illustration by (my former office-partner) the talented Helen Goodchild – from the article published in Internet Archaeology).

The well got me thinking about holy wells.  And that got me thinking about the “deep time” of the site.  The well stands as a metaphor for the through-line of the site’s use – a deep shaft cutting through all of the levels of the site’s human occupation.   So, I headed off to the Mesolithic which is the temporal “bedrock” of the game.  You don’t go further back than this. 

In Archangel’s Mesolithic period – and, I suspect, long before that – the well was a spring.  A spring means vegetation: reeds, grass, bushes, and trees.  This was something I have been avoiding creating because tree modeling scares me – I’ve never done it and it seems impossibly complex.  Fortunately, quite a while ago I had experimented with a foliage shader based on techniques used in The Witness.  I had made some simple bushes, and, personally, I like them a lot.  I adapted the bushes to become the leaves of trees.  Birch trees, to be exact.  Why?  I saw one once, 11,000 years old, felled by beavers, lying fresh as the day it fell, preserved by peat, uncovered by my former colleagues at the Mesolithic superstar site Star Carr. 


A birch trunk from Star Carr – felled about 11,000 years ago. (Image: Nicky Milner, Chantal Conneller, Barry Taylor, University of York Department of Archaeology)

So, when modeling the Mesolithic in the game, of course, I discovered that the spring fed a nearby stand of birches.  Turns out, they’re not too hard to model.  Using one base tree I soon had five individuals, a sapling, plus a dead trunk.  From these, I grew a little forest.  The spring was now a little pond.  In addition to the trees, it was surrounded by the bushes I had already made.  I then created some simple cross-plane mesh reeds and cattails to stand in the shallows of the spring.  All along the tiny stream that ran down from the spring was a scattering of tufts of mesh grass.  I also discovered that the spring was just below the crown of a small hill that was naturally (magically?) empty of trees.  It was a high point visible from most places on the island (only, the island wasn’t and island in the Mesolithic).  This was attractive to Archangel’s Mesolithic people, who practiced “sky burial” and so built their funeral scaffolds atop the hill, near the (obviously sacred) spring.  The cemetery was marked out by a few naturally placed boulders and a set of four large posts to mark time.  Down the hill toward the spring, they responded to the sacredness of the site by setting up a tall monument with many faces, carved from wood and painted.  (I based this on the Mesolithic Shigir Idol.)  Whatever the birds left after the sky burials (usually skulls and long bones) were gathered up and placed with their ancestors in some of the small caves nearby.


Sky burials, tiny skeletons (we’ll talk about that in next week’s post)

Now that I had a good idea of the site at its “beginning” the periods between then and the collapse of the parish church tower some 115 centuries later started to emerge in my imagination.  As time wore on at Archangel, the site retained its cemetery associations.  Eventually, as Archangel was settled, the settlers built their own, artificial burial cave on the hilltop – a round barrow.  The holy spring continued to flow.  Much later, the settlers created two concentric earth rings around the hilltop with its barrow and spring.  The hillfort was small, but large enough for the very small, now island, community.  The spring was lined with timber and earth was banked up around its edges, creating the first well.  Burials continued, but now they were inhumations.  The barrow was robbed for building stone.  Later, animal sacrifices were made, thrown into the well, and the well “sealed”.  For a while it slept.


The Neolithic barrow on its own on the hilltop.

The years passed, the settlement Romanized.  The hillfort was abandoned as the settlement began to drift toward the sea and a fishing way-of-life.  But, the inhabitants still returned to the hilltop for their final resting place, the wealthier among them setting up small temple-tombs.  As the site became crowded, the hillfort ramparts were filled in, creating more space at the top of the hill.  The well began to bubble to the surface again.  Some of the temple tombs were dismantled, their stones used to re-line the old well.  Offerings to the spirit of the spring/well resumed.


That little Roman mausoleum.

Years passed.  A new message arrived on Archangel.  Some of the islanders sought to be initiated into the new rites, their new understanding of divinity and humanity.  Perhaps it was a miracle that made the waters of the old hilltop well sweet again.  Perhaps it was centuries of slow purification.  Its waters baptized the islanders.  And they set up a stone cross on the hilltop near the well to mark the place. 

Later, they built a wooden church.  Then quickly replaced it with a stone one.  And that begins the building the ruins of which, centuries later, the player reconstructs.


