Dark Dice Video Game: Post Mortem (Patreon)

Greetings Adventurers,I'm here to report on a top-secret project - a video game we were working on called Realmweaver. I hope to briefly outline the project, what went wrong, and give some insights into what came out of this experience. This is, ultimately, a post mortem report, as you will discover.

In 2019 we were approached by a team of indie video game creators (with substantial financial backing) who were interested in creating a Dark Dice video game. We were asked to be the creative directors: overseeing artists, musicians, writing, game mechanics, and ensuring the overall tone of the project while the programming director managed his team, and the project manager kept all of us on task, on schedule, and on budget. Since this was a fresh collaboration between multiple video game creators (some of which had previously worked on titles for Ubisoft, Blizzard, and New Blood Interactive) and us at Fool & Scholar Productions, we formed a new studio called Lost Outpost Studios.

The goal of the company was simple:Create a "vertical slice" (a playable demo to encapsulate the experience of playing a larger game) in the hopes of pitching it to a publisher. The publisher would fund the team necessary to build the full game. All of the main folks involved (and many of our collaborators) had industry contacts that would entertain such a pitch, and "all we needed to do" was make a playable demo that encompassed the elements we wanted to include and showcase.

Game Concept. The concept of Realmweaver was a tactical RPG, where players could recruit unique party members, explore a new world while adventuring further from the safety of civilizations, and undertake dangerous adventures, while experiencing a deep, harrowing story. We aimed for an interactive realm, where players could opt to leave party members to grow and defend/build remote settlements, gaining access to more options (like shops or healing) upon their return. For example, leaving your bard in the village of Ilmater's Hope could earn passive income every time you visit, or a leaving a huntsman could create a unique leather/monster goods shop. So the gameplay was a mix of resource management, RPG, and (to a very small degree) city-builder.

Classes. Characters largely revolved around pre-defined classes, which we mapped skill trees and abilities for. Each class had 3-4 "specialist classes" they could upgrade into, similarly to the Borderlands 3 talent trees - allowing character to mix and match, or go all in to a single specialization. Each path would offer combat and non-combat skills as well. For example: Scholars could become an Alchemist (a healing physician who can also create alchemical attacks - applying elemental effects to weapons), Occultist (part archaeologist, part biologist - dissecting dead monsters as a learned treasure hunter) or a Wizard (learning highly specific incantations which result in predictable magical outcomes). All characters remain within their specific "class," and some classes could only be recruited in certain regions of the world.

We finished art and basic skill trees for every one of the core classes and specializations, which included Commoner, Faithful, Fighter, Outlaw, Ruffian, Scholar, and Realmweaver (a class unique to the hero). Notably, some classes had skills better suited for non-combat situations, such as the Porter or Merchant (both of which are Commoner specializations), which boosts party morale, reduces their use of resources during travel/camping, etc.

Each class also had a unique visual identity so they could easily be identified in "spite form" at a glance - because the game was intended visually isometric, blending 2D character art with 3D environments. We programmed all animations to allow for customization of clothing, but eventually simplified enemy design choices to ensure that their "class" would be clearer to players at a glance. That way a wizard doesn't suddenly hit you with advanced fighter abilities. We also changed our combat system away from a tactical RPG akin to D&D4e (or Final Fantasy Tactics) into an Encounter System that was less combat focused.

Story. You play as the Realmweaver, a human weaver from Eastwood cursed with the inability to die. The sole survivor of a supernatural disaster in your hometown, you travel the world with an every-growing party, working to "weave" the fractured human realms together so they can survive the surrounding horrors. You battle literal monsters, nature, giants, elves, undead, and humans (religious factions, local lords who see a unified human kingdoms as a threat to their personal autonomy, etc.), while protecting new locations as you travel the realms. A portion of the plot also revolves around the legitimacy of your authority, "given" to you by noble factions, heirs, and similar groups of authority. Will you lead humanity toward peace, or conquer your rivals through violence "for the good of all?"

Demo Story. The story for our demo focused on Act II of the game, where you are already well-respected as the Realmweaver and are traveling through the Bright Vale toward a larger quest. Your party consists of the Realmweaver (you), Ainsley the Bard (who you're paying to be there, and viciously mocks any unselfish choice), Sir Blane Ozmark (a noble who is endlessly devoted to your mission but does not view "the poor" as people), and Wick Whittaker (a dwarven snake-oil-selling healer and barber-surgeon who is effectively on a prison "work release program"). You can also recruit a monster hunter named Soren. Your team is constantly bickering, but when blades are drawn they are dependable.

In the demo you learn about the missing residents of Ilmater's Hope (a few hundred years before the Dark Dice podcast takes place), and uncover a kidnapping plot - whereby villagers are being traded to a nearby elven village as sacrifices to keep a "nameless evil" sealed. You travel from Ilmater's Hope to the larger city of Lopovichel (and a few other locations) to either aid the kidnapping plot, tolerate it "for the greater good", or put an end to it. All outcomes bring about different benefits to your growing realm... and the elves swear that blood sacrifices will be needed to keep your villages in the region safe.

