My levels are totally shit.
At least that's how it used to be. Take a look at my oldest levels for that matter. Mostly, they are bad. I didn't care much for other people's opinions, I thought they were good
Some of them did turn out better than others and people did enjoy them.
Even through MBP I can tell you it was basically bad, and only very few levels ever got positive feedback. With PlatinumQuest, the staff were pretty harsh on each other and I got a lot of bad feedback, but some positives too. It was only when Phil told me that all my levels were flat and didn't have height differences that something clicked, and it changed a lot of things for me.
Some of my inspirations nowadays come from either my brain coming with a unique thing (queue HiGuy groaning) or seeing what others made and implementing the basic ideas in my own level and create something new altogether. I always have some cool ideas I'd like to do but are, at the moment, completely out of reach due to either the tools I have, the time required to input or having a more "thorough" understanding of what I want in the level and how it should be built.
When things did click together though, I pretty much knew what I wanted to do. I could sketch things and have better visualisation, but a lot were just stored in my head until they were used.
You might have played PQ's Vibrancy Grounds, and that is probably the level I spent the most time on. The design came from MBAdvanced' Color Twist.
This is how the original looked, and the base inspiration:
This is my bigger design of this level. Added some more stuff and expanded it, and added some PQ features to it. It played okay which at first we accepted it for, but it was pretty bad and boring and eventually scrapped out.
Afterwards I decided that it needed a proper remake. The garden design took the longest to make, followed by construction and the top part. I didn't have anything sketched but I wanted to make something unique and cool. Add in key scenery design, lots of gameplay and balancing, you end up with easily one of my top levels of all time.
Another triple-remake is the very first PlatinumQuest level, Training Wheels. Originally called 'To the Finish', the idea was basic movement but add in some pitfall challenge. The original Tutorial level designs were based on the fact you couldn't go Out of Bounds (supported by colmesh and the trim to be key-design aspect).
Eventually we settled on the fact we needed something better and didn't want to go with the direction of 30 levels to introduce basics. I came up with the 'learning to roll' level design coupled with powerups, but also advanced marbling techniques. I wanted a completely unique level design in the basic movement that no other level has done: teach you the Forward/Backwards/Right/Left/Jump one by one so it can be beaten without camera movement. Advanced marbling techniques and powerups followed.
The idea was pretty good, but last year the entire staff went hard at that level stating it wasn't a proper novice-teaching level, too hard, and too confusing. I eventually crumbled under the pressure (read: literally everyone was against me) and agreed to remake the level without the split-path combination, and forgo the advanced marbling techniques. I then decided to take the 'novice' direction further by making it much harder to go Out of Bounds (unless it was part of the level design).
I also wanted to add in a gravity design, so I made a unique gravity-changing path on the get go and taught the player about the gravity modifiers later on. Threefolder added in the smooth slopes and curves that QuArK can't do, and with more feedback from staff and testers, the end version was born.
As the level's General Hint says: "Would you believe me if I said this was the third revision for this tutorial level? It keeps getting better every time though!"
And it does.
I can go with more level that have had revisions or have had unique designs, but then we'll have spoilers
So, that was some of my story. Now the tasty tips:
- Know when to end a level. I know it seems great to keep pushing a level longer and longer, but eventually you just run out of ideas, and you keep forcing yourself to add more. This is actually noticeable in the level, and it turns a good level sour.
I used this very tip on my own levels, especially a gigantic one that I just cut off at a point that didn't seem to make sense, but unless you were me you thought it was intended and was a good ending. I simply ran out of ideas and didn't want to make it crap.
- Quantity vs Quality. Back in the old days, quantity > quality, but to a degree. Look at Henry's levels. He contacted us years later telling us how his duo 100-level packs are absolute cringe. Starting a few years ago, quality > quantity. People appreciated a level that has had more effort put into it and was fun over a level that was long, bland and boring. Good texturing didn't help at all.
Some of the best levels in the past are now frowned upon. That's how times changed.
- Focus on feedback. There aren't a lot of people here, so the vocal ones are also the ones you see more often. Still, if you can join a group conversation and ask for feedback, it will help you a lot.
- Simple, yet effective. Some of the best levels are those that don't add anything new to the table. They don't use new features, or challenges you haven't met before. Their design aspect make them fun though!
