Articles | Starsworn, A Project

This article is independent from the two The Digital Tabletop Experience articles (one of which is not public at the time of this writing).


It is 6:13 AM, April 10th, 2019. I have a severe migraine building up, as is typical with days I am offered only a few hours of sleep. In less than two hours I must place trash outside for crows to gleefully tear apart. There is too much trash. My grandmother is throwing out all kinds of perfectly useful things, like coushins and blankets. But, they're also too much to hang onto in the event of catastrophe. The atmosphere is tense at the best of times.


It's no secret that times have been tough since 2014 and, more recently, 2017. I've tried to distract myself in true All-American spirit, but it's hard to keep one's head up and eyes forward. Currently, I'm grinding out 600k or so points out of a Fate: Grand Order event. I haven't had much time to devote to it except for the last few days. From building a new computer funded by HKS to dealing with yard work and politics, life has been unexpectedly hectic for a while. Things are looking to settle down before the storm that is summer and its insufferable heat, allowing me a chance to catch up on my biggest project since Retribution.


Indeed, even as I smash golden Nobus and flying saucers with whale meat and the migraine pounds ever more feverishly I am plotting and planning and scheming for my next week and a half of work. Just because I haven't had time to commit to my designs yet doesn't mean they have stopped.




Although my time with D&D as a player is long over, Starsworn itself yet lives and lives vibrantly. I have run it every weekend I have had availabilty from my players - a dicey prospect at times, given one is an archeologist, one hunts for corpses in vans, and one is a normie - and this schedule has been relatively consistent except for summer and winter holidays.


Last night I was reviewing videos of past sessions once more. The game is currently inbetween subchapter strides. The party has spent much of the time the last few sessions in one relative location, though they've experienced some wild and wacky excitement in this one spot. The project also recently saw an enormous upheavel when Fanburst, the "service" that occassionally provided music for the prototype digital tabletop platform Roll20, decided to kill itself a month after its suicide deadline. Without any alternative for music I was forced to adopt streaming instead, which meant redoing my year and a half of work on Fanburst locally - a prospect I was ultimately going to be looking at anyways, given HKS has been steadily making progress on his replacement for roll20 itself.


The logistics of Starsworn's assets mean that traditional solutions, like paying enough to buy a 3TB hard disk for a 2005 USB stick worth of storage on scam services, are completely unrealistic. Even managing the assets is difficult - I've scarcely scratched the surface of what Starsworn's fanburst presence had, and been flexing heavily for recent sessions. With almost 3 terabytes of music to sort through, edit piece by piece for loopability, and organize according to domain, mood, event and character, just fleshing out the game's soundtrack is a herculean task. One that hasn't seen nearly enough attention it has sorely needed since the Fanburst post-mortem.


It is a crossroads, and one I felt inclined to, once more, reflect upon the experience with. Rather than analyze or assess the game as I do in the extensive Digital Tabletop articles, I wanted to look at specifically the game, its players, and my own growth in this time from a somewhat different perspective - particularly since I've also been dabbling with Warcraft 3 in recent times, and it has reminded me just how much I hate modding.


Had you told me I'd be running a D&D game for the most unlikely of groups - Mucky, Thalraxal, Schwa and HKS - back in 2014, I'd have laughed you out of the pub. Yet, today, it's hard to imagine not running the game. The mental process of planning months ahead and preparing assets and material in kind while carefully dissecting the sessions and players is all but second nature, now.


Memfis Castle had been in preparation for around a year prior to the party actually reaching it. Seen in this image is a composite of assets - the tokens the players chose for their characters, tokens I hunted from the internet, a Starcraft 2 map created by Lollerskates, random junk off of google image search, and aught else. Indeed, the contributions from Lollerskates and Bloodyshade by hunting and ripping assets on the side have helped alleviate the workload somewhat, but areas like Memfis - spanning dozens of maps - require a great deal of manual labor to make happen.


