Procedurally Generated History: paper by Rhye & Soren Johnson

Camikaze

Administrator
Administrator
Joined
Dec 27, 2008
Messages
27,335
Location
Sydney
As reported earlier this year, Rhye, the creator of Rhye's and Fall of Civilization, along with Soren Johnson, the lead designer of Civ4, have presented a paper entitled "Procedurally Generated History: building a game system through autoplay". The paper was delivered at the 'Foundations of Digital Games' conference in June.

The written paper, also co-authored by Pieter Spronck, has been made available to us, for any interested readers. The abstract is as follows:
Generating interesting game worlds is an ongoing challenge within Procedural Content Generation (PCG). While the use of algorithms for automatically generating content could potentially bring many advantages, the content itself is often random and not well-crafted. In this paper we explore a potential solution to this issue by creating meaning through procedurally tracing a “history” of the game world while constructing it. We introduce a novel approach, called Procedurally Generated History (PGH), which consists of building a game world history through controlled autoplay. The resulting game ecosystem is expected to be engaging and balanced as hand-made content, while still preserving replayability. PGH was implemented in a modpack for Sid Meier’s Civilization IV. Tests with the modpack demonstrated the soundness and balance of the results of the PGH approach.
Download it by viewing this post in the News Updates forum.
 

Attachments

  • Procedurally Generated History.pdf
    1.5 MB · Views: 1,105
Well, if comments are allowed in this thread, let me be the first to say I've read it and I generally get the idea behind it. I'm not sure this can be properly implemented in Civ5, however, I'm very curious to see how this mod plays out, seeing as how little I invested in Civ4.

If the concept catches on, we may be looking at a really big thing if it's implemented in Civ6. It would require SERIOUSLY good AI though
 
I don't know if anybody noticed that but the table qith test results features both HOly Roman Empire and Germany, although the former wasn't in the original RFC. What does this mean?
 
Too brief for my taste.

We've been 'availed' of it as a privilege, so it feels peculiar to analyse its actual content, so instead I'll just direct my commentary unto the community vis a vis the news story.

That conference is quite unnerving. Just the titles alone are eerie , then add up the fact all this research is being pooled into, used to be Microsoft, now it's.... ah. Facebook.
Can't see any dirty hands in this pot, but there's a very disquieting lack of self-awareness into all the research in motivation, engagement, and plausibility. The esa is, at least, naively innocent.
 
I don't know if anybody noticed that but the table qith test results features both Holy Roman Empire and Germany, although the former wasn't in the original RFC. What does this mean?
The table with HRE (table 4) doesn't actually have any entries with Germany, so I assume that it is just a stand-in for Germany included when some pedant who didn't know about which civs are included in RFC pointed out that Germany wasn't a political entity by the date on the table.
 
The attachment isn't working for me; could this have to do with the server migration?
 
The table with HRE (table 4) doesn't actually have any entries with Germany, so I assume that it is just a stand-in for Germany included when some pedant who didn't know about which civs are included in RFC pointed out that Germany wasn't a political entity by the date on the table.
Table 4 actually has Germany in some rows.
 
Top Bottom