Tales of the Rampant Coyote

Adventures in Indie Gaming!

Frayed Knights Dev Update: Looking Back

Posted by Rampant Coyote on June 20, 2012

I guess it’s time to start talking a little bit about Frayed Knights 2. Having weekly (or nearly weekly) updates on the original on the blog was actually pretty handy for the development process, and we’re getting close now to the point where I may be doing the same thing with the first sequel. We’re not there yet, still not quite in full production mode, but we’ve had some breakthroughs recently that make me feel we’re a lot closer to it than we were only a few weeks ago.

What? We’re not in full production mode yet? What have I been doing the last several months?

I’m glad you asked. This is what the next two or three updates are about. But for this one, I thought I would use the wayback machine and revisit Frayed Knights development. Let’s see where we’ve been, so we might get a clearer idea of where we’re going. This isn’t exactly a post-mortem, but more of a way of providing background information that will explain what’s happening with Frayed Knights 2 easier.

It all started with a contest. A “Game In a Year” contest. I’d just made some very frustrating discoveries about some limitations of the Torque Game Engine (TGE) in a game that was in late alpha, that would have necessitated some pretty serious engine development work to bring it to completion. Since it was really supposed to not be much more than a ‘learning project’ for me anyway to get me to learn the engine,  I consequently decided to shelve it “temporarily” (I still harbor plans to re-make it using different technology, BTW). What I really, really wanted to do – what I was “building up to” – was to make an RPG. The other project was really not  And after investing a bit of money and a lot of time into the engine, I was gonna put it to use and make a game out of it, goshdurnit!

I was about a month late throwing my hat in the ring, but I went for it. Some of the design for Frayed Knights was dictated by the engine. Particularly after running into some of it’s limitations, I decided to build the game around its strengths. In particular, the first-person view. As much as Garage Games proclaimed the engine’s flexibility, deep down in really wanted to be an FPS.  So I leveraged that. But since so many mainstream RPGs also want to be an FPS (or third-person brawler, I guess) deep down as well, I made a deliberate decision to go back to an older form of first-person-perspective RPG that had been abandoned by mainstream (and most indie) developers, one in which I felt there was still a lot of potential left to explore: The first-person party-based RPG, a la the Wizardry, Might & Magic, and Bard’s Tale series.

But I didn’t want to just rehash the past. I wanted to invoke the old style of game, but add my own voice, and a modern approach, to the mix.  The idea of having party members performing “table talk” commentary during development came very quickly to me. I wanted the party to have a personality of its own, so that the player truly felt like he was guiding a group of characters on their adventures rather than just maneuvering a ‘blob’ of stats through dungeons. I wanted the player to care about the characters in the party as individuals, not only by their combat potential. The humor part also came naturally to me. Aside from “The Bard’s Tale” parody that had come out a few years earlier, there really wasn’t too much by the way of comedic RPGs.

But I didn’t want to make this game just a parody. I wanted it to be humorous, but I wanted the humor to derive more from the characters and situations, not just an endless parade of absurdity and silliness. Again, I wanted the player to care about the characters and the story, something hard to do when it constantly descends into “stupid humor.” I also wanted it to be a serious RPG under the surface. It had to stand on its own as an RPG even if the story and humor fell flat on their face. It had to be a good game, first.

I had a lot of experimental ideas I wanted to try, too. For example, the whole game mechanic of picking locks or disarming traps has always sucked IMO. I love the concept of the rogue class, but implementation-wise, it’s often pretty boring, resolved by a single dice-check (if that). What are the consequences of failing a lockpicking check? It’s so pathetic that many games have done away with it entirely, making it a deterministic check. I also wanted to play around with the idea of encouraging players to “play through” tough situations, rather than just reloading from a previously saved game.

The result – for the contest – was “Frayed Knights: The Temple of Pokmor Xang” – a “pilot episode” for what I expected to be the full game completed maybe a year later (HAH!). The idea of a pilot episode was a lot like making the pilot episode of a television show – something to test out on audiences to see if it will fly, and to see what needs tweaking before going into full-on production mode on a series. That was exactly what The Temple of Pokmor Xang was. I was experimenting and getting real, live audiences to try out a free game and provide me with feedback. While I’ve been a game developer for many years, and an RPG fan for even more, making a “real” RPG (Hackenslash doesn’t quite count) was still a challenge for me. I gained a lot of sympathy for the designers I’d often criticized in the past over their design decisions. Once you have to make those same decisions yourself, and truly consider the trade-offs, the alternatives aren’t quite as superior anymore.