Starsworn superficially is simply just a D&D game, but I treat it as an experience besides. All players have some history in consuming anime and other media involving mystery solving, mood setting, and strong characterization. The values of the media they consume - such as Fate, Berserk and JoJo - synergize strongly with my values as a designer. My goal to create powerful and memorable moments, give players the tools to make profound impact in a volatile world, and set the stage for grand spectacle has begun to take shape in Starsworn, 46 sessions in, back at the beginning of the year. The slow, steady buildup to Memfis has been over a year in the making, and the confrontations and resolutions to come as a result had been everything I had hoped for.


Between the Living World engine and the players decisions, the world of Starsworn has proven itself to stand entirely aside of my previous projects. Designed to be improvisation-oriented, Starsworn's preparation work, being the building of assets, statblocks and maps, is exclusively about giving myself the tools to deliver the content when it is needed. One can't always predict what will happen in the game, and Starsworn's world leaves a great deal open to chance that simply cannot be accounted for before a session begins.



In the battle with the Fallen Blades, HKS makes a critical decision that will reshape the entire game to come. It was a decision I never anticipated even in my most farflung predictions of how the battle would go. However, because I had prepared ahead of time with all of the assets necessary and knew how to granulate my world's mechanics in D&D terms I knew how to tackle the complex interaction of the catgirl and the box from the church. This simple, desparate act is such a moment of power that Starsworn has been engineered to facilitate. The fact that things like this can happen - and have, several times - places power in the player's hands that simply cannot be replicated in other game platforms.


This has meant that I've had to be five steps ahead of the curve, understanding the probabilities of the myriad interactions that can occur within the world I have created. I understand the full ins and outs of every world system, from how the elements work at all levels from physical to conceptual to understanding how precisely the abilities of every single character and piece of technology function. Consistency is paramount, especially in an unfamiliar setting. What the players may initially perceive as random never is - and as they uncover more and dig deeper they learn more and more how to make the world work for them. Having the world's framework as a written entity built and heavily polished prior to Starsworn's first days as a concept enabled my ability to flex to otherwise unpredictable results that may have resulted in traditional DM's bullshitting their way past - which can invite disaster by breaking consistency.


Delivering this complex information and breaking it down to the players in manageable ways was the greatest hurdle of Starsworn, a hurdle that costed me over a dozen friendships and the last community I was a part of. Once the game actually started, however, I realized all of the stress and anxiety was for nothing. It was merely a flawed approach that brought about this catastrophic result - and once I returned to my base instincts and heard the call of the wild, I was ready to track. Twenty years of modding had prepared me for this moment, I needed only trust in my ability to work under pressure - something that wasn't made easy by the fear of repeating the past.


With this group - composed of familiar faces - however, there is no doubt, no hesitation. This means I can handle the enormous toolset and palette I've crafted for myself since the dawn of the concept in a somewhat coherent manner, and roll with the punches of a prohibitive, archaic online platform designed to scam the ignorant rather than deliver content from creators to players.


Reyere's mental domain was a possible place the party could explore, but I didn't expect her mental state to degrade so rapidly and the party to instigate a reaction so soon. I was caught completely unprepared, and in a flurry of motions I had to composite three separate maps and a dozen different statblocks right as the party was getting sucked into it during the game. Thankfully, I knew ahead of time what the domain contained and what would happen within it given the circumstances. The phantasmal Greenborn had already been prepared ahead of time - with new statblocks that had been built directly as a result of feedback from the Test Sessions, which seemed like an eternity ago, even when this session was played early 2018. The maps were either things I had lying around waiting to be used or submissions from Lollerskates on my various desktop backups. I hardly even had the chance to read the game's text between my furious dumpster dives.


The players have presented a generally predictable and oft times hysterical balance to the game in terms of both personality and gameplay. Designing encounters has been extremely challenging, taking the better part of a year to understand to any reasonable extent, especially since official material was proven to be useless as a reference point from the very first test session. I've since built my own understanding of the game and game mechanics, and have designed almost every statblock, map and otherwise entirely from scratch with no outside influence whatsoever.


