Hi people! Apologies for yet another later-than-I'd-like post - here's the May Fossil Sweeper "mid-month" status update!

Over the period that this post covers, I spent most of my Fossil Sweeper time continuing to work on globe asset production workflows. I also explored an alternate presentation for Dig sites.

Dig

During one of my weekly dev streams, I was talking about remaining work in the game after the map is finalised, and when the topic turned to "rare specimen" Dig sites, which offer a higher stakes challenge than the existing "full specimen" Dig sites. In "full specimen" Dig sites, there's one, and only one of each of a skeleton's fossils that have a fancy material and my hope is that by making them uncommon, I can help convey the idea that the odds of finding a complete fossilised skeleton in the real world is also uncommon. You can make a mistake and smash a fossil, but there's a chance it'll be a misc fossil and not one of the skeleton's bones.

"Rare specimen" Dig sites go a little further, and will only yield a fossil if the entire puzzle is completed without mistakes. I want the rewards to feel special and suggest something about a specimen's behaviour or identity, drawing inspiration from things like the Berlin Specimen of Archaeopteryx (which is called the Berlin Specimen because it's in a museum in Berlin - it's from the Altmühltal Formation in Bavaria), the Suncor Nodosaur specimen of Borealopelta (named after the company, not the worker who found it - which is a shame because Shawn Funk is a rad name), or the Fighting Dinosaurs specimen of Protoceratops and Velociraptor (which is comparatively well named, but easily confused with the Duelling Dinosaurs). For the accompanying video, I was going to do paintings of each of those, but drawing some quick scribbles and painting highlights on them ended up looking fun and evocative (and thankfully was much faster to do, since I was already running late with the video).

I'd been thinking about ways to make the Dig sites themselves feel distinctive during play. In another context, I'd be open to considering having "rare specimen" Dig sites be a different type of puzzle (something like a Shakashaka puzzle would be cool), but Fossil Sweeper's primary purpose in existing is to explore the concept of a Minesweeper style game with layers of metaprogression, and any alternate gameplay should be in service of/support the core Minesweeper style puzzles (in the same way the Assembly and Museum phases reflect and incentivise your engagement with the Dig phase's puzzles). A different puzzle would need to fill a role that didn't undermine that core part of Fossil Sweeper's identity. On top of that, Fossil Sweeper has taken a lot longer than I'd originally intended, and if I had planned to do another puzzle type, I'd probably be holding off at this point so that I could cut it later if necessary as a scope reduction effort.

With that in mind, I'd been thinking about whether there were ways I could present regular Dig gameplay that would make it feel novel, and while going over that on stream, I decided to do a quick proof-of-concept experiment with the camera rotated 45 degrees so that I could have an example to talk about of how presentation can make a difference. I'm not 100% settled, but there's a tiny bit of extra cognitive load in recognising patterns that I think makes feel like a different context or a different puzzle where the player's experience and knowledge of the game remains relevant - which is the kind of thing I'm looking for.

Map

Continuing with work mentioned in previous posts, I've been working on revised globe asset production workflows. I wrote a Blender addon to manage automatically switch the references I've sketched depending on how far through the animation timeline the playhead is. In the version of Blender I'm using, animated values in materials don't trigger an update in the viewport unless I enter and exit edit mode. It works fine for renders, but for the purposes of switching references automatically, it's not helpful. After a bunch of trial and error, I was able to get around that in Python by changing the values as part of a frame_change_pre handler.

I also wrapped up preliminary first pass drift animations for the remainder of the globe's landmasses going back to the beginning of the Mesozoic, including making islands like Hawaii and Fiji disappear under the ocean.

I mentioned in the previous post that (during the period that this post covers), I'd added the Black Sea and Caspain sea, which modified the mesh in a way that caused the shape keys to get busted up. The offsets that previously belonged to some other vertices end up applied to the new vertices, causing them to skew off in unexpected directions - in this case, punching through the Earth and out the other side in the Southern hemisphere. I already knew I'd need to redo all my current shape keys, so this doesn't represent anything significant, but it's a nice example of one of the concerns that led me to slow down on globe stuff and rethink how I was doing things.

And that's about it for this post. Thanks as always for reading and for your patience, if you check out test builds and have any feedback, I'd love to hear your thoughts!