Why only the first level? Because the first seconds/minutes of your experience is incredibly important and the hardest to design. As a game designer, when you spend years building your game it becomes difficult to have a clean slate approach and understand how first-timers will interact with your experience.
Having only played 28 minutes of this game I definitely am bringing a clean slate perspective. I hadn’t even watched the trailers then so I was completely new to the experience. I will try to explain what I think of Splashteam’s design decisions and provide alternatives and suggestions in areas where I think they fall short. Most comments I make are pedantic but I feel little things incrementally add up in the long run.
Without further ado, let’s get started. This is how the game opens right after the loading screen.
Did you catch the first mechanic that was introduced?
I hadn’t because it was too fast and I wasn’t paying attention. One thing to note is that during the loading screens peoples attention stray, we check Instagram, think about dinner etc. Putting something time sensitive that you want your players to see right after a loading is not good practice.
The first resting scene is crucial. What I am referring to is the screen that stays put when the game opens and waits for further input. In the first resting scene, the player should be safe, with no immediate threats and the game should give the player some space to explore the controls.
Splasher does give us the safe space, but also has that huge shooting turret. If you caught what happened in the video up above you can understand the mechanic of the turret shooting the gears which in turn opens the door. I, however, when I played the game didn’t. Thus, I was somewhat confused by the arbitrary turret shooting the green gear like object.
I move forward when I am ready and see the tooltip telling us how to jump. It is done relatively well, there is no text and the tooltip doesn’t break the flow of the game by taking control away. It is very important that the games avoid taking away control wherever possible whether this comes in the form of tooltip pop-ups, cutscenes or unopenable doors only the companion npc can open.
The lowering of the platform gives a sense of dynamism which I like. I go ahead and test my jump height in that obstacle and understand that I need to use the platform. The obstacle hight is ALMOST the same as my vertical jump. In a vertical jump, this is not a problem because there is no skill involved. You keep pressing the button, see that your character cannot reach the edge and move on. However, as a general rule of thumb, it is better to make obstacles significantly bigger than the jump radius. This ensures that people read certain obstacles as “inaccessible” and don’t see it as a test of skill. If a gap is 1 pixel wider than the maximum jump distance, several of your players will spend minutes trying to make the perfect jump, under the assumption that the jump can be made.
I take note of the jump capabilities of my character and move on. I notice two things: The person hanging upside down and the red goo.
Red is usually used to represent danger and people tend to avoid it. In this case, however, I decide to touch it to see what happens, mainly prompted by my prior knowledge of this game through the art and other contextual clues. Great, it slows me down and reduces my jump radius. I make a mental note to avoid it. This avoidance idea is also somewhat reinforced by the person I saw trapped in the goo just seconds ago.
Now that I played the game I know that the red goo is not an obstacle but rather a mechanic that allows us to attach to other surfaces, however, the game does not do a good job introducing it.
The first time the red good is introduced the flow of the game moves us away from it.
In reality, what the game wants us to do is to learn the goo is sticky and then use that knowledge to go back to help the person hanging upside down.
This implementation has a few problems. First of all, the game doesn’t force the usage of the new mechanic and actually lays down a path forward that avoids it. Second, the game’s intent of us helping the poor fella is not communicated well. The player has to break their momentum to go back to “help” him. This kills the flow of the level. Furthermore, when the player moves to the second platform the target is already out of the frame, so our attention is not primarily on target and the game doesn’t make it obvious that we can actually save the guy.
The first level of Mario does this incredibly well. When you pop Super Mushroom out of the brick even if you try to jump over it, you will be knocked right into it thanks to the clever level design. Funneling your players to go through the interactions you want them to experience is very useful in tailoring what the player learns about the game and when they do so.
Alright, after learning that red means sticky and saving the poor fella, I keep moving forward until I face the same turret that we saw earlier. It is shooting. Usually, in games, you avoid bullets. So I try to jump from my vantage point, trying to avoid the blue shots and squeeze in through the main body of the turret.
I do not get killed by the bullets! But instead, die when I came in contact with the thing that now I register as an enemy. The enemy introduction is alright, it has a spike which justifies my death. It also has a shaded tooltip showing an explosion, while I didn’t register it in my playthrough, it is nice to have.
After learning that enemies are bad and blue shots are not that bad, I try to tease out the way the enemy operates.
