Claymates!

Welcome back everyone! Thank you for joining; it’s great to see you. My latest speedrun project just wrapped up, so grab a coffee or a tea or a cookie, get comfy, and I’ll tell you all about how it went!

TLDR

If you’re just here for the headlines, Project Claymates was very successful! I got the world record in both of the existing speedrun categories of the game. You can check out the two runs here:

Here’s a snapshot from my speedrun profile:

The other thing to mention in the TLDR is that I became a Twitch affiliate during this Claymates Project. That means I am now able to make advertising revenue, have paid subscribers to my Twitch channel, and a bunch of other cool stuff. I’ll talk about that more below. So, let’s get into the details of how Claymates went down!

Game Selection

A lot of different factors weigh in on which speedrun project to choose to work on. Projects are somewhat long - my last two projects were 36 and 49 days long - so it’s important to be interested in the work ahead of me.

When it came to selecting Claymates, the #1 factor that drove my choice was that I wanted to be able to contribute to strategy-finding. Claymates is a lot less optimized than the previous projects I have worked on. By that I mean: since fewer people play Claymates than say, Mickey’s Magical Quest, less is known about Claymates’ inner workings. The strategies in the world record have undergone less stress-testing than the world record strategies found in other games. There has been less overall pushing of the envelope.

picture of the cart I did runs on

In Ninja Turtles, Zombies Ate My Neighbors, and Mickey’s Magical Quest, I came up with very few original strategies. I added or tweaked one or two really small things in Turtles. I added literally nothing at all in Mickey. Zombies had an original strategy here or there, and actually is a game where I think new research and strategies could make a big difference. In my personal Zombies experience though, I was simply too burnt out by the end of the project to put too much effort into independent research. Saving all the neighbors took everything I had at the time. As a reminder, I only ever completed one run of Zombies Ate My Neighbors 100% and it took 49 days to finally get it.

When I arrived on the Claymates scene, the Warpless category only had one run. That one run, also known as the world record, was an impressive 35:18 achieved by Error72 on May 1, 2016. Error72 performed Claymates warpless at Summer Games Done Quick later that year, which as of this writing remains the only time Claymates has ever appeared in a GDQ event.

Error72 did a fantastic job in his Claymates era. But, iron sharpens iron. Having stood unbeaten for 8 years and counting without competition, my intuition was that Claymates was a speedrun where I could dip my toe into routing and strategy finding for the first time, and have it be fruitful. That was the hook that brought me into Claymates.

The Lab 11/12 - 11/30 (19 days)

During the lab Erika and I made a raspberry lemon layer cake, and two separate lemon cheesecakes. LEMON. The cake, made right at the start of the lab, is the only one that got photographed.

The lab also covered Thanksgiving. Erika and I hosted four others for a six-person thanksgiving dinner. We made a whole turkey, potatoes stuffing - you know the deal. Again though, no photos. I’m off my photo game. I vow to start taking more photos.

Back to the lab though, after learning the first level, I thought this whole thing was going to be a breeze. Within a day of practice I was consistently beating the level 1 in the WR by 4-5 seconds. For reference, there are 18 levels.

The next few days painted a different picture, though. Levels 2-5 were all levels that, despite a lot of practice, I tended to do slower than the WR. As I continued on, I understood that this wouldn’t be an easy task at all. After learning 9 of the 18 levels, I estimated I would be around 9 second behind the WR, on a good run.

Level 10 was an opportunity. Lily Pods. Error72 takes a hit and lives with the time loss. Claymates is a long game to try and beat without taking any hits. But, here were 12 seconds up for grabs.

Level 11 tended to shade in Error’s favor. By the end of level 11, I estimated I’d be even with the world record on a good run. The late game is where I saw opportunity, though. Via a combination of new strategies and more aggressive play, I estimated I could play levels 12-18 a combined 40 seconds faster than the WR was able to.

With that landscape in front of me, I started to stream attempts.

Show time 12/1 - 12/11 (11 days)

Over these 11 days, I live stream Claymates attempts for 20 total hours. I walked away with a 24 second improvement to the world record, and a personal best time of 34:54. I was also able to improve the any% world record by 31 seconds, with a personal best time of 16:16.

In Claymates, there are two categories people speedrun - with and without warps. My main focus was speedrunning without warps, but I took a diversion to set a new world record for the with-warps category since beating the game with warps is a subset of beating the game without warps. There’s no new strategies to learn, so I did it sort of for completeness sake. The 34:54 in warpless is the run that means more to me, though.

All the while, the channel continues to grow! I mentioned at the top that the other major milestone reached during the Claymates Project is that my Twitch profile qualified for the affiliate program! Being an affiliate unlocks a lot of different creative options for the channel. First and foremost, I now have emotes! Here are my three emotes that you can use today on Twitch.

Longtime readers of the blog may recognize the ‘star’ of my emotes is the same main character from the video game I’m always working on but never finishing. For the non video gamesters, GL is Twitch slang for “good luck”, while GG is for “good game”. I will say, it is very cool to see this little character actually live and moving around on Twitch.

If you’ve been reading between the lines on this blog, you may have picked up on the fact that I’m starting to get back into game dev. Someday speedrunning and game dev will collide. They are two branches that are getting closer and closer to touching all the time. The emotes in a way felt like the beginning. The first cross-contamination. Needing emotes required me to dust off all my pixel art equipment and dive back into that work. Not surprisingly, after making the emotes, I stuck around and worked on some art for the game too. I’m not doing a ton of game dev work right now, but it’s showing signs of life. I’m not making any promises, but it’s something to keep an eye on.

I think that’s it from me. As usual, thank you for reading - it was great catching up. I am currently leaning toward my next speedgame being Adventure Island 3 for the NES. Stay tuned!

My updated resume