Flame Sword #19: Sizzle Reel



With all the level scraps from showing off on Twitter, I took the 10 segments and divided them into two levels. The first felt scrap level felt obvious with a vertical theme called “Everclimb” and the second”Conveyor Central” poured the new conveyor belt segments into it. I worked on Conveyor Central’s 5 segments first and let me point out that to test these level segments, I make a hub of five transporters to quickly take me to each segment. Then the smoothing process happens. Constantly playing each segment and smoothing out the issues.


Once I have something that’s enjoyable in one form or another, I speed run each and establish how long it takes me to quickly get through each segment. In the case of Conveyor Central it was 7 minutes, which is long for any level to be a speed run. If it takes me 7 minutes, it can take a first time player 21. I cut it down to four segments and then dumped that fifth segment into a third scrap level to be dealt with later. This reduced the time to 5 minutes and I added conveyor belts to segments that didn’t have any.  This gives the level more cohesion to the theme that this is testing out the conveyor belts.

Other demo levels in the game also have conveyor belts and they’re some of the grandest segments with wide open spaces, enemies falling from the sky, and switches that control the flow of the conveyor belts. They are actually more fun and less of a challenge than these segments made for Twitter. That’s why at the end of forging these four segments into a level, I made this a secret level. It’s more challenging and less fun, even if it’s far less of a pain than it originally was. Its hardly a success story and needs improvement. After the segments were done, I added some rest zones where the player can stop, rest, and hit a checkpoint if they want. Checkpoints are optional and I try to put them slightly out of the way where you’ll spend at least a few seconds extra to hit them. The earlier demo levels have abundant checkpoints front and center that can easily be hit.


Everclimb was the next level to get focus. It already felt good, but climbing vertical levels is always a greater task for the player. A lot of the Twitter dazzle came from a lot of enemies, so I removed a lot of enemies for the easy mode. Oh and I should mention that there is no longer an easy difficulty, just a normal difficulty. It was confusing testers that medium was pretty hard and they didn’t want to play easy since they felt it would be easy.

Back to the level itself, there are plenty of horizontal segments the player can run through to enjoy the speed and thrill of simple jumping on moving platforms. When the five segments were soothed out and streamlined, it came to a whopping 8 minutes, which again is way too much for a speed run. Two of the segments felt out of place for a level named Everclimb, so it was easy to cut and thrown into the third scrap level. With the level under five minutes, I added some rest spots with checkpoints and included a rest spot half way up the final vertical segment for players who explore or notice the doorway to the right.

The second scrap level was forged into something enjoyable, even if half of it involved climbing. It was far more fun than the low points of Conveyor Central. I suppose it’s more fun to run and climb than it is to wait on conveyor belts while killing enemies and giant boxes.


There’s still one last scrap level to assemble into something enjoyable and cohesive. The game doesn’t really need more demo levels as there haven’t been new mechanics to keep the game fresh. So I might just add these three segments into the other levels as secret areas. This will help give a special feeling to grander secret areas even if players might feel let down when there’s nothing special at the end.


I highly recommend every week or two that any developer get forced to replay their entire game. Oh and you have to record it and watch it again to get refreshed of issues and how things look. It’s easy to find a glitch or issue and work on it right away, but if you do that for each when they happen, the workflow seems like a lot less flow. Playing the entirety of a game probably doesn’t work for longer games that are 40 – 100+ hours of course.

Replaying everything helps to know what should be added and more importantly what should be cut. During my last play through it had quite a lot of new bugs. What you know to work on one level might not work on another level. A new issue for me is with the camera zoomed in close for some areas, simple, safe jumps can turn into blind jumps so there needs to be incentive or some sign of safety below. It’s easy to just add  a platform a bit lower to let the player know it’s safe.

Off the top of my head, here are some glitched things fixed:
 – Enemies go to sleep when far enough off screen, which caused an issue where enemies would hang in the air and you could see their feet.
 – Lots of conveyor belts caused a slow down.
 – Conveyor belt seams were glitched like entire conveyors that were the right edge of the conveyor belt instead of left, center, right.
 – Missing elements like guides that tell platforms to speed up or slow down.
 – Areas that were too far to jump, but that’s why there’s a debug for a mega jump for situations like this.
 – Some secret areas had visible seams or plants hanging out from beneath them.
 – Leaving a room and returning within a second kept the enemies and respawned new ones (been a problem for a while)

Having to play through 60 minutes of the game again, I cut some things that didn’t feel good. This game is all about the feel of things. You would think that I’ll get better at playing my own game so many times, but it’s more about feeling when my joy to play starts to sag and finding redundancies. I have constructed a lot of new segments for Twitter and they all start to feel the same, so I’ll tweak them and cut out redundant parts and replace them with something more creative. It trades a bit of intensity for interesting areas that feel fresh. It’s tough feeling fresh playing a game for a full hour that has a single environment. Also finding places that feel cramped and opening them up to breathe to be more enjoyable.


After my glitch filled 60 minute play through, I played the game again the next day once the glitches were fixed and it was a far smoother, less glitchy play through. The second play through shaved 10 minutes off the entire thing. With the second play through, I took the recorded footage and turned it into a 2 minute sizzle reel to show off how high octane the game can be. The game still has mundane moments, but the action and craziness will really show it off as a spectacle that’s fun to watch and play. The sizzle reel started off this post, because it’s a doozy and I don’t think anyone will read this far.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s