My new fresh sites are getting CPU timeouts.

We currently run [andbounds.com](https://andbounds.com) on Deno deploy using an oak based system and brute force client side js. Now I rebuilt everything to work in fresh. You can see it at ~[fresh.andbounds.com](https://fresh.andbounds.com)~ (edit: now live at andbounds.com). I also have another fresh site at [agent.andbounds.com](https://agent.andbounds.com) that is supposed to support an iOS app. They both experience 502 errors and CPU timeouts regularly. It makes the sites very hard to use, I was expecting to migrate to this new fresh version, but random 500's and 7 second loads for client side js is not gonna work. I see this error more in places with dependencies like three js, but it can strike anywhere.

I would be happy to let you all take a look at the source of both sites, perhaps I have done something wrong.

Or maybe we need pro plus option where we get a little more CPU time?

Start by reproducing the 502 errors, CPU timeouts, and seven-second client-side JavaScript loads on andbounds.com and agent.andbounds.com, comparing behavior around the reported three.js dependencies. Review the source for both Fresh sites if it becomes available; done means the cause is identified and the sites no longer fail or regularly exceed the CPU limit.

Written by the indexing model from the issue text.