Open world Czech Bohemia-inspired orienteering game Prologue: Go Wayback! from PlayerUnknown Productions and PUBG creator Brendan Greene is out now in Steam Early Access – and that’s very inconvenient, because we still don’t have consensus on a nickname. Back in 2017, RPS jocularly renamed PUBG “Plunkbat” – a show of feistiness that surely resulted in no angry emails (genuinely, I don’t know if it did), and which I recently consecrated by writing it on the wall of a random shed in Scotland. We are still massaging our temples about the new game, however.
Obviously we can’t just call it Prologue: Go Wayback! That exclamation mark is some drive-by Yu-Gi-Oh bullshit, and besides, the title’s arguably a tautology and I would like to be able to mention this game to my partner without sounding like I think she doesn’t know what “prologue” means. So here’s our working list of alternatives, mixing in a few other informal RPS titling conventions. Please pick favourites or suggest your own in the comments.
- Proback
- Prologoback
- Prayback
- Progo
- Ploway
- Proooack
- Puwack
- ack!
- Pooo!
- Prolo!
- Plab!
- : !
- Prologue Colon Go Wayback!
- Pro Colon!
- Colon!
- Colon Go!
- Colon Go Way!
- GoW
- Anti-Social Peak
- A Good Walk Rewardingly Ruined
- Gunshy Plunkbat
- Revengeless Liam Neeson Simulator
- Czech Stranding
- Not Generated From Anything Stolen
The last part is a reference to the game’s map generation tools. As disclosed on Steam, the game makes use of proprietary machine learning software to create wilderness maps. Here’s that disclosure in full.
We are developing our own in-house machine-learning model which is trained on publicly available open-source data and vetted to prevent the use of any copyrighted material. This ML model is used to generate base terrain maps in Unreal Engine 5 which are then populated procedurally with custom assets such as the trees, rivers, rocks, and hills you explore. Our artists and designers use our ML model to allow us to generate 64km2 maps that look and feel natural and realistic. The game generates a new map offline on your system every time you play.
It’s one of the more informative such disclosures, responding to the broad concern about genAI using other people’s art and media for their datasets without the creator’s consent. It muddies the waters a bit, however: “machine learning” and “generative AI” aren’t the same thing, as Embark recently explained to us in an interview about robot animations in Arc Raiders.
The disclosure also doesn’t address worries about the emissions cost of ‘training’ generators by having them repeatedly go over datasets to extrapolate patterns – yes, it’ll run offline on your PC, which suggests a lower energy footprint for the user, but it’s not clear how much power was consumed during development. There’s also the question of whether the map generation here contributes to overall normalisation of efforts to automate jobs and shrink team sizes, even if the game’s handling of such technological terrors strikes you as above reproach.
Such caveat! Much proviso! More context than I can easily get into here! And it’s only going to get more complicated. As the developers recap on Steam, this is “the first of three games that lay the foundation for our studio’s ultimate vision: Project Artemis”, which is “a platform for massive, living multiplayer sandboxes where millions of players can shape their own worlds and emergent stories.”
Somewhere inside all of these metaversal convolutions and investor-facing language there is a promising wilderness walking sim. I interviewed Greene and played the game earlier in the year (amongst other things, he told me there are in-game lore reasons for the title). I swam across a river in the dark and it did feel slightly magical.
