I'm a former DigiPen student and a working game developer, nowadays working as a manager and recruiter / interviewer. I love DigiPen student games, and the coursework is such that I treat DigiPen students more as peers than children, and respect DigiPen student games to where I have decently high expectations and evaluate them not too similarly to a release game. Obviously, student games are student games, made in more limited time than professional projects by inexperienced people still learning, and DigiPen games often have a severe secret handicap in that they're often made in custom game engines made from mostly scratch instead of an out-of-the-box standard engine. Making that custom engine is an incredible learning experience, but it means that a lot more development time is often devoted to tech than polishing the game itself. All that said, however, I have high expectations for DigiPen games—I expect them to be better than even the best internet game jam games, at least—and I prefer to give very honest constructive feedback in hopes it might be helpful to the developers in any small way, instead of wearing kids gloves that I might with high school students or true beginners in game jams, etc.
Split Spirit is disappointing as it could be improved dramatically with some small changes.
[list]
[*] The overall art and packaging is [i]fine[/i], but it desperately needs improved game feel. The game has very few and basic mechanics, so they should be polished to an lustrous shine, and currently movement and throwing feels very sluggish.
[list]
[*]Simply increasing the game entire's world speed by 20% or so (maybe more) would dramatically improve the overall game experience without affecting level design.
[*]The jumping is very floaty and awkward. Ideally, gravity should be higher for better-feeling jumps, but that would likely impact level design and require significant rework. It seems the levels were art-driven, instead of grey-boxed and designed/tested/iterated with fun and game feel in-mind before overlaying art.
[*]The mechanic of throwing the soul similarly feels slow and awkward. It doesn't feel like it follows the preview path shown very well, falling with gravity faster than estimated by the preview. The soul travels [i]very[/i] slowly, especially after it starts bouncing around, which feels silly and out-of-theme, and annoying when a throw is miscalculated, slowly waiting for the soul to return from a failed attempt. Clicking seemed to sometimes return the thrown soul to source early (like a restart button), but it felt inconsistent, unclear if there was a cooldown before restarting, or if so how long. And either way, if clicking was helping to return the soul early for a retry (if it wasn't just me imagining it based on coincidence when it returned on its own), that wasn't communicated to the player. I'd prefer if the soul moved more ethereally, perhaps with an aim-assist, heat-seeking missile behavior, and/or slight waver up-and-down as it traveled to feel more like a spirit than a baseball. It'd be nice if the soul changed sprite dynamically mid-flight for more feedback and visual interaction too. But, it would also feel better simply with a faster initial velocity and sharper ricochets, moving more like a bullet than a lobbed slow-ball. I'd also really like if we could throw the soul while mid-jump as a deer.
[*]The hitboxes were very forgiving—the soul often connecting to a flower or deer when it was nowhere near it—but the hitboxes were so forgiving to the point that it felt awkward and buggy. I think tighter hitboxes would have been frustrating due to the finnicky soul throwing mechanic in its current form, especially when the soul needed to bounce around, so I'd prefer if these hitboxes were kept large, but represented visually so it felt less buggy, e.g., with a transparent aura or ethereal outline around the targets. Such an aura / outline could appear dynamically with a shader only when the soul approached it, reacting to the soul's presence, drawing the player's eye, and indicating the hitbox threshold where the soul would successfully enter the host. As a point of polish, it'd also be really nice if there was a simple animation for the soul entering a host, instead of just snapping to next to its head, like at least programmatically tweening the soul from where it touches the hitbox to its rest point on the host plus some particle effects. But I think I'd prefer a dedicated animation, like a simple one made with Smear FX by CodeManu, a big smeared sweeping motion spiraling around the flower/deer/host as the soul shrinks inside, then maybe a quick twinkle (like Team Rocket disappearing into the sky after blasting away) and/or the host glowing ethereally for a moment with a basic shader. The active host constantly glowing softly, acting as a light source, and/or shedding magical motes of particles would do well to emphasize the theming and highlight the active host to the player.
[*]There's a decent particle from the player walking, but Split Spirit needs more particles. A pretty trailing effect for the soul being thrown through the air, an impact particle effect when the soul bounces off walls, jumping and landing particles (maybe those were there but I didn't notice them/forgot them), etc. More particle effects is such a cheap and easy way to improve visuals and game feel, particularly highlighting every button press and interaction from the player's actions.
[/list]
[*]Most of the game (especially level 1) is slowly walking or jumping forward through a straightforward environment. I was genuinely disappointed when it said 'continue to next level' because it meant there was still more game to get through. But level 2 is significantly better than level 1, adding more interesting mechanics and level design that gives more for the player to deal with. This is common if levels are developed linearly; level design improves over time and new/better mechanic ideas get added. But it's important to go back and update earlier levels with the same quality as later levels have, or in the case of this very short game with only 2 levels, I hesitate to say that there should even be 2 levels, especially given how much of the first level is repetitive movement. It's generally good to expect your player to have seen WASD movement and basic platforming before and not drag it out longer than a first interaction tutorial. I would've liked to see more intricate use of the soul throwing (or perhaps better to say: soul splitting) mechanic earlier, integrated with the platforming. Given the content in the 2 levels, I feel like it'd be best if they were simply merged. But again, I get the impression that art guided the level design more than level design and playtesting, which is problematic (I could be wrong... but that's perhaps worse).
[*]In a perfect world, there'd be more mechanics, of course—more host types (e.g., fish, bird, rock), more world objects that interact with the thrown soul, puzzle design reasons to go back-and-forth between hosts in an area to solve a problem instead of just shedding your hosts one-by-one to move right-and-up, etc.—but making best use of what's there, I'd emphasize what's good and trim the fluff to maximize awesome per second and a fun experience over a slightly longer experience. Not that Split Spirit is long—it took me 11 minutes to beat, but it could've been a tighter 5-minute experience instead, which would be preferable. Leave them wanting more. Less but more polished content is better than more but less polished content.
[/list]
I could keep rambling ad infinitum, but the long-and-short of it is I think the developers can do better, and likely will! It's a good feat to release any game in a playable state, especially without much bugs, and I'm sure the devs of Split Spirit are already out there making ever cooler things. But standing on its own merit, with critical honesty I have to thumbs down Split Spirit.