It got my feet wet. I discovered a bunch of failures in my design from that one. I don’t have the link “live” anymore, though I’m sure you can hunt it down if you really want. It’s been surpassed completely by the free demo version of Frayed Knights: The Skull of S’makh-Daon – which fully encompasses everything from the original release, but is far more cleaned-up. But it’s the same adventure.

The contest was a success in every way but one. I learned a lot. I got a somewhat functional RPG engine built in TGE. I even won the contest. Unfortunately, the promised prize money for the competition never arrived, which was a bummer. The company supporting the competition dissolved just as the contest was drawing to a conclusion. The remaining organizers announced the winner – with their private apologies – as about their final act as an entity.

I prefer to chalk that up to the organizers being clueless newbies who discovered that sponsoring the contest didn’t drive their business revenues the way they’d hoped. But as I was operating under the assumption that I would not win the contest anyway, the lack of prize money wasn’t a direct impact on the game.

Indirectly, however, the combination of disappointment and my day-job (in the games biz) ramping up to full-on extended crunch for months and months really did hit my morale and development time. I’d anticipated a year to bring the full game to conclusion. Between development lethargy, a surprising number of changes that had to be made to the original game engine and the full game, and simply MUCH LARGER DEVELOPMENT TIMES than anticipated for adding new content to the game (spoiler alert: That’s italicized and all-caps for a reason – see future FK2 updates), progress was really slow.

A year later, not much had changed with the game from our pilot episode. We had a lots of bits and pieces of new content, but things were just not coming together. And considering the scope of the game, it was clear that we were in trouble.

A bit on the scope:  We’d had creep. We’d had creep, and I was learning that my projections were all wrong. I’d built a storyline and setting that I thought would be appropriate for a decent-sized indie release. It was about one-third the size of an original plan, which I’d recognized as being far too ambitious. But I realized that my scaled-down plan was also very ambitious, especially for a ‘first’ game (even with the pilot).  While the amount of content would probably be appropriate for an action-RPG that you were expected to blitz through, for a detailed, more methodical game with an emphasis on exploration and discovery, I’d still gone way overboard.  What I thought was only going to represent maybe twelve hours of total play-time was actually going to be more 60.  Or more, with the way we kept expanding and filling in things.

So I made the decision to break the story in three pieces, and make Frayed Knights a trilogy.

This was a really good thing, as the game would still be in “early alpha” today if that hadn’t happened.

My idea was to just re-use the same engine for all three games (with tweaks and enhancements), with new content. In fact, much of the development of Frayed Knights 1 included testing higher-level characters and spells than you can actually attain in the released game. There’s not much content to support such high-level play, but I’ve had 30th-level Ariannas, Chloes, Dirks, and Benjamins running loose through the Plane of Anarchy laying waste to everything they see with little effort or danger.

The change also necessitated a re-working of the story, as each piece of the trilogy now had to stand on it’s own. In retrospect, this was a good thing, because the middle part of the story (which has become Frayed Knights 2) really didn’t have much meat to it – it just drug things out while the characters leveled up to get ready for the exciting conclusion.

Development still took a lot longer than expected. Content creation was still very slow. Testing was likewise very slow and painful, as we basically pulled together all these elements together into something resembling a “game.” It needed a lot more work to stand up as a cohesive whole than I’d expected. Again, all learning process stuff.  DGM kindly provided me with copies of some of the more amusing feedback we’d exchanged during this process, which I’ll have to edit and put up as updates soon. You can get an idea of some of the craziness we had to deal with during this process.

The game engine remained a problem. It had been “sunset” by its manufacturer during development, and the problems of obsolescence still keep coming back to bite me (particularly the Mac port – which is proving very difficult to put together on modern systems).  I simply can’t produce both a game and tons of maintenance on an aging engine at the same time as a mostly solo programmer on the project. A lot of the “bugs” we encountered were engine problems arising from unforeseen compatibility issues with modern PCs.  Fun.

Somehow, we managed to get the whole thing released. Frayed Knights: The Skull of S’makh-Daon has been the winner of multiple awards and lots of acclaim from a contingent of old-school western RPG fans. It’s been a labor of love which managed to turn quite a few heads and talk. Plus, I think it was an idea whose time had come. It is riding something of a revival wave in the indie “scene” of old-school western RPG ideas, nestled among recent and upcoming titles embracing bits and pieces of the same theme: Legend of Grimrock, Age of Decadence, the Eschalon series, Wasteland 2, Swords & Sorcery Underworld: Gold, the Darklight Dungeon series, and more.

