Jedi: Survivor
Senior Technical Gameplay Designer
DEC 2019 - OCT 2023
My role on the Star Wars team expanded drastically from Fallen Order to Survivor. As primarily a Tech Designer up until this point, my focus on Red Dead Redemption 2 and on Jedi: Fallen Order had been more oriented towards technical problems and support, but early in Survivor’s development I participated heavily in the prototyping and ideation process for new player and level mechanics, and by the time we entered full production I was straddling the line between technical and gameplay designer, which has continued since.
During the first months of preproduction most of my time was spent on ideation and prototyping of various new mechanics and toys for Cal. Many of the sequel’s new abilities originated with me, including Force Dash, Phase Gates, Bomb Droids, Merrin’s Reconstruction Power (originally an advancement of Cal’s Psychometry ability), and several more were pitched by other less-technical designers and handed to me in doc form to create initial prototypes for, such as our new reactive platforming elements, the Grapple Gun, and Merrin’s Force Dash Conduits. As the project moved forward I started working with Jedi’s established traversal experts and took full ownership of many of the new traversal mechanics that had passed through my hands during the prototyping phase, which set the course for the remainder of production.
My primary responsibilities on Survivor were ownership of the new Force Dash and Grapple Gun mechanics, the Mounts (Glider, Nekko, and Spamel), and all the new Reactive Platforming elements, things like collapsing platforms and ledges, bouncy or bendable rails, etc. This included all these mechanics’ various supporting elements, things like Phase Gates , specialized Grapple interactions, the Mount Stable, and so forth. Additionally I continued to provide a lot of level design support in the form of toys, tools, and one-offs; things like specialized Force-interactables, puzzle elements, and so on. Virtually everything I worked on had very limited or no code support, and were engineered entirely in Blueprint (sometimes for better and sometimes for worse).
I won’t delve into everything I touched, but some of the highlights:
Force Dash
Force Dash (featured prominently in this Force Tear guide to the left), allows Cal to tap the evade button in midair to quickly push himself in a given direction, acting as a kind of extra or triple-jump. It also allows him to bypass certain kinds of obstacles or interact with certain kinds of objects he can’t otherwise (move through air vents without being blown away, pass through phase gates, or use his momentum to push certain kinds of objects). It’s a classic platforming ability that I adapted for Jedi and pitched early in pre-production while we were exploring new lock-and-key mechanics for Survivor. You can dash at anytime in midair, but only once before touching the ground again.
In this video you can see several specialized Force Dash interactions: phase gates (the green barriers), which reset the player’s ability to dash in midair, Phase Conduits, which Cal can dash into and be swept along a predetermined path, and push-surfaces, which rotate if Cal dashes into them. These all came out of my prototyping testbeds, where I spent several months early in development just making random toys that interacted with Force Dash. There are a lot more that didn’t make it into the game (we could only get so weird with it, we have to fit the Star Wars IP after all), but I experimented with everything from dash-extension volumes, mirrors and bounce pads that could redirect dashes, nodes that the player could dash into and hop from node to node, volumes that let you steer your dash in slow-motion for long distances, etc.
Force Dash is not some wildly new or innovative ability, but I’m very happy with the result: clean, simple, and adds a lot of fun to traversal—in large part, I think, because it gives the player a lot more freedom to recover from mistakes. In Fallen Order if the player missed a jump or fell off a ledge, oftentimes the recovery window was very narrow. But with an extra burst of movement that the player can weave into their aerial acrobatics, mistakes could now often be fixed with quick reflexes, turning a potential moment of frustration into a feeling of nimble recovery. It also gives the player more than one way to skin a cat, as it were: the dash can take place before, between, or after the two standard jumps, and at any point while falling, meaning the player has MUCH more control over their overall trajectory, and can more easily land on strangely placed ledges behind or even under obstacles. Moving through their air becomes improvisational in a way it wasn’t in Jedi 1, adding a lot of intrinsic fun to just getting from one ledge to another.
Grapple
Grapple is another tried-and-true platforming staple, and its simplicity is the kind that took a lot of hard work between design and animation make it feel effortless. The mechanic (seen in the video to the left) is very straightforward: tap L2 in range of a grapple point to zip to it. The major challenges here were twofold: one, we wanted to be able to grapple to and from any kind of traversal state Cal might be in: from swimming to flying to sinking in tar to the back of a mount, and two, that Cal wouldn’t lose momentum during any part of the grapple interaction, entering or exiting, keeping the transitions fluid. Unlike the Force Dash, which was practically shippable with just a handful of animations, Grapple needed everything we could throw at it.
Entering the grapple was comparatively simple: Cal simply needed to exit any navigation he was doing (let go of ledges, drop off wallruns, leap off the back of a mount) and start performing his grapple sequence. There were still plenty of problems to solve here but nothing worth digging into. We turned down Cal’s gravity a little bit if he was in mid-air to create a little bit of a freeze-frame effect as he fires the grapple (and to prevent him from gaining to much downward velocity while waiting for the grapple to attach), but once that was done it was just a matter of moving him through the air toward the grapple point (with a little bit of a swing/arc for flavor). Some basic collision sweeping let us slide him past most obstacles. In extreme cases, if Cal managed to grapple something in such a way that an obstacle prevented him reaching the point, we would detach the grapple after a maximum amount of time elapsed, but that was a rare edge case.
But once Cal reaches the grapple point things get weird. In Survivor there are many types of grapple points, because there are many ways Cal can end a grapple. Cal can grapple into wall-runs, climb-surfaces, ledge-grabs, ceiling-hangs, you name it. He can grapple onto balloons which allow him to steer his jump and launch into the air. He can grapple into swingbeams and jump off like a trapeze artist. Each of these had to be handcrafted to create a fluid entry animation, release Cal back to the player’s control without losing momentum, and allow for immediate followthrough into the new navigation style.
Mounts
And last but not least: mounts. Mounts in Jedi were a bit of a mixed bag unfortunately. A lot of time and effort went into them, and at their best I think we succeeded in making the glider, nekko, and spamel (the bike was not under my particular umbrella) fun to navigate and traverse with, but there were some tech issues which kept the mounts from greatness—namely, some choices made early in Jedi: Fallen Order’s development which meant that the only character controller we could use, fundamentally, was Cal’s. As such, what mounts could do, motion-wise, was limited to what Cal could do, just tuned with radically different parameter values and collision. Re-engineering such a fundamental part of Jedi’s original motion code was just not in the cards, sadly.
The result of this was that while the mounts were fun to use and handled smoothly on open and level ground (or, in the case of the glider, in open air), things were a little less smooth when operating on rough terrain or colliding with geometry. For the most part these were minor issues, but in the case of the spamel, with its massive frame and extraordinarily long legs, sometimes things could get pretty gnarly. We did our best to work through these issues, but in the end I think things fell short of the bar we all wanted to deliver.
That being said,