The first stone church – with the older, Roman mausoleum nearby (perhaps the tomb of a proto-saint?), the stone-lined well/spring is on the left.  The foundations of the church over later centuries is shown, too.  Time is weird, I told you I still don’t quite understand it in this game.

Oof, though
That’s a lot of modelling to do, and much of it is pretty tedious.  Three versions of the church were time-consuming enough, but the landscape itself has proved to be the most time-intensive.  I’ve set myself a perhaps too ambitious task of showing the landscape like an archaeological site.  When sectioned, the player should be able to see that there are holes and ditches that have been filled in by later changes.  It turns out that this is hard to do – or, at least, tedious, requiring sculpting and then dodgy Boolean operations, eventually requiring hours of hand-stitching polygons together.  And then there’s the layer of topsoil over that.  Oof!  I think I have to come up with a better – or at least quicker – way of building landscapes.  We’ll see how what I’ve got now works for the demo.


The whole hilltop, with its different epochs all layered over each other and the church foundation for reference.


An exploded view, showing each occupation period separately.

What has been very enjoyable, though, is learning how to model the natural environment – not just the individual trees and plants, but scattering them across the landscape, in and around the human-created features.  The process uses Cinema 4D’s hair and cloning tools to paint trees, vegetation, and rocks onto the surface.  I can randomize these based on my small pool of originals.  Best of all, I can export all of this in a very simple text file for Unity where the prefab placement tool I discussed in last week’s post populates the game landscape.  This has been very satisfying.

Anyway, the process of discovery, of learning that the real anchor of the site was not the church, but the well/spring, has led to some really exciting developments in the game’s design, which will be the subject of next week’s devlog.

Twitch streaming
Also, I’ve been dipping my toe into the live-streaming waters.  I’ve noticed some devs do this – just broadcast themselves while they do level design, just chatting.  I thought I would give this a try because … why not?  But, I do all of my work in the early morning, so I think very few people will watch.

If you’d like something perhaps a little dull to lull you back to sleep, then look me up on Twitch.  I’m “amasinton”.

[JobLeonard]

Heh, I guess that since you incorporate the time element into your game, it's like you add an extra dimension to how many art assets you need. Which can grow quite explosively (I have been thinking of Big O behavior of algorithms at work today, which is probably why that is my first association)

The well got me thinking about holy wells.  And that got me thinking about the “deep time” of the site.  The well stands as a metaphor for the through-line of the site’s use – a deep shaft cutting through all of the levels of the site’s human occupation.   So, I headed off to the Mesolithic which is the temporal “bedrock” of the game.  You don’t go further back than this.

Unrelated to the game, but in Religion for Atheists Alain de Botton argues that maybe we should have secular temples to remind us of the fast that there is a greater universe out there, to teach us to "get over ourselves" basically. One suggestion (again, IIRC) was the temple of deep time, where you would stand at the bottom of a long shaft looking upwards through many geological layers, arranged in a way to overwhelm us with a sense of insignificance similar to how cathedrals are supposed to impress us when we look up

This comment reminded me of that

[amasinton]

Heh, I guess that since you incorporate the time element into your game, it's like you add an extra dimension to how many art assets you need. Which can grow quite explosively (I have been thinking of Big O behavior of algorithms at work today, which is probably why that is my first association)

And now I have a name for the feeling I get when I consider the work involved every time I want to add a new time period.  I'll remind myself I am experiencing that "Big O Feeling" and perhaps think again!

I've always been aware that each time period at least doubles the assets I need, but the cumulative effect of layering up the remains of many different times in the same place does exactly what you say - adds another dimension to as well as explosive growth in the number of assets needed.

In next week's post - the final part of this massive 3-parter - I talk about some strategies to limit this growth of assets.  But they require me to make hard decisions about the overall design of and my goals for the game.  If I want to finish making the game, I need to restrict the number assets, and to do that I have to restrict the player's spatial and temporal freedom in various ways.  I think I'm okay with that, though!

... temple of deep time, where you would stand at the bottom of a long shaft looking upwards through many geological layers, arranged in a way to overwhelm us with a sense of insignificance similar to how cathedrals are supposed to impress us when we look up

I like that!  The island in the game is in itself a kind of Temple of Deep Time.  There's the well I was talking about, but the whole island is this way.  If the player gets a feeling that the ways things are now is NOT the way they have always been and that it is NOT the way it will always be - that would be a huge success.  Change is constant and that's okay.  Maybe we need this sense now more than ever ...