- Short and small challenges. A trend that started in 2010 was that a level would be made up of several small challenges, each taking 3-5 seconds at most, and rarely you'll go to 5-10 seconds. You would then see a level made up of easily 15-20 challenges that flowed smoothly, were easy and made sense. A lot of outstanding levels in 2010-2012 era were made just based on that trend, and it features prominently in PlatinumQuest levels as well.
It made for good level design that was simple, refreshing and didn't annoy players.
- Learn from the worst and the best. In the olden days, Pablo, Phil, Andrew, Oaky were some of the shining stars of this community. Their feedback gave a lot of weight as their levels were rated the highest. I took Threefolder to PlatinumQuest by playing his levels as well, and seeing the refreshing creativity and design he came up with to make some seriously fun levels. I noted his hunt-based levels were horrible, and put him off of those in PlatinumQuest as well. Only one of them actually got a thumbs up from me.
Learn what they did right, what they did wrong, and learn from their mistakes and achievements. I did that and got some of the best levels I could ever hope to do.
Likewise, I've learned from the worst level designers and the worst levels I ever played. I learned what not to do and the community feedback on those levels to see how I can improve and not repeat such... offenses.
- Criticism sucks. I hate to get bad feedback just like everyone else and would defend my choices to the death. However, it is also wise to accept some of it and implement it into level design change. The above two levels are important lessons that came through harsh criticism: everyone was against me, and I (eventually) responded by accepting defeat and redoing. TWheels especially was a nightmare and what started as a "I hate you all" redesign ended up as a well-loved welcoming level.
It is amazing how much we fight for what we believe should be how things are. PlatinumQuest's original design is quite different from where it ended up, and I fought a lot to have some of the original aspects remain. I lost, but the end design is better. Sometimes it is better to listen.
- Find the way you work best. Some like sketching things out, some like thinking on their head. The above people do their level design in completely different ways, and you might find that some of the stuff they do applies to you, and some does not. That's fine. Be yourself.
Off on a tangent: it's the same for me at work. We have a regular 4-people team that, pending stress and need, has half of the lab available to reinforce. Every person who is qualified for the team has his/her own way of working, and new people who get training are often told to learn from everyone and end up make their own style of work that fits them the best. Guess what? Anyone who is qualified really does have their own style of work that fits them the best.
- Know how to reuse, but never ever remake. I won't lie that I haven't reused what other people made. I outright stolen ideas and came up with my own design for those, but also had to make things wildly change. You can't just copy and paste. Remakes don't work either and 99% of the time, remake-based levels were some of the worst I've seen. VGrounds was no exception, and needed a serious makeover with whole new design and challenges ideas to become something new and beautiful.
- One idea. Some levels have a recurring theme to them. Gyrocopter Monster Course started its own trend as well as those Skill Course based levels. I won't lie... a lot aren't good. It takes really smart thinking to come up with a theme to a level that is recurring and maintain itself throughout. You'll have to be extra-creative!
- Time. The less you spend on the level, the more it shows. You can spend very little on a level as long as it's short and fun, or you can spend a solid 36 hours over a few weeks perfecting every single key aspect of it. VGrounds is both the former and the latter. Except it wasn't fun, just 2 hours of QuArK.
I can personally tell you that Threefolder, Andrew, Phil, Oaky, Pablo and many of the greatest level designers all spent a few days or weeks coming up with their creations, and would do hours at a time in each session to ensure their level met the highest criteria.
- Bigger eyes only with experience. Start with small levels, and go bigger and more detailed the more experience you get. If you try and do what Uil does (as RDs showed), you better have the thousands of hours of experience to back it up. That said, not everything will work out. Accept that.
- Singleplayer Vs. Multiplayer. One of the least understood things is that making a level that is for singleplayer is wildly different than making one that has two or more people in mind. This includes the level layout, challenges, content (scenery) and more. Co-Op levels also differ. When you make the level, know where you aim.
There are more, but I should shut up now
Enjoy!
(side note: I just realised I've been making levels for
10 years... damn.)
Edit: just want to mention I often see myself less than most others here. I don't have the creativity they have, so I often spend more time in constructing my levels and coming up with far more unique ideas than what normally is. I have an uphill battle every time, and it's something I made a goal to beat every time. Gained a lot of experience fighting such battles over and over.