"Moral" Encounters. The demo was intended to present the kinds of difficult moral choices you would face throughout the game while giving you a taste of the basic mechanics. In the opening encounter of the demo, you meet a bounty hunter who states that your party's healer (Wick) is wanted in Vinnelhaven for selling expired potions, which led to people dying.

During the encounter you can:A. Give Wick to the bounty hunter (for free, or negotiate a "finder's fee"). If you do this, you will no longer have access to a healer! Sir Ozmark will be relieved to see justice served, and Ainsley will now openly question your loyalty to your "traveling companions."B. Intimidate or negotiate with the bounty hunter (and his hidden team) to leave your party alone. Options include "proving" that Wick has reformed, that she's no longer a threat, that her current conscription is more vital than an execution, etc. Wick and Ainsley will be grateful, Sir Ozmark will begrudgingly assist.C. Fight the bounty hunters. This could also lead to options A or B.D. Execute Wick yourself. Perhaps she did deserve to die after all.

Gameplay. The core "Combat System" for Realmweaver was actually an "Encounter System" designed for seamless interaction from "conversation" to "combat" to "negotiation" where you gain (or lose) accord with others by talking with them, fighting, or giving them gold/food/resources. Hurt their overall morale by outnumbering them, damaging their health, winning in a particular situation, or leveraging information ("We know you're bluffing!").

In one example encounter, you try to win over an audience during Ainsley's music competition as the other bard tries to out-perform, sabotage, and humiliate them. Other encounters include stealth missions, negotiations with bounty hunters, walking upon a bandit camp, and (potentially) confronting the mayor of Ilmater's Hope.

Writing. We initially started with a writing team of about 15 people, which included audio drama writers like Tau Zaman and Jordan Cobb, industry writers from Blizzard and Ubisoft, and amazing consultants like Jennifer Kretchmer, James Mendez Hodes, and Nala Wu, who all worked with us to expand the core species, factions, deities, realms, and world-lore from the Dark Dice notes we provided. We had initially envisioned a complex dialogue system for Realmweaver similar to Hades, but part way through the project we shifted the scope of the Encounter System to make dialogue options less involved. Instead of "every option in every encounter" requiring unique dialogue we opted for a generic button in each encounter that's simply "Appeal to Pride" or "Appeal to Justice" in which the larger conversation is implied - except during key story encounters. This shift ensured a much smaller budget would be required once the game was greenlit and helped reduce the scope of what had to be written to accommodate the demo. This is one example of where our Project Lead helped us avoid "mission creep (aka making a project too big for its budget)."

Visual Development. As mentioned earlier, each realm within the world was designed with a color scheme, emblem, costume style, and architecture in mind. We also earmarked a section for each group to focus on common disabilities (e.g. in a world of gigantic monsters, many elven warriors will live long enough to lose a limb or acquire some form of hearing loss). The first sheet seen above is an early example of a style sheet, and we quickly adapted toward a style that was closer to the sprites that would appear in the finished game.

Keeping to Budget. Most of the game would focus on the core human realms, but we also pre-planned what the expanded game could look like (with a bigger/smaller budget) for the purposes of a pitch deck. We also developed ways to maximize assets - especially for character animations, sprite sheets, music, and character portraits. For the latter, we batch-drew character portraits using basic templates. In this way the "scrawny lady template" (seen above) was used for a grandmother, a farmer, and a priestess at a 25% cost (for time/price) after the initial template was designed. By working efficiently, Mercado & team created over 80 character portraits in just a few months.

Soundtrack. Music was similarly designed to have many uses with overlapping layers that could be added/removed and remixed. Each region of the world was given to a composer who worked with Steven Melin to ensure that the overall soundtrack would be coherent. A core "Humanity" theme was designed - aka the "Theme of the Realmweaver" which would have been easily identify throughout our soundtrack. Composers also selected a specific arrangement of fantasy instruments for their realm, such as:

Northrealm - David Wise selected the Deduk with a focus on German chanting.

The Darklands - Brandon Boone wanted a haunting female-focused choir singing in Infernal.

Eastwood - Very violin and string focused with a more traditional choir.

The Bright Vale - A strong focus on hurdy gurdy and lower-end vocals.

These elements make each region easily identifiable, even when listening to the Dark Dice podcast.