Storywriting focused on dialogue plays into my strengths, though at times I fear I am too willing to engage with it.


The party delegates on how to handle the mirror of Yu'Lee they've fought tooth and nail to reach while also trying to avoid dying to traps that surround her. Screenshots like this, from HKS' recordings of the game, are always weird to me because they look entirely unlike how the DM side looks. On my end I have a veritable army of random monsters and effects in case they are needed, colors overlapping the screen and auras everywhere. I can rarely tell what is going on at a glance; only the players see the scene as it is intended. Thankfully, I am accustomed to working in absolute chaos. Indeed, I have forgotten what it's like to even play D&D. It's easy to forget what the players side looks and feels like.


While linearity and predictability was to be expected given my conversations with previous D&D players and DM's, that isn't to mean one should be complacent with expecting the party to make obvious decisions or do things as you expect. A lot of my prep work doesn't get used, but the biggest, heaviest pieces overlap in such a way that they eventually come up. The world was always designed to facilitate an organic progression path that flowed to how the players moved and didn't move. The relative small scale of the sky city is still linked to the world at large, so I have to account for several layers of interactivity. Oft times things I've written off eventually return much later because of some interaction I hadn't foreseen, and I thank myself for having the material on-hand to adapt as necessary.


This means that wild and wacky swings can happen at the flip of a dime, and the players seem rather willing to dip their hands into the beehive to find some toothy honey.


Coming from projects geared to both multiplayer and singleplayer content, D&D has been something entirely unalike any project I have ever involved myself with. Not simply because of the lopsided nature of time-sensitive responsibility and how it persists, not simply because of the staggered workload and realtime nature of the project, but because of the reward. To see four players file into the game that you have created time after time has a certain satisfaction to it. Knowing that, in spite of the many trials the game has been through, people still want to see what I have created and participate in the project is ultimately what keeps the oil burning whereas in other cases I have dropped all involvement with project creation. Indeed, it is because I know the people involved that I can feel satisfaction knowing I am connecting with individuals with whom I share a common goal, and the project's continued development is that goal. Starsworn is a co-operative project. It is wholly, utterly unlike anything else I've ever worked on - even projects with far more people. Team projects like Carbot's Starcrafts was a thoroughly soulless and detached experience, and my own projects were frustrating more than anything.


D&D feels a natural progression from modding. It puts much of what I've learned and set in stone to the test, challenges my theories and my experiences, yet takes advantage of the few skills I managed to retain over the years and yet encourages me to advance further. There is always room for improvement in Starsworn. From the assets, such as the maps and tokens I painstakingly reverse engineer, color correct, clean up and remove watermarks from, to the social experience and presentation - pacing, soundtrack manipulation, and information distribution, Starsworn has many arms all of which I have significant room to improve on but all of them I have worked with before and thus weren't unprepared for.


Shown here is a spattering of work from all across the project's lifespan. The enemy tokens consist of a Tower of Saviors sprite, a Japanese kitbashed RPGMaker sprite, and a Chinese mobile promotional sprite that once had a watermark on it. That sprite was my first experiment in removing watermarks, and without comparing it side-by-side with the original you may not be aware a watermark existed at all. The map tiles are all from google image search and not particularly amazing, but they fit with the reclaimed assets and the starcraft 2 map to a satisfactory point.


Production value is an important element of establishing immersion, as is consistency. As opposed to the test sessions, Starsworn's campaign was designed to be as consistent as possible given the limits of my skills and the time needed to prepare. I restrict how much time I commit to preparation to a few hours a day tops, often only during 2 days a week. I may flex more time during a slow week, but I consciously try to maximize short expenditures as often as possible. That isn't always possible, though.


This battle and the subsequent resolution took longer than expected, and was one of the more difficult battles in Memfis. Which was good, because the door to the right - while it did have maps - was entirely incomplete on the other side and would have prompted another scramble. The party ignored or bypassed many of the encounters in Memfis Depths and then facepulled the Templar, which went rather contrary to what I expected them to do. As a result I had to rethink how I was working on Memfis which, even as the party finally breached its halls, was over 70% incomplete in spite of having started on it many months ago.