Thanks for the insights!
 

[amasinton]

Content, Content, Content (Part III) – Design


My first tree sitting beside the tower ruins.  Doesn’t have much to do with the rest of the post, I just thought it was pretty.

Welcome to the final part of this three part mega-update.

Losing human scale
One thing that happens in Archaeology a lot is that we lose the people.  Sites with human skeletal remains might suffer from this a little less – although if you read the skeletal reports the humanity doesn’t always shine through.  It’s all too easy to see pot sherds and postholes as phenomena in their own right, without the human agents who fashioned and discarded them.  This is unsurprising given that Archaeology is the study of material culture.  Still, from time to time, you do catch glimpses of real people.  It’s surprising where you find these moments.  They’re usually in the little things.  My former colleague Penny Spikins excavated a Mesolithic flint-knapping site on March Hill, West Yorkshire.  Pretty dull stuff.  Except that the amount of detail she was able to tease out of the soil allowed her to get down to “bum prints” – the places where the knappers sat to carry out their work.  Suddenly, you could catch a glimpse of a single moment in time almost six thousand years ago as a small group of people sat down to work on a project together, doing something “normal”.  I love archaeology moments like those. 

I think about how an Anglo-Saxon silver ring one of the students found on the excavation in Burdale got lost in the tiny settlement there.  And how long the owner searched for it.  How old they were.


This ring was lost for 1400 years. (Source: Julian D Richards, Steve Roskams (2013) Burdale: an Anglian settlement in the Yorkshire Wolds [data-set]. York: Archaeology Data Service [distributor]https://doi.org/10.5284/1021540)


The trench at Burdale, 2007. I’m in this picture. A million internet points if you can spot me.
(Source: Photo by Ben Gourley in Julian D Richards, Steve Roskams (2013) Burdale: an Anglian settlement in the Yorkshire Wolds [data-set]. York: Archaeology Data Service [distributor]https://doi.org/10.5284/1021540)

The Resurrection has a human scale problem.  Its focus is pulled-back, almost isometric, looking at buildings and landscapes.  Big things.  At this scale, the people get lost.  I knew this was an issue when I started creating the “cutscenes” I’ve written about in earlier posts.  The issue was reinforced when I made the “sky burial” platforms for the Mesolithic period.  The skeletons on the platforms almost disappear.  And when you zoom close enough to see them, they lose their context with the rest of the site.  I’ve even cheated a little bit in that I’ve scaled up the skeletons to between 1.2 and 1.5 times actual size.  It helps, but not as much as I’d like.

I think this is a design issue.  At least, an art design issue.  The best way to solve it may be to continue to fudge the scale – making the humans much bigger than they would be in reality, just to keep them within the visual context of their sites.  Humans figures are such powerful storytellers that they can only be used sparingly in a game like this.  And so, perhaps, their out-of-scale size will not be consciously noticeable.  Medieval and Renaissance artists always manipulated scale.  They would scale down the horizontal and scale up the vertical, giving England and the Netherlands mountain ranges where before they barely had hills.


The Z-stretched landscapes of Pieter Bruegel the Elder. See how they swallow up human scale? (Which is kind of the point here.) Also, this is not Egypt, which Bruegel had never seen; this is the Netherlands with the height scaled up.The Flight Into Egypt, 1563, oil on panel. (Source: Wikimedia Commons)

Finding the “third act”
The other design … opportunity? … all of this modelling of the past has revealed is how I might include something I’ve really wanted for a long time: a “third act twist”.  It’s more of a third act twist in mechanics, though.  I think it will make the gameplay deeper and the player’s investment more … meaningful.  I’ve needed a story hook to hang it on – one that didn’t require the player to swallow even more unjustified magical thinking than I’m already asking.  The modelling and the discovery of the importance of the spring/well I talked about in Part II has provided that hook.  The hook has provided a mechanic. 

Right now, I’m hesitant to talk about it in too much detail.  There’s a moment at the end of the 2008 Prince of Persia which has stuck with me.  It’s kind of like that.  The point here is that it was the process of modeling the site’s past that has revealed a way forward in the design of the gameplay.  This always happens when I’m modelling a real site’s past incarnations, too.  Modeling, like drawing, turns out to be a really useful way of coming to a better understanding of a place, and of seeing things you normally would have missed.