Composers designed a "main theme" for each region, as well as a City theme, Village theme, and Encounter theme. Through the aforementioned layering, it was also made possible to seamlessly remix each song (especially encounter music) to be larger or more sparse as the intensity of the encounter grew or waned. We assembled an all-star team of composers, including Nobuo Uematsu, David Wise, Austin Wintory, Hitoshi Sakimoto, Brandon Boone, Dallas Crane, and Yuzo Koshiro who each developed a portion of our world's aural identity. This project is also a LARGE reason why the Dark Dice soundtrack was able to be created! We produced well over 2 hours of fully original orchestral music (60+ tracks), much of which still has not been shared.

Production. The project went smoothly (and rather secretly) for about 2 years. We maintained weekly production meetings, followed generally accepted industry practices, and were prepared to scale up production and expand the team once the demo was complete and full funding was acquired. During this "Development Time" we shifted from Tactical RPG to the gameplay style described above, we world-built, we went through hundreds of applications and hired our key artists, animators, writers, programmers. We continued to meet weekly, even through the start of the war in Ukraine (we employed Russian artists who we could no longer legally employ, and Ukrainian artists who had to relocate their families from the front lines - and were thankfully not harmed). We had an amazing team of talented folks who were paid fairly, and we encountered no interpersonal issues over the entire life of the project.

Cracks in Communication. Our project manager was a highly skilled, documentation-by-the-letter programmer who questioned everything and believed fully in the "Waterfall" Process as well as various other methods that hold up the corporate I.T. world. In other words, he's the kind of person who effectively stopped "mission creep", kept the project on budget, broke down long-term goals into weekly markers, and pushed through BS on the art or programming sides to ask "is this necessary?" Our project manager was "the stick" and the fuel that kept the project moving through thick and thin... This changed when he was offered a "real job" at a price that he simply could not refuse. We were very happy for him, but sad to see him go, and did our best to continue with less and less of his involvement. Without our key manager to keep the programming team in constant communication, things gradually degraded. (We at Fool & Scholar were already working full time and could not give more of out time to the project. It was not feasible, and it was not within our skill set.) Our Programming Director was also an incredibly talented individual, but not the strongest leader. At this point in the project we also needed to hire more programmers, which he didn't have the time to do because of his existing duties, which exasperated the team's problems.

Losing Traction. Our weekly team meetings became "art, writing, and worldbuilding reviews" and folks from the programming side eventually stopped showing up. The quality of programming documentation eroded until it was non-existent, and at the monthly company-wide meetings the promise of "we'll have a real demo next month" repeated for years as our existing programming team gradually left, largely due to lack of communication on goals, and the lack of waypoints toward the major goals. While we were happy to participate where we could, we did not go into this project alone, and the parts of the team who could build the game lost direction. The point of this is NOT to speak poorly about our programming director but to explain that without proper leadership Realmweaver ultimately faded from sight until every last person on the team (including our director) was gone. As of 2024, even getting an email reply from our director took about 2-3 months and it was clear that he had quietly moved on. On the art side, we accomplished most of our goals, but again - we are full time podcasters, and it was not our job to fill the needed leadership role for the video game production company.

Internal Review. The project's funding (from Lost Outpost Studios' dev team) dried up in 2024, and at the start of 2025 Realmweaver defaulted to the ownership of "the Dark Dice IP holders" - Fool & Scholar Productions (us). It's both known and foreign to us; some pieces we recognize and created ourselves, other sections are languages we don't speak and indecipherable magics (code) we cannot undertake the task to repair.We were not really sure what to do with it... The existing demo (internally dubbed "Bikini Bash" because despite having pixel art assets, all the character models are just the same guy in a bikini) - is not polished enough to present to a publisher. It's buggy, and most of the art assets we'd spent years creating were never implemented. The back end to the encounter system is also something of a mystery for us, as it has evolved over a few years of internal playtesting and tweaking. It is not a "bad" demo, but it falls short of a full proof of concept. It is not something we can present to industry professionals.

We asked a friend to look at the code, let us know if the project could be salvaged, and to estimate how much time it would take (in a very basic sense) to make the demo functional or even just aesthetically presentable. His report was grim. The last two years of code were poorly documented, very self-referential, and there are key bugs that cause crashes which cannot be easily located... There's a chance of salvaging it with a few weeks of work, but he believes that it would be easier to re-code the entire demo from scratch with better documentation and practices. We are podcasters, and we have no interest (or authority) to use our income to start a video game company - so with this post the story of Realmweaver ends.

...However, it's worth noting that the assets created by the project - the music, the art, lore, etc, will serve to make Dark Dice better. Those of you who play our D&D adventure "Domain of the Nameless God" may uncover the mystery surrounding Ilamter's Hope's missing villagers, and may meet other NPCs in future stories, podcast releases, and pen & paper adventures.

We hope that this post helps others stay on the path of production. We are now at a point with our work where we no longer have time to take on additional projects, but ideas that grew from this now cold corpse-code have helped us form future stories for Dark Dice. And there: in your headphones, on your speakers, itching in the back of your ears, is where we want to be.-Fool & Scholar