* writers note: This post was written across several days and several emotional states. It oscillates in tone widely. hashtag-deal-with-it. *

There’s a moment that sometimes occurs as a game designer that can sometimes be difficult to explain to others. You have this game prototype. It plays well, it looks good, the systems are coming together, you have a plan for the asset build out. It WORKS. You begin to show it off, to small groups, your trusted friends. Maybe you soft launch it online. And then it strikes you.

This is not the game you want to spend the next year+ of your life making.

This is the long and short of why I disappeared shortly after launching this development blog. The feedback and support I received was incredible. Folks at the Oculus Connect conference thought the control felt incredible. No-one got nauseous (which I’m still quite proud of).

But, in news which will be ‘bad’ for some of you: The demo shown in the prior blog post (that some of you got to catch in DK2 glory at Oculus Connect) is no more. I decided shortly after Oculus Connect to start again from scratch. In what follows, I will hope to outline some of my thinking, show you where the project is at currently, and where I feel things might go.



So the first elephant in the room to talk about first here is Descent, the originator of 6DOF in interior spaces (i don’t know if it was actually the FIRST first, but that doesn’t matter much to me, its the one we talk about). If you’re making a 6DOF game, its either being compared to and spoken about in relationship to the legacy of XWing/Wing Commander, or Descent.

I didn’t realize how much of a problem this was for me until the Kickstarter for Descent Underground was launched. Presented by ex-star-citizen-guy-and-friends, with a purchased license to use the Descent name/brand/logo, they cast their vision of a follow-up out into the world. Note-to-self for the future: if you can’t be the first to plant a flag, buying the biggest flag works too.

Anywho I leapt on over and watched the Kickstarter video.. and was horrified by what I found. At least 8,164 folks obviously had the opposite reaction, and opened their wallets wide. Now, I don’t want to jump on a hate-train regarding their vision for their game. Such things aren’t terribly interesting. What IS interesting (to me at least), and in many ways central to the past 6 months of my work on Null Operator, was WHY I had the reaction that I did.



Their vision of Descent was nothing like mine, in tone, in focus, nearly everything.

Had we even played the same game?

Maybe we hadn’t…

I was 11 years old when I played Descent. I was terrible at it. I mean.. loathsomely bad at it. Even with a 2-axis 2 button joystick and keyboard, I struggled with the controls for weeks. I got lost constantly. I died when the reactor exploded so many time, unable to get to the exit in time, panicking as the timer approached zero.

Confession time: I never beat the game without cheats. I probably got up to about the 12th level or so ‘naturally’. I even had a strategy guide and I couldn’t do it. After that, I tended to just load a level at random, give myself a bunch of high-power missiles, and just enjoy the experience of flying through levels aimlessly, listening to the FUCKING FANTASTIC redbook audio soundtrack. SideNote: 11-year old Anton never knew he was listening to Type O Negative and Skinny Puppy, but he loved it. Explains so many fucking things…

ANYWHO over time, the thing that brought me back again and again was the experience of movement first and foremost, the topology of each maze and flying guided missiles around them. Shooting that irritating Guide-bot over and over again. Hunting and killing that damn thief-bot. The sheer joy of 6DOF movement.

— shot of descent

Descent is a sparse game if you go back and look at it. Its a testament to minimalism relative to our current digital toys. There’s not a whole lot of ‘there’ there.

To author in its tradition is to first recognize the amazing gulf between its technological and ludic complexity and that of the current era.

Thus, to create a ‘spiritual successor’ to such a title is to necessarily latch on to the one or two things that stood out to YOU in this past experience, what grabbed your imagination, what kept you coming back, and to fill in the 90+% of ‘modern’ game that one needs.

This will inevitably mean that one’s vision of how a modern sequel to a game of one’s childhood would or SHOULD look and feel like, is going to be so very different from another’s.

What I came to realize shortly after Oculus Connect, staring back into my demo and trying to decide what I wanted to do next, was the sheer number of decisions I had made authorially that were about its proximity to Descent, and really nothing more.

First it took place in a mine, then in a mine-like structure, then just in an asteroid, then in something less explicable.

First you were a dude in a ship, then just a ship, then maybe a robot-thing in a ship, then I-don’t-even know.

I couldn’t figure out who was the aggressor in the situation.

I couldn’t figure out what I wanted the point of your presence in this space was.

Why were you destroying these robots?

What mental gymnastics was I engaging in to have the player archetype not just be the prototypical psychopath mass-murderer protagonist?

Why do we think its less violent to blast robots apart than people? I like robots more than I like most people.

I don’t even know what I’m trying to make anymore.