The Soul of Memfis had an enormous statblock, though a lot of it didn't get used. The statblock was simply a tool, and the way the world had changed prior to the battle ultimately dictated what happened more than the stats themselves. Furthermore, the party had racked up an army of NPC's by this stage, and had managed to smash most of Isabelle's exhaustion, making her effective in this battle even without her spirit weapons. Between all of this, the encounter was significantly easier than I had originally envisioned. A well-received encounter, the Soul of Memfis ultimately was a very simple phasic battle with a few twists towards the end thanks to the players actions elsewhere in the world. I was quite content with the result, and will disseminate it in detail in later publications.


I stopped looking for third party maps entirely after the Silent Summit. Other than submissions from Lollerskates, I deemed the internet too worthless to bother plundering. Too many deviantart dropouts think selling jpegs is a financially stable future for them, so real artists typically steer clear of the digital tabletop.



Isabelle's cave. Most graphical effects, including Bloom and depth fade, don't function in the orthographic renders necessary to create suitable images for D&D, which greatly prohibits what I can do with the maps visually.


I found it incredibly bizarre how, back in 2009, I still dreamed of making games. The system I built at the time, with the i7 920, would be the system I'd be using all the way until the dawn of 2019. That system had been built with the intention of getting into Project Offset, a game engine that never saw the light of day except for shovelware once Intel acquired it. The Unreal Development Kit, a far superior solution, would show up later in time, but by then my balls were broken and I had no hope of making a game project ever again. All pursuits for machinima and CGI died with Retribution and TheCoach's destructive deception, and my passion for modding died with Apex. Unreal 4 - the greatest opportunity to finally pursue my lifelong dreams of building my own games - came out, and I did utterly nothing with it other than testing old assets and toying with materials and lighting.


Only with Starsworn did I again find the drive to learn new things and try the unexplored. I focused on the few skills I garnered during my momentary UDK exploration with Retribution, and the single most important asset I had the power to actually create for D&D - maps.


Because of the prohibitive lightmass render times and the time-sensitive nature of my forcibly alloted mapping time, most of my maps are constructed in Unlit and I approach lighting in passes. Since I got the new computer from HKS in Q2 2019, however, most lightmass times were cut to a tenth of their previous lengths, making this largely unnecessary.


D&D reminded me how much I liked to create things. Characters, when I still played, and now maps and NPC's. New abilities, items, and concepts that probably won't actually make it into the campaign given their complexity. They are actionable, unlike in my mods, since I have all the skills necessary to make them possible. I'm a terrible spot writer, so I rely on assets to convey most of my descriptive storytelling. This leads to wordy exchanges, but the players seem to enjoy them.



Lighting a 2d orthographic scene is a nightmare. I think of lights as gameplay reads first in this case. This leads to invisible lighting in weird places I have to justify in the written word - a big nono for actual level design. Since I can't use dynamic lighting or player-based lighting, dark scenes are a careful balance of attaining the desired immersion while keeping the map readable on where tokens should be able to go and shouldn't. Starsworn aims for a realistic look, which means lineart and other tools used by the extremely simple stock material available for starter D&D campaigns to lay out obvious paths is entirely unavailable to me.


Point lights in weird places, especially with an orthographic top-down view, fuck with the eyes and create optical illusions. It is a troublesome thing, given I've seen the maps in perspective while working on them and subconsciously know all of their details but the players don't. Players experience the fucked up result as the first impression, and I have to know precisely how they'll perceive something without ever telling them it even exists. This makes the prospect of predicting and analyzing how they handle information I give them ever more important. I can't ask beforehand how they'll assimilate certain information, and some of the players aren't particularly vocal about how they are handling the advancement of the assets and information around them.