SPOILERS for Prince of Persia 2008. Only watch this if you’ve already played through PoP 2008 or you want to spoil the very best part – which you really shouldn’t do. (Source:

)

Adrift on a Sea of Content
On a more practical design note, there is potential for a lot of content required for this game.  For every period in a site’s life, there’s the rest of the island to model, because the player is not restrained to a single place when jumping through time. 

But, maybe they should be. 

Considering how long it has taken me to get only this far (although I’ve been doing far more than just modeling), creating dozens of versions of the island seems foolish (I talked about this briefly last time).  There might be some procedural options to ease the burden.  But, one thing that has recently occurred to me is that I might simply limit a player’s scope for some sites and some periods, thereby limiting the scope of the content as a whole. 

The things the player unearths unlock other times periods.  I’ve always intended some artifacts to function as keys and some to be mere decoration.  The in-game reason for this is that some artifacts simply don’t permit time travel because their “time signature” is too weak.  It would not be too much of a stretch to make the signature of some artifacts strong, but unstable, so that they only permitted time travel in a specific area, or for a certain amount of play time.  Beyond just a hack to relieve me of the duty of modeling the island over and over again, there are some interesting gameplay options which arise from this design decision.  Artifacts and architectural fragments could be collected into sets which, when complete, create a sufficient signature, for example.  The part that artifacts and architectural fragments play – and therefore the incentive for exploration – becomes more compelling.  It also gives a kind of natural hierarchy of sites, too.  Some are “common” and their impact on driving the story of the island limited.  Other sites have deep histories and the player will understand that their actions in these places are important for the story of the island overall.

Epilogue
In the days (weeks?) since first writing this I’ve been thinking a lot about the benefits of limiting the player’s movements.  I think it’s a good idea.  An open world concept does not really serve what the game is trying to express.  Limits also help justify the “diorama” visual feel – which I have had trouble justifying for quite a while now.  I think tossing out the total open world design principle will prove to be helpful creatively, as well as helpful practically.  And I think the gameplay may be a little more interesting for the player as a result.  Win, win, win!

Thanks for reading!

Happy Father’s Day!

[oahda]

Amazing posts! Really interesting updates on the game alongside really interesting information in general. Love learning about these things. Seeing the layers at large is so cool, but I like that you're also remembering to consider the individuals once occupying those places.

[amasinton]

Amazing posts! Really interesting updates on the game alongside really interesting information in general. Love learning about these things. Seeing the layers at large is so cool, but I like that you're also remembering to consider the individuals once occupying those places.

Thank you for your encouragement!

That tension between the landscape and the human scales is proving to be a real challenge to resolve.  I think that's part of the fun in developing the game, though - regardless of how it turns out.

[JobLeonard]

Considering how long it has taken me to get only this far (although I’ve been doing far more than just modeling), creating dozens of versions of the island seems foolish (I talked about this briefly last time).  There might be some procedural options to ease the burden.  But, one thing that has recently occurred to me is that I might simply limit a player’s scope for some sites and some periods, thereby limiting the scope of the content as a whole. 

Just make the Player Character an archeology undergrad who doesn't have full access to the site

[amasinton]

Just make the Player Character an archeology undergrad who doesn't have full access to the site

Brilliant! And realistic.  

[ChrisLSound]

Don't have much to say other than all these updates have been fantastic and I'm always excited to see what you have to say/show. It's interesting seeing things get more refined and focused. Great work!

[amasinton]

Don't have much to say other than all these updates have been fantastic and I'm always excited to see what you have to say/show. It's interesting seeing things get more refined and focused. Great work!

Thank you, Chris! 

I'm inching forward toward a demo - all of the work has been 3d modeling for the past couple of weeks now.  I'm eager to get it all into the game and to start bug-fixing (really!).

[amasinton]

July Update

I wrote this two weeks ago, but then I had to figure out time travel again ...

Time for the monthly update!  In this edition I try out Addressables, get back into charts, speculate on future archaeology, and report on progress in content creation.  Also, I’ve been double-timing because I really do have to finish a project for a client, so this game hasn’t quite been getting all of the attention it’s accustomed to.  Nevertheless, the march of progress here, though glacial, goes ever onward.