I figure out that proximity doesn’t trigger any reaction, but rather, going into their field of view forces them to run away. I recognize that it gets clipped by the turret and then get utterly destroyed by it. Aha, the blue droplets, water, hurts the enemies but not me. One thing I would have done differently would have been to ensure that the player sees the fact that the enemy doesn’t walk out of the platforms. While it is almost guaranteed that players will come to that conclusion that this enemy doesn’t walk off platforms, it would have been better to design the level such that the focus of the player never leaves the concept that the game is trying to introduce, in this case, the enemy.
I keep on moving forward and pick up another mechanic.
This mechanic is actually the same as the one I missed in the opening scene of the game. In here it is delivered much much better. We can see the mechanic in action not once, but rather in a cyclical nature. More importantly, the player has to use the mechanic to progress. This ensures that the player makes the connection “Water bullets make things move.” unlike the earlier attempts. When it comes to writing advice “Show, don’t tell.” is very common. Similarly, albeit a lot less elegant, “Make the players use the mechanic, don’t show.” is apt for game design.
I keep moving forward and get our first checkpoint and see another poor soul to be saved (or more bonus points). Immediately afterward we encounter our second enemy: A bull-like creature that immediately charges at us. The design of the enemy does a good job telegraphing what it would act like.
When I jump over the enemy it skids significantly before changing direction. This and my previous knowledge of bull like enemies push me to try and drop him into the toxic wastes. I am successful, the red goo machine activates and paints my way up.
What is missing from this video is that the death of enemy releasing the prisoner. The fact that there was toxic sludge on both sides makes it possible for the players to miss the unlocking of the cage.
The expected way to deal with the enemy probably goes something like this:
The skilled player jumps over the enemy, the enemy falls in, voila colors and bonus points. I am certain that almost all the developers while playtesting this level did this exact sequence and failed to capture the possible pain point. Another reminder that it is extremely important to test with people who know nothing about your game, as much as possible. Also, to this day I don’t know if the blinking of the enemy is an attempt to make a connection with the blinking of the lock icon. Possibly, I definitely did not read it as such while playing the game first time around. [Note from the future: After having played the game some more, it does seem to be the case.]
I move forward a bit more and get greeted by the timid enemy we had seen earlier. I scare them into running towards their doom.
Did you catch that their death unlocked the padlock? I had not. Maybe I am not the sharpest tool in the shed but it would have benefited this enemy – padlock connection if the turrets, and by effect, the demise of the enemies were much closer to the door. Ensuring that the connected concepts are both within the focus area of the player is essential, otherwise, the connection can be missed. If putting the concepts in the same focus area is not possible, it is the designer’s task to guide the attention of the player from one concept to the other. Doing so, in this case, would have allowed my attention to encapsulate both the enemy deaths and the changes on the padlock, which I initially had missed.
Getting into the groove the red sticky goo, I climb up another wall and rush my way through the ceiling…
…and promptly miss another mechanic being introduced. I am talking about the yellow goo. Once again, the fırst level isn’t eager to introduce the new mechanics by forcing the player to use them. Rather, the game just sprinkles them around.
Another interesting ordering decision is where I learn that water erases other types of goo.
This time, the game teaches me by forcing me to react to this knowledge. Thankfully, I manage to keep on climbing. After I get past the obstacle, the game shows me how the mechanic works, in a non-intrusive fashion. While it was nice that I was never under real threat (aka, there were no spikes) the ordering is curious. It would make much more sense to show the interaction and then ask the player to use it. One can make the argument saying that It feels better to react the mechanic (as I managed to do) but then why would you show it again, in a way that the player doesn’t use? I think the better decision would have been either to switch the ordering or remove the turrets that spray the ground.
Moving forward now we get to the real introduction of the yellow goo.
This one is done in a proper fashion, introduced in a safe environment requiring the player to use it to progress. Even adding in some kinesthetic joyride in the form of the tunnel which feels really good.
After the joyride, I come across the same puzzle and I go about solving it in the same way
and immediately die. Asking your players to react fast is definitely not a bad thing. However, it is important to ensure that the deaths don’t feel cheap. If the player attributes their failure to an external cause, something outside their control, this will annoy them significantly. In, this clip you can see that only after you start your last jump the circular saw enters the field of view. Which means the player has very little time to react. Furthermore, the obstacle is right in the flow path of the game, which can teach some players to not give into the flow of the game. I haven’t played enough to see if this is a theme throughout the game. Different players have different reactions to these types of “out of nowhere” deaths, but I would recommend against them in general due to them feeling cheap.
Moving forward we get greeted by a short cutscene that introduces the black holes. I jump right in hungry for more points.