I had a minimum threshold for producing a sequel, which I didn’t want to make a big deal about. If Frayed Knights 1 hadn’t sold a certain number of copies, I would consider it a failure and would move on to something else. Other games, probably another RPG (I have like six early designs I really want to make…). But since the story stood on its own pretty well, and at that point hardly anyone would have played it, it wouldn’t be a big loss to anybody if it ended up as only a single experimental project.

Fortunately for the fans of the game, Frayed Knights blew past that threshold in its first week. At that point, it made more business sense to build on Frayed Knights, improving on the concept and the technology, than to start over with something new just yet.

But the deficiencies of the engine, our content pipeline, our process, and even the game design (and yes, interface) itself were also exposed. The sequel must be more than just brand-new content.  We had a story, we had a game system, we had a lot of things we knew which worked, we had fans, and we had a loose design ready to go. But we had to make changes and improve things for the next time around, building on what made Frayed Knights: The Skull of S’makh-Daon great with all-new tech and processes, and lot better understanding of what the hell we’re doing now than we did the first time around.

But that, folks, is a story for later.  Actually, several stories…

Oh, yeah. And in the meantime, if you haven’t tried out the original Frayed Knights yet, you can check out the demo absolutely free right here:

Frayed Knights: The Skull of S’makh-Daon Demo Version

Alternately, you can grab it from Desura. Whatever works best for blasting pus golems into oblivion…

Filed Under: Frayed Knights - Comments: 8 Comments to Read

  • Albert1 said,

    Don’t you think that using a 3rd party engine slowed you down more than having crafted an engine by yourself? I mean not a full, super-flexible engine, yet a framework complete enough to serve your purposes.

  • Rampant Coyote said,

    Sadly, I’ve asked that question myself several times. I don’t know the answer for sure. TGE gave me a lot of very cool functionality “for free,” at many times the learning curve for the engine – or for different aspects of the engine – and the extra work I had to do to “fight” the architecture really seemed to hurt me more than it helped.

    After I created my home-brewed engine for Void War, I decided I wouldn’t create my own engine ever again – I really wanted to make games, not engines. But I spent something like 3 years doing almost nothing while tinkering with TGE prior to beginning Frayed Knights – lots of throw-away projects and stuff.

    But I did get some cool effects and possibilities that wouldn’t have existed if I’d had to do it all myself.

    So as of right now, I’ll call it a wash. If it had truly been easy to port it to Mac, I’d have given the nod in the engine’s favor. But because the engine hasn’t been supported in years, now, that’s become surprisingly difficult.

    I don’t think the take-away for this particular engine should be considered a lesson for *all* engines, however. I’m working with Unity now (more on that in a future update), and while I’ve run into a few limitations and problems there, comparatively it’s a dream to work with.

  • Albert1 said,

    If I were to start the 90s indie FPS I’ve always wanted to do, I probably would:
    – not write an engine: I’d write some base code to make simple navigable environment with SDL+OpenGL, then add features only when needed
    – not write a level editor: I’d use Blender, exporting levels via simple Python scripting – Blender would be good enough for lightmapping too
    – not write it in C++: I’d write it in ANSI C, using a single struct for every entity – yes, I know, oop purists would kill me, but the fact that an entity is an enemy doesn’t mean that those weapon related fields are wasted space – maybe when you kill it, it’ll drop some ammo, hence the need for that ammo_count field
    – not follow the “Quake path”: i.e. not writing an operating system, not writing a virtual file system when you can make a simple utility that accepts a list of files/folder name then packs them into a big one and emits a include file to add to your source. I know, this last point would make modding difficult, but I think that modding is fading away, and if you still want a little more flexibility you can split the game in exe and a dll containing the replaceable logic(filenames, AI, etc)
    – making it single threaded: it probably won’t be the latest in graphics&technology, so I don’t see the need to complicate my programming life…
    – not writing bsp compilers, etc: a simple sector/bvh approach would probably be fine enough for levels so simple
    Sure, this approach would make very difficult to extend my simple FPS into more complicated ones, but I prefer to ship a game with hard to mantain code base, rather than keep a never ending gem of reusability and good practices lost in my drive!