Underwhelming Addressables
So, I built a little Addressables demo and the results are that I don’t understand addressables.  The demo loads cubes into the cells of a 10 x 10 grid based on the position of the camera.  That part was fun to code and I’ll definitely use it.  This system divides the world into numbered cells 1 – 100 with 0 in the lower left corner.  It knows which cell the camera is in and loads cubes into the surrounding eight adjacent cells.  It can load in the 15 surrounding that, and the 21 surrounding those, etc, depending on the camera’s distance to the cell at the center.  It also figures out which cells need to be unloaded, which kept, and which loaded fresh.  It’s a neat system and was fun to figure out.  But the Addressables system is what loads and unloads each cell from an array of prefabs and this is really slow.  Like, you can watch the cubes pop in and out of existence one at a time.  I was expecting the loading/unloading to be a lot faster.  Regular instancing/destroying runs much, much faster (at the cost of garbage allocations).  But maybe speed is not the point of Addressables?  Maybe it’s the efficient loading and unloading of data from memory?  I think I don’t understand the system very well.  It’s not very well documented yet, either as it’s one of the many, many potentially wonderful systems in Unity that are in “preview” – and have been for a couple of years.




It works?  (Predictably, there is a problem with my logic.)

A Map of Time
Charts again!  In an effort to keep track of the paths through the demo I made a chart.  I have no idea how to map this out because time AND space are important here.  It was a helpful exercise, though, as it solidified for me how the game is shaping up and the key tools for storytelling.  The chart helped me to see the structure of the game as a dungeon where the different time periods are the rooms and the artifacts are the keys.


An illegible, indecipherable chart! Click the image for a giant-sized version of the image you can actually read (for what it’s worth).

Dungeons tell stories according to the order in which rooms are unlocked.  Others have made useful attempts at charting out dungeon structures.  Mark Brown’s Boss Keys series charts every dungeon in every Zelda game (and now, other games, too) as a function of choice, time, and resources.  Nathan Savant has a series of articles over on Gamasutra which, in part, charts game narratives as functions of time and space. 

I have tried to do something similar to help me get my head round the structure of The Resurrection.  While the chart looks like a linear series of events, it also has an implicit web-like structure with time periods as the nodes and artifacts providing the connections in the network.  That’s messy, and it’s hard to shape the player’s experience.  The most direct way, of course, is to control how and when the player has access to keys.  The difficulty is that, even with just a few keys scattered across nodes, the player quickly builds up a collection which breaks the linearity of the playthrough.  I compensate for this, or brake its momentum, with the rule that a player can only use the keys in their inventory that lie outside their current time period.  (There are some related rules, but this is the main idea.)
But, there’s a hitch …

There are TWO time frames!

That’s right, it’s relativity.  There are two time frames running concurrently through the game.  There might be more, in fact.  The two reference frames are the chronology, or history, of the game world and the player’s own playthrough.  The player’s time is linear and happens in realtime.  It’s simply the time they spend playing through the game.  The game chronology is also linear, but from the player’s perspective, is non-linear, circular, and sometimes branching because the player can jump from period to period using the keys they find – and because their actions change the chronology.




This video by minutephysics is magnificent and also to blame for my time travel graphs.

I’m hopelessly lost.

Let’s make a game to help me understand the game
While writing this blog post I realized that I didn’t really understand how the rules in my own system impacted how the player could move through the game’s chronology, and therefore, I didn’t understand how the narrative tools I’m building actually work.  So, I stopped writing this and built a new project in Unity which models the rules of time travel in The Resurrection.

… two weeks later …





This simulation is just a grid of buttons – each one representing one of the artifacts/keys from the demo.  Pressing one of these takes you to the time period when it was created.  The keys have a toggle which indicates if that key is in the player’s inventory.  It also has a timeline of dates.  Finally, it has a button which allows the player to switch between ends of their time travel wormhole.  The keys and the dates in the timeline turn on or off depending on the time travel rules.  Now I can see which keys are in a player’s inventory at any point in their playthrough, which time periods are accessible by the player at any given point in their playthrough based on the keys they have collected, and I can see when players collect keys for the first time.

