Domo Yoro

Why The Demo Was So Polished

Added 2023-08-17 01:12:36 +0000 UTC

The secret ingredient is playtesting!

The Playtest Process

For Soda-Powered Penguin, there’s an important point in the development of a level wherein I start the playtesting process: right after the first draft of a level is done. This is when I'll start reaching out to playtesters, one at a time, to schedule stream-calls on Discord. From there I'll watch them play, listen, and take notes on just about everything.

In playtesting, the most valuable data comes from watching a player’s actions. Seeing what approaches they take, where they get stuck, etc is more important than listening to their words.

The very first draft of a level is always unpolished, and only a proof of concept. Playtesters will have to navigate a janky, frustrating, duct-taped level with ugly, half-finished placeholder visuals. And I don’t feel bad about it! (Mostly because I’m there to explain the intent behind everything.)

This room in particular was frustrating for playtesters- even when I fixed the issue at the bottom of the screen.

Once they're done playing, I'll answer questions, ask them about their general thoughts and if anything stands out, and then thank them for their time. Afterwards I'll go back and narrow down my notes to a list of things that need to be done- make this part of the level more consistent, improve that mechanic, etc. Once I feel like the major issues of the previous build were addressed, I'll reach out to another playtester and start the process again.

These are probably difficult to understand- I didn’t intend on showing these notes to anyone else.

I'll repeat this as many times as I can- ideally, until there are no major issues with the level design or the mechanics. It’s important to revise the game before starting another playtesting session, as any major unsolved issues may overshadow other problems that you'll want found.

Problems With The Process

Of course, this process isn't without its flaws. Since I’m doing these playtests one-on-one, I can only reasonably do a handful of playtests before it stops being worth the time investment, or I can’t stand to watch another 2-hour playthrough.

This playtest was so long that the timer broke; it’s trying to display 1 hr + 102 minutes + 41.549 seconds, meaning that this was nearly a 3 hour playtest. It was very painful to be there for the whole thing.

It can also be easy to overcorrect for outliers- the mistakes of 2-3 players might get a room design thrown out entirely. Additionally, I likely do not have enough playtesters / can conduct enough playtest sessions to reasonably represent the spectrum of players that will play SPP.

As a side note, it’s really hard to not comment or answer any questions while watching a playtest- sometimes playtesters will pull off things I never imagined happening, or solve problems in really weird ways. If they’re stuck, I have to let them be stuck. I won’t be included with every copy of the game when it ships, so I need to make sure the game explains itself adequately enough.

Going From Playtesting To Feedback

The goal for these active playtests is to refine things enough that there are no major issues: there will definitely be bugs if you look for them, but otherwise, the level should be “good enough for now”.

This is when it’s best to expand the scope of playtesting, and focus more on feedback. Players will still find issues, but with all of the major problems solved, each player is likely to find different issues. Solving these issues is usually the last bit of polish the level needs. As it turns out, this is also the perfect opportunity for Patreon builds!

This is also why Patreon builds can’t be monthly; if I were to upload a build with known major issues, everyone would report the same problems- it’d be unproductive and a waste of time.

This is the part of the post where I give shoutouts to Thaw (Abst) for very thoroughly testing the game prior to the public demo. They found countless issues, exploits, and bugs- and I was able to fix many of them prior to the SAGE ‘22 demo.

The feedback process winds up being fairly close to quality assurance / bug testing, but extended a bit further. If there are issues with a specific room design or the controls, I’ll be sure to address those. Scheduling these fixes is always a bit of a challenge, since I have no way of knowing how many problems will be found, so I often schedule it as a secondary priority alongside something else. Or, if enough things stack up all at once, I’ll set aside a few weeks to tackle them.

So, I hope this paints a clear picture of how I tackle playtesting for SPP- and how the game managed to be so polished for the first public demo.