I don’t want to make the ‘Descent’ some people seem to want at all.


Now what was I going to do.



Since that point, I shifted gears for a while. “Back to the drawing board” moments for me are often spent taking a project 100% off the screen, off the machine, and back to my sketchbook. Smoke some trees, do some drawings, repeat until something meaningful comes out. Thankfully two sketchbooks and twice as many months later, something started to.

I was working on a demo for the Alloy Shader Framework, when some ideas and impulses that had been swirling in my aesthetic mind began to come forth. I was about 2 weeks into this demo; a scene of industrial sci-fi ruins within a large cavern, lit from above. Everything was covered with sand. The whole scene had taken on an almost.. grave like quality to it.

I decided I wanted something dynamic in the scene, something that had a large effect on the lighting situation, and drew inspiration from my sketchbook to create a large robot. Procedurally animated, I set the construct up to wander the scene’s periphery, casting its search-beam about. It felt sad, wistful, almost gentle to me.


This, this was it. This was the feeling I wanted in this space. I didn’t want to fly around shooting it. I wanted to help it. I wanted to know why it was here, what it was looking for. I wanted to ask it where everything else had gone. Without explicitly looking for it, I had found a new sort of tonal center for my explorations.


I decided to continue to riff off of what worked so far, taking sketch ideas from my notebook and creating simple robots, each with an entirely generative movement and behavior set, sound dynamics driven by the same set of equations. The robo-jellyfish-like thing, and tink-tink came out of the process. I love them both dearly.

More importantly, what struck me most powerfully upon creating these three constructs was that I wanted to make something more meaningful involving them. I didn’t know what yet. But I’m not afraid to try. The gaming world will always have a surplus of shooty things. ‘Losing’ one should be a problem for noone.

So, what now?

I can’t say I’m sure right now. I know I want to continue building from the bottom up, staying loose, and most importantly trying to ‘forget’ Descent even as it continues to inspire me emotively. When I look at the mess of demos, experiments, sketches and the like that I’ve produced the past six months or so, a few things stand out as points of interest that will be central to what comes next. Note that each of these is going to be getting a blog post of its own in the future.

6DOF Movement – absolutely central. I just adore this way of moving about an environment. Maybe its the 3d modeller in me, but I can’t stand being limited to just walking around a space. I want to come at it from any angle, explore it from any vector. More so, I love dedicating so much of the controller to just pure movement. So much expressive potential. I also think one of the draws for me is the amount of patience it takes to really master. And frankly, if you as a viewer/player/whatever lack patience, well I don’t much fucking care to cater to you.

Physics – Especially with the PhysX upgrade in Unity 5 (my fucking noodley appendage it’s soo much better), I just love driving as much as I can through actual physics interaction, and building higher and higher complexity systems that interact via physics. For me its all about the messiness, the potential for emergent results, unique experiences. This leads to:

Procedural Animation – I can’t STAND keyframe animation, and avoid it as much as possible. Though the unity profiler screams at me, pretty much everything that moves in my work is driven by parameters, noise manipulation, relationships, raycasting, physics, etc. If I can make something situationally dependant, I do. I don’t like being limited to cross-fading ‘locked’ in clips with my animation in an attempt to reach a believable/interesting result.

Robits – I love them. Don’t feel like justifying it any further at the moment.

Weakness – Totally done with making digital power fantasies. At 31, I’ve just hit tilt on this. I’ve role-played as super-capable fictive characters enough for several life times. We’re all weak. We’re all fairly powerless relative to so many facets of life. We almost never explore this through this medium. We need to grow the fuck up. In the last demo of Null Op, I tried simply making the player frail in combat. Making that engagement as weighty as possible, as consequential as I could. I think my desire to please people, and inflexibility in thinking about a 6dof experience bereft of combat kept me from seeing the obvious for too long. Being weak sometimes means not being able to fight back.

Coming Next

Now that I feel back on track with this project, you should expect more frequent updates to this blog. I’ve begun building several different system prototypes, the first of which I’ll be doing a public alpha test with very soon.

It feels good to be out of the creative rut. Hope you’re still interested in what I have to share with Null Operator.



14 Responses to “Turns Out Second Blog Posts Are Even More Awkward”

  1. Kein

    I didn’t want to fly around shooting it. I wanted to help it. I wanted to know why it was here, what it was looking for. I wanted to ask it where everything else had gone. Without explicitly looking for it, I had found a new sort of tonal center for my explorations.

    This sounds fucking fantastic. You choose the right direction with this one.