For a long time I have been worried about the potential ratcheting-effect of these rules because the further back in time the player went, the harder it would be to travel forward in time because the player would not be able to use artifacts from future time periods.  Now that I can see it in simulation, that’s, of course, entirely NOT how the rules work.  There IS a minor ratcheting effect if I want a player to visit a specific period at a specific time in their playthrough, but I think I have come up with a workaround.  Otherwise, the rules generally leave the player with a variety of options for periods to visit most of the time while also leaving me with the ability to direct their playthrough with the availability of keys.

Okay, fixed time travel (again).  What’s next?

Future Archaeology
All of this time travel stuff is great, but it’s a sideshow to the main work for most of the past month: creating models for each period.  I’m almost done with the past, but am finding myself stuck on what to do about the periods that are in our own future?  Look how much the world has changed over the past three centuries.  People’s needs haven’t changed, but the way in which they satisfy all of those needs have changed quite a bit and have left very distinct traces in the archaeological record.  What traces will there be 300 years from now? 

I had thought about putting in some wrecked cars but then I realized that cars, as we know them, probably won’t be around even in a hundred years.  We’ll still need to transport stuff, but I wonder how much of this will be by things we would recognize as automobiles?  Will transportation even need vehicles of any kind?  And if that’s the case, what kind of record does that leave?  Either the world ends buried in garbage with lots of material for future archaeologists to work with, or garbage is a passing period of cultural development and humanity will have moved onto other ways of accomplishing its basic, unchanging goals.  300 years is unimaginable.


A Romano-British temple based on one at Pagan’s Hill, Somerset. Maybe it would be future archaeology if I put some neon tubes and advertising written in kanji all over it?

I find myself confronted with the well-known fact that speculation about the future has nothing to do with the future but everything to do with our concerns about the present.  Maybe I just roll with it and put in ruined future cars with wheels or hover engines or whatever and move on.  Maybe I’m overthinking it.

So Many Content
The parish church scene now sits within a landscape of 20 x 20m tiles in a 5 x 4 grid.  Each tile occupies 11,500 years of time scattered across five key periods and three or for sub-periods here and there.  That’s a ton of content in a small area.  It’s a challenge to fill all of that space and time.  Especially the landscape itself.  But I’m refining some workflows which have helped speed up that process, and I’m learning to not sweat the details, especially the accuracy.  Also, not every place has to be a treasury of wonders at all times.  It is a long process, though.

Sketching the Island
On the macro-level, I’ve experimented with a workflow for sketching the island as a whole, too.  I’ve had some encouraging first results using Sebastian Lague’s  Hydraulic Erosion tool.  I can sculpt a basic landmass in C4D, export it as a heightmap, import that into Sebastian’s system as a grid of values, which are then worn down by virtual water, then export it back to C4D for further work.  I like it!  It works for me for now.


Archangel Island c7000BC with hydraulic erosion applied. (Inset: the heightmap.)

Conclusion
So, that’s actually quite a lot – more than I thought I had to report.  I can see the demo moving nearer and nearer to reality.  Still a lot of work left, of course.  But, given that I only get a few hours a week to work on it, it’s going pretty well.

[JobLeonard]

Oh yeah, with complex systems you want low-fi prototypes as early as possible to untangle them.

Otherwise, the rules generally leave the player with a variety of options for periods to visit most of the time while also leaving me with the ability to direct their playthrough with the availability of keys.

I mean, I basically figured that's what you were going to do, no? Design a kind of "metroidvania" backtracking system but with time-travel where areas are unlocked as puzzles get solved

[Sp1ke]

Very original concept, but I think it's originality may be also a trap for playability.
Good luck

[amasinton]

Oh yeah, with complex systems you want low-fi prototypes as early as possible to untangle them.

NOW you tell me!

Yeah, I should have done that right at the beginning.    I didn't realize I was making a complex system.  I thought the consequences of my rules of time travel were pretty straightforward.

I mean, I basically figured that's what you were going to do, no? Design a kind of "metroidvania" backtracking system but with time-travel where areas are unlocked as puzzles get solved

You saw this WAY before me!  I saw it first in terms of rebuilding ruins, then in terms of time, and it didn't occur to me that time really can act just like space in the sense of time periods acting like rooms in a dungeon (which I really should have known all along because that's how historians/archaeologists work - by locking themselves in arbitrary rooms of the time dungeon).  But, yeah, I've always had this kind of experience in mind - I just didn't see it as a "Zelda-like" or "metroidvania" until now.  Sheesh - I'm slow!