In the “bonus level” we encounter the same enemies we dealt with before. I can easily outmaneuver them and it feels satisfying. While I am jumping around dodging bulls I miss the newest mechanic that is introduced.
When I destroy the second enemy I am greeted with a LOT of particle effects.
To be honest, I was a little bit overwhelmed and could not parse everything that was going on. I saw the prisoner drop down that I knew I wanted to save. I saw dozens of yellow bonus points rushing to my character. I saw the turrets activate and spew water everywhere, spawning even more yellow particles. It was difficult to track where they were coming from. While the scene looked amazing it is important to remember that too much juice can obscure the information you are trying to convey. Through the confusion, I gathered that if you put water on the yellow bouncy goo it turns into bonus points!
And that was wrong. The yellow “sparkly” goo was different than the yellow “bouncy” goo. In this bonus level, with added screen distortions and the pressure of dodging the enemy, it really was not the time to introduce the sparkly bonus goo. Furthermore using the same coloring and adding sparkles on top I think is a poor decision. Even further, giving no place for me to touch the sparkly goo made it easier for me to recognize it as the bouncy one.
I leave the bonus level using the exit portal that just appeared. Moving a little bit more I get introduced to the main mechanic of the game, our very own water gun!
I think it is a rather nice introduction. You can miss it, it has a meaty cutscene attached to it, and you immediately learn what button to press. What I dislike, however, is using the enemies as content checkpoints. When you kill the enemy several things activate. This teaches the player that they should kill all the enemies and if they are failing at a puzzle the key might lie in the death of an opponent. The uncertainty of which enemies being linked to what activation reduces the feeling of agency and activates a brute force method of “Gotta Kill Anything That Moves Otherwise Might Miss Content”.
With my newly reinforced bloodlust, I move on to see a gear that now I can activate.
I shoot the gear and I am a bit surprised at the outcome. While the shaded background does show a connection between the gear and ground, I didn’t expect it to move. This goes to show that it is important to be diligent in making sure the game is readable and the mechanically important objects are separated from the background.
At this point, I have learned that I should shoot to kill and I take the elevator up, sniping helpless enemies on the way, towards the end of the level. When I reach the end, I am greeted by the final bonus prisoner that I need to save. I understand that the game asks me to shoot the container until it fills up.
Unfortunately, I either missed an enemy or didn’t clean up some of the sparkly yellow goo off the ground so I fail to save the final prisoner. Sorry, Bob. I tell Bob that I will be back and take off to find the final 50 points that will grant him his freedom.
I am unfortunately faced with a concrete wall that I can not pass. And this annoys me. If I want to get a “perfect score” I would need to restart the whole level. What is annoying isn’t their requirement of doing it in one go, but rather, their lack of commitment to the idea mechanically. Whenever you grab points you can actually see how many there is in the level. It was 700 for the starter level. Telling us how much the maximum is, but also not allowing us to go back gives the player different signals. As in, if I just not stepped through the final door, I could have scoured the land for the remaining few sparkles. And I knew that there were a few remaining sparkles, so the final door is not a deterrent of a playstyle but one that forces people to mistrust every single direction pathway.
I didn’t play the level again to fully complete it.
Overall I actually really enjoyed Spasher and intend to play it more. I know the game has much more to offer in the coming levels but doing this “blank slate” reading of the first level allowed me to remember what a first-time player feels like. Here is a summary of my findings:
- Spend time thinking how you are introducing your mechanics. Asking your players to use them in a safe environment is much more effective than just showing them in action.
- Understand where the player is paying attention on a given screen. If possible keep the connected mechanics and concepts always in the field of view of the player. Make sure the connections between concepts are reinforced.
- Make sure your design decisions don’t go against each other. If you are going to block the player from going back after a point, remove the signposting that could allow them to make that decision before the block.
- Playtest with people who have never played your game. If they don’t play many games overall even better!
- Avoid cheap deaths by making sure the player sees them coming.
- Do not put anything that disappears right after the loading screen. Your player most likely will not be paying attention to the screen while the game is loading.
If you are interested in more I would recommend these resources on the topic:
- Extra Credits – Tutorial 101
- GDC 2014 – How to Make Great Game Tutorials
- Mark Brown – Half-Life 2’s Invisible Tutorial
- Mark Brown – Super Mario 3D World’s 4 Step Level Design
Thank you for reading. I hope some of the things you read you disagree with and would feel compelled to share your perspective.
I am still trying to find the optimal format. I am especially curious what you think about the short clip format and the orange highlight of the “outcome” texts. Please do give me your harshest criticism and thank you for reading.