  2. vampirewalrus

    This is *exactly* what I want from a Descent-inspired game; you’ve completely hit the nail on the head for me. I don’t even want a Descent spiritual sequel, I want someone to take that one part, the part that everyone else is ignoring, and develop it with this kind of clarity.

  3. DataStorm

    I played all Descent editions a lot back in the day. I was good at it, I was always aware where an exit was, what was relational to me in what direction even if I was looping around a enemy robot to evasive movements etc. I cannot count the times I was out of rockets, and needed to get the robots to shoot each other and then loot their rockets they dropped. I used my Joystick for all degrees of movement. The main direction with the first joystick XY, the second joystick was used as horizontal plane strafing, and turning at the same time to keep the target on the enemy robot, the throttle was used to go forward/backwards or keep steady. I used the HUD to look around, and 6th and 7th button on my MS Force Feedback joystick to strafe vertically.

    Wing Commander series was also a game I remember fondly. But the maze of Descent was always a fun environment. As for goals, just exploration of such would get boring fast imo.

    I would love to see this come to light and play it.

  4. silentw

    This is the game I want someone to be making. I have 500 million ways of shooty-shooty-bang-bang, but very few ways of just… being in an alien space.

    • anton

      You and I both :-). It can be hard to design away from the shooting paradigm. Nearly everything about the tools we use is built around the heritage of first person shooters. I think the key to correcting this balance is realizing that we already have other great design models for satisfying reflex play (navigation, pattern matching, aiming for other purposes, etc.) capable of supporting a wider emotional and intellectual spectra than shooting. Its like… everyone celebrated portal/portal 2 yet no-one reeeeally took it seriously from a design perspective.

  5. DasBilligeAlien

    All your thoughts in this post feel very familiar. I think you are on the right track.

    I am struggeling with my direction in making games for some time now And the conclusions I had were the following:

    Reducing the violencne or change it’s focus. Like, instead of fighting enemys directly, attack thier moral and command structure or even rob e player of all fighting capabilities. Or having to deal with cosequences like refugees or prisoners of war.

    Movement is the key in my enjoyment in playing games. The important part in this is using all 3 dimensions in your game. Not just as a decoration. I hate it when enemys can crawl around the level but i can’t even open the friggin door.
    I am mostly a 2D Artist and it hurts me that my ability in 3D cant keep up with my desire and speed to create.

    Procedural everything. I just love the diverity it can create. To have a loving world you need a decent amount of content creation and simmulation. Something that can interact with the world on a fundamental basis not just as a moving cardboard target to shoot at.
    Again I am laking the skills to achieve this. So I am stuck on this at the moment.

    Sorry for random rambling in your blogpost. It just resonates with me greatly, I hope you find fullfilling in your new direction and i can’t wait to play and see more of the game.

    thanks for sharing your thoughts.
    best wishes

  6. Paul Drummond

    This looks very interesting, with a similar sense of melancholy and decay to something I’ve been working on. I’m using Unity for development too. How are you combining IK and procedural animations with colliders and character controllers so the agents can interact with the world? Combining the two is quite a challenge!

    • anton

      Hi Paul. For all of my constructs in this project, everything is structured as full physics rigidbodies using compound child colliders. No joints (they can’t handle high speeds), no mechanim (doing all animation purely by code). All of the agents fly via a combination of sensor/waypoint based navigation using physics forces and a lot of horrific vector math. I agree it’s quite a challenge. I’ve been slowly working on the codebase that’s become these things for ~3 years now. Sloooowly and paiiinfully.

      • Paul Drummond

        I just bought the FINAL IK system for Unity to handle the procedural animation side. I need to wade through the demo code to work out the best way to combine it with some kind of collider system. It looks like it’s possible, but there isn’t much in the way of documentation. Trial and error seems to be the way forward.

        • anton

          In general, you can move sub-colliders of a rigidbody (ie. that are part of of a compound collider), BUT there are two things to be aware of.
          1. You will pay a cpu cost as it recalculates the aggregate collider for the rigidbody. To keep this cheap, only use primitive colliders (capsule, box, sphere) if humanly possible.
          2. If you are moving a sub-collider INTO a collision, it will not register that frame, or any frame during which its moving. If you need sub-colliders that move to still collide with things, only move them in fixed update, and only move them every other physics-frame. This will give the aggregate collider ‘time’ to interact with the world.
          And yes, its all trial-and-error. Physics programming it seems is 20% following the rules, 20% throwing darts at the wall, 20% evil vector math, and 40% luck.



  1.  Descent Gone Melancholy In Null Operator | Rock, Paper, Shotgun

Leave a Reply

Your email address will not be published. Required fields are marked *