Also, now I have a genre to tag this to, right?  And that's good because I really had no idea how to elevator pitch this!

[amasinton]

Very original concept, but I think it's originality may be also a trap for playability.
Good luck

You know, I've been thinking about that all along and it's a genuine concern.  I think you put it very succinctly.

The originality and obvious trouble I'm having coming to grips with just what this things is also give me concerns.  Neat idea, but how compelling is it to play?  Is it a toy or is it something that piques curiosity in the player so that they come back to it?  We'll see ... 

I have very modest goals in terms of number of people who ever play this, at least - so the bar for "success" is pretty low.

Thanks for commenting on this aspect.  It's genuinely helpful!

[JobLeonard]

If it makes you feel better, I didn't fully see it until I put it into words in the last message either - I kind of felt it as you introduced new ideas in the topic. Never mentioned it because my reasoning was that you had a lot more ideas than you were sharing with us and that you were still working those out.

BTW, "metroidvania" implies a very specific platformer-with-combat style game, if you are going to refer to it as a shorthand be sure to emphasize that only the "unlockable backtracking" part is covered.

I think what might be more interesting is to think of how good/bad that "unlock areas with insights gained from solving puzzles" maps to real-world archaeology where answering questions unlocks more questions - you're the only one here with enough experience in the field to really tell us I think!

Neat idea, but how compelling is it to play?  Is it a toy or is it something that piques curiosity in the player so that they come back to it?  We'll see ... 

To me the premise of the game is very narrative driven, where as you play the game you become more and more invested in the history of the location. I think that can be very compelling on its own

Honestly, maybe it's better to keep the puzzle elements "light" and lean into the walking simulator genre a bit there - reconstructing the historical fiction-based-on-nonfiction is probably compelling enough on its own!

[oahda]

Hooray, an update!

While they are also very different, there are so many parallels with what I'm doing with my game right now, haha. Will see if I can cover it in a smaller update back in my own thread. Also fun to see you referencing all these videos that I am also familiar with.

I feel like the metroidvania description made something click for me too. Makes things clearer, and I like it! The game might be a bit niche as Sp1ke pointed out but if people play Dwarf Fortress and flight simulators, I can't really see much of an issue finding an audience for this too, especially with such a gorgeous presentation. It seems very appealing to me, at least!

I had either not realised until now, or forgotten if I ever knew, that there is travel into the far future as well. Hope you're able to come up with some fun speculation and solve the struggles you described. Seems like a complex game to deal with, but I think it might be worth it!

[amasinton]

If it makes you feel better, I didn't fully see it until I put it into words in the last message either - I kind of felt it as you introduced new ideas in the topic. Never mentioned it because my reasoning was that you had a lot more ideas than you were sharing with us and that you were still working those out.

Thanks, this is really encouraging. I've had some core ideas at the start, but others have only revealed themselves as I've gone through development. And there are a few I still have in reserve, cooking away. We'll see what makes it into the final dish.

BTW, "metroidvania" implies a very specific platformer-with-combat style game, if you are going to refer to it as a shorthand be sure to emphasize that only the "unlockable backtracking" part is covered

Yes, the backtracking part - after unlocking a new artifact, in this case - is appropriate. Platforming, not so much! 

I think what might be more interesting is to think of how good/bad that "unlock areas with insights gained from solving puzzles" maps to real-world archaeology where answering questions unlocks more questions - you're the only one here with enough experience in the field to really tell us I think!

So, here I think you've hit on another, deeper idea: archaeology as puzzle solving. There is a lot of puzzle solving in archaeology, but I don't think any archaeologist expects to actually "solve" any puzzle (unless you're re-fitting a flint core or piecing together a pot, maybe). The big puzzle: "What happened here?" I don't think you ever solve in the classic sense of finding the one correct solution. In games, though, that's often what you do - find the correct solution - it's also what goes on in mystery and crime fiction, too - and what they often try to do in true crime. I think because it's very satisfying. But in this game, I've tried to embrace the idea of no "right" solution. It feels more true to archaeology (which is satisfying for me, but maybe me only), and - what's more important - hopefully ends up being satisfying for the player because the right solution is the one the player pieces together in their own mind. Everyone's answer to "What happened here?" will be a little bit different - and that's the successful solution.

How many times have you been let down by the end of a TV series, or trilogy of movies, or books, because the final reveal - the conclusion that the author(s) present - is never as satisfying as what you had but up in your own mind?