  • ChrisH said,

    Wait, wait, wait. The grand prize for the game in a year contest was never even awarded?!

    I entered a game called Bit Battles and made it into the top 10 before communication with the DreamGames company/judges completely died out. My last email was from Dave Young (head DreamGames guy) months after the original ‘end date’ for the contest saying the judges hadn’t reviewed my game yet. I never received another email, point tally or even an announcement declaring a winner and I certainly didn’t receive an apology.

    I vented (ranted, really) about this contest and the InstantAction contest in an old blog of mine so I won’t bother rehashing things. I just find it amazing the contest ended in such a way and yet most of the other contestants seemed very indifferent about it at the time.

  • Rampant Coyote said,

    Yeah, well, now you know. You didn’t really miss out on anything. I got an apology, at least. I didn’t really see a point in making a big deal out of it. The whole contest ended late and anticlimactically with a fizzle and Dream Games going out of business.

  • jwmeep said,

    Since we have a nice little retrospective here, I’ll list the things that worked for me, with the game, and the things I didn’t care for.

    Things I liked:
    The lock picking, trap disarming system: Brilliant in my opinion. Most games either make it a stat roll, or a minigame that’s nothing like the rest of the game, and is annoying. A turn based mini-game that’s almost like combat, but with unlocking, and uses your stat rolls? Again brilliant.

    The Drama stars: It did it’s purpose. I saved often, but didn’t reload often. I took the hit of failure, and wounding several times to get those stars. Definitely reduced save scumming, and made it pretty interesting.

    The combat: Many people knock first person, turn based combat in RPGs, but you pulled it off. It definitely felt tactical. Not just messing around with menus.

    What didn’t work for me:
    The interface: Seemed clunky and awkward. Also wish you could have used your mouse more in it, like clicking on the enemy you are trying to target.

    Random combat: As much as I like the combat, it got awfully repetitive. And given how the combat in this game is difficult, having it happen so frequently made it a meat grinder. I’d get get out of a few difficult scrapes, run back to town to heal, and get the gold ring up, only to have random combat kick in and get killed before I could successfully run away.

    The humor: It was hit and miss. Loved stuff like a the Rat quest, but some of the jokes really fell flat. Somebody on another board said that Frayed Knights is more Silly, than funny. I agree with that, but still, silly worked for it. I came to love the characters, so it didn’t bother me too much.

    Linearity: I know part of it is due to the story line and characters, and you are probably saving more non-linear stuff for that nebulous, super ambitious RPG sometime in the future you’ve hinted that you want to make, but a few more optional areas you could get into without having quest pre-requisites would have been nice.

  • Rampant Coyote said,

    A lot of those things you touch on there are pretty common things in the feedback I’ve received. Definitely stuff I want to address in the sequel – keep the good, see what I can do about the not-so-good.

    The humor will always be hit-or-miss, to be honest. Although I think I learned a lot from the process, and I think I’m a better writer today than I was when I started. Hopefully that will translate into better writing for games two and three, though it’ll still have generally the same flavor.

    On the linearity side – that’s one I can argue a little bit with, because at least on the optional content front I tried to throw in an ‘optional’ dungeon (or two) in almost every area (the Rat quest was actually optional, too). But non-linearity means more than just optional content. It’s something I hope I’ll be able to address in the sequel as well. One of the big issues we had with the first game was – as I said – the difficulty of providing content. A lot of what I’ve been working on the last few months – besides getting up-to-speed on the new engine – has been tools to make the content creation much easier (and more robust – a key issue when testing a Big Game!). While reality always beats the crap out of theory, my hope is that it will allow us to be a *little* more liberal with alternate and optional paths.

  • KombAtMiBroh said,

    Man, I can’t believe it’s already been more than half a year since FK came out. I was itching to replay it again after finishing the first time to try out different character builds but life and work got in the way.

    Overall, I had a great time with it…unfortunately it’s been half a year since I last played it, so I don’t remember details only lasting impressions which is pretty much summarized as: “Had a fun time playing it, want to revisit…”

    Things I would improve have already been mentioned: Interface and linearity. I remember the way partially filled drama stars filled in tine by tine after using a star still felt off, but otherwise a great design choice.

    Linearity can also be partially addressed in map design…you can look at games like Thief and Deus Ex for examples. Dungeons are linear by design, I know, but other areas could have multiple entries, allow for multiple solutions and access to optional content.