For tokens, my process is much less refined. I often have to edit token images rather heavily. I don't settle for subpar - and almost all "renders" on the internet are, at best, subpar. It's not that my skills in image editing are the least bit satisfactory, or that I possess any hidden tricks. It's that I am extremely stubborn.


The token for Nightmare Jebrefekt took around a day's manhours to edit from its isvoc source. Most of the hair was incomprehensible when zoomed in. It was my second time ever using Krita for this task and it performed admirably, a task that could have taken a week or longer in Photoshop due to the slowness of Photoshop's interface. I wanted to preserve the glow on the tips of the wings, but didn't know how to re-create it.


Any third-party art I use for my tokens usually gets edited in some manner. Even if it's a simple waifu2x conversion to chisel some jpeg compression out. Anything transparent, however, is done pixel-by-pixel. Sometimes I have to redraw portions of images, an extremely daunting task. Sometimes I have to shred metadata.



Lanczos resize (could never get spline64 into xnview and the rest butcher these) is not kind to such images, especially when Wilson doubles the chromosomes with jpeg and video compression. Even so, Nightmare Jebrefekt commands a presence that cannot be obtained with top-down, low-quality sprites typical of those on tabletop marketplaces. I was content.


The ability to endure pixel-by-pixel editing of so many tokens, converting assets and remastering them for Unreal to use in maps, and the patience to deal with the arbitrarily limited and undeveloped web middleware that brings it all together. This is the ability driven by the desire to create and deliver content. I could never bring myself to do it for people I don't know. Perhaps that's why the test sessions don't weigh heavily on me as projects that failed. Failure was to be expected.



As someone whom has given their life to the goal of creating things, Starsworn is a very bittersweet victory amidst a completely barren decade. I realize today, April 16th, as I push to get last minute time into my maps and tokens for the party's approach to Tenneth, that it really has been a decade since I last produced a functional project. Again, it was entirely inconceivable this is where I would be sitting. I knew that, were I to go on for a decade trying to do what I was trying to do, I'd travel down some truly dark roads. Starsworn, however, has been a beacon amidst those roads, and one that by nature I have returned to time and time again, my purse open and my makeup at the ready. In spite of everything that has happened, I still find myself willing to continue building, even as all the other lights in the neighborhood die out and the sky blackens to nothingness.


Indeed, there was never really much question once I asked if the remaining players wanted to continue. It was for their desires to move forward that I moved forward. In that sense, it is not the project itself that grants me substance. It is the knowledge that I've given the players substance that in turn grants me substance. I have, by mere chance, stumbled upon the recipe that creates developer-player connections. The truth of what a true co-operative project, a project with staff, is really supposed to be about. And the many truths of why the composition can fail, or why typical projects don't ever succeed for teams of randoms.


D&D and similar projects are devices of unsung trust between their members. Without that trust, without that connection, they cannot form into a cohesive shape. And, without no shape, there is no project.


Starsworn is, and has always been, about translating an existing idea into a medium by which I can connect to people through. The many errors I created in its design were always about muddying that simple idea without even realizing it.


Nothing lives forever, though. Once Starsworn is complete - something that could take a couple years, as D&D campaigns often do - I will most likely turn away from D&D and never look back. It is likely that other than pet projects I may never have the courage or mental strength to put such energy into a real, serious venture ever again. But, that is the future, and for now I am content with Starsworn. A project. An active project amidst so many years of nothingness and disappointment. I once dreamed of grand things, but life taught me I'm not built to create grand things. I was depressed over this for a long time, but Starsworn taught me there was one last thing I could do. Even if I wasn't sure how to do it, and even if I still don't really understand, I still feel confident in its cohesion and consistency.


And that is a difficult thing for me to feel. It is an alien sensation. Yet it is a welcome one. Amidst all the troubles and uncertainties of the world, Starsworn remains with conviction. Starsworn is a pillar in the murky years of hell following 2014 and 2017. For that, I am thankful. I am thankful for the players whom gave me a second chance. I am thankful for the time they have given me and the dedication they have brought to the gameboard time and time again. This is an experience that shall never leave me.