... where as you play the game you become more and more invested in the history of the location. I think that can be very compelling on its own

Yes! Exactly. You're writing the history of the location as you play. The history doesn't actually exist until you create it while you rebuild the island. I think that has the potential to really build a personal attachment between the player and the game. I think it's also a really hard trick to accomplish on the developer's part - but that's the goal.

Honestly, maybe it's better to keep the puzzle elements "light" and lean into the walking simulator genre a bit there - reconstructing the historical fiction-based-on-nonfiction is probably compelling enough on its own!

I agree! The puzzle elements, such as they are, are mainly in how you traverse time with the artifacts (the keys) you have. You may not even know you're working out a puzzle, in fact, and that's okay. The idea is not to 100% this game - its to tell a story - so it's okay if you don't puzzle through how to unlock every time period in every location!

So much good stuff here - thank you for this!

[JobLeonard]

So, here I think you've hit on another, deeper idea: archaeology as puzzle solving. There is a lot of puzzle solving in archaeology, but I don't think any archaeologist expects to actually "solve" any puzzle (unless you're re-fitting a flint core or piecing together a pot, maybe). The big puzzle: "What happened here?" I don't think you ever solve in the classic sense of finding the one correct solution. In games, though, that's often what you do - find the correct solution - it's also what goes on in mystery and crime fiction, too - and what they often try to do in true crime. I think because it's very satisfying. But in this game, I've tried to embrace the idea of no "right" solution. It feels more true to archaeology (which is satisfying for me, but maybe me only), and - what's more important - hopefully ends up being satisfying for the player because the right solution is the one the player pieces together in their own mind. Everyone's answer to "What happened here?" will be a little bit different - and that's the successful solution.

This reminds me so much of the process I went through when I switched to studying art/design after dropping out of physics. Physics is based on the idea that there is one external objective truth about how the universe works that we can get closer approximations of by better measurements and building better models. The result is that it is accepted in the physics community that eventually our understanding of the universe will converge on one universal theory of everything. Meanwhile art is all about acknowledging that no two subjective experiences are exactly the same, and that knowledge and interpretation is contextual.

I guess archaeology sits somewhere in the middle: when it comes to historical events, something has objectively happened at one point. We can get closer reconstructing to it via contextual knowledge, but must accept that there will be many gaps and different possible paths that we cannot exclude with certainty.

Interesting how this makes your game a complete contrast with, say, Return of the Obra Din, which is a deduction game where One True Answer will come up via a process of elimination.

How many times have you been let down by the end of a TV series, or trilogy of movies, or books, because the final reveal - the conclusion that the author(s) present - is never as satisfying as what you had but up in your own mind?

So basically, your intent is to craft this game in such a way that people aren't just discussing what the open ending means, but what is the ending to begin with, right?

I agree! The puzzle elements, such as they are, are mainly in how you traverse time with the artifacts (the keys) you have. You may not even know you're working out a puzzle, in fact, and that's okay. The idea is not to 100% this game - its to tell a story - so it's okay if you don't puzzle through how to unlock every time period in every location!

It might be interesting to approach this from a reverse point of view: imagine a "completionist" type gamer playing this game. How would their typical playstyle "ruin" your game and the intention of it for them, and what could (or should) be done about it? Do you want to nudge them to frame the game differently?

I say this because I'm also sort-of a completionist: I want to explore every possible story thread that a game has to offer. It's not the same as getting 100% on a game but still quite relevant to this game I'd say

[ChrisLSound]

Gosh, so much cool stuff here. The pieces are really starting to come together for me, to where I think I have a good idea of the feel you're going for.

Interestingly, "metroidvania" has recently been super helpful in designing my own (very early, side-project) game - using it as a sort of model/philosophy to think about not-typically-metroidvania mechanics. The links in your posts have also made me realize I could probably learn a lot from Outer Wilds.

that's how historians/archaeologists work - by locking themselves in arbitrary rooms of the time dungeon

Now that is a cool pitch!

I have very modest goals in terms of number of people who ever play this, at least - so the bar for "success" is pretty low.

FWIW, lots of my favorite games I've played recently have been ones like this. Tightly-focused, highly-curious passion projects with low stakes in terms of dev cost. I think it's rad as hell to just explore ideas like this!