Cohost's final week of existence has, for me, primarily been spent working to re-implement debug rendering functionality in Psychonauts.

All the lines drawn here (not the text) are handled by a class which has all of its functionality stubbed out in the release build of the game. That is to say, it should be impossible for this screenshot to have been taken from a release build of the game. And yet, it was.

Since the Linux build of the game still exists, and most of the calls to these functions remain intact somehow, I'm able to hook into those stubs and recreate the functionality.,

There's still a bunch of line-based primitives I need to reimplement, and then I need to reimplement debug tri rendering, but getting something drawing here is a huge milestone. I had a couple-hour long headache during this that turned out to be me completely misunderstanding the layout of the game's main GameApp singleton.

Alas, Cohost will have passed away by the time I get around to writing a full blog post on this topic. That... really stings. And makes me sad. I'll be talking more about post-Cohost stuff later today, probably.

For a serious answer to this question - I don't think it really is but I do think it's misunderstood and underrated. The main concept of the game isn't really like anything else I've played.

The localisation in questionable, the story is filled with cryptic keywords, the story missions are for the most part not great (except for the final boss)...

But the core game is pretty novel and a lot of fun. Only really lacking in variety. It's not even like it's "not a fighting game" - I mean, it's definitely not one in the traditional sense but it does have 1 on 1 combat sometimes with actual fighting game combos and stuff still.

It reminds me of Brutal Legend in both gameplay and reputation. The gameplay similarities aren't too deep (but there are quite a few, hell, even the aforementioned lack of variety is a shared issue) but I think both games have gotten a bit of a similar unfortunate mixed reception - because of its differences with the rest of the series in GG2's case and because of differences in EA's marketing vs. the actual game in Brutal Legend's case.

It is pretty funny that so much lore that sets the stage for GGXrd and Strive is tucked away in GG2 though.

Model by @xenonfossil

I wanted to give a quick status update on what I've been working on in regards to Psychonauts, since it's been a little while.

Brain Tumbler

A large chunk of my time spent working on Psychonauts stuff since last time was spent on working on a Blender addon (titled "Brain Tumbler") designed for use alongside Oatmeal to make working with Psychonauts-specific data in Blender easier. If you read my post on making custom levels, you'll know that "Brain Tumbler" was originally what evolved into Oatmeal. The original iteration was planned to be a full suite - importer, tools and exporter.

In this new design the responsibilities are split - Oatmeal is handling format conversion, leveraging my existing PsychoPortal library. Oatmeal embeds Psychonauts-specific data into glTF extra data but I can't well expect prospective world builders to edit that data by hand to set up flags and such. Thus, a Blender addon is necessary.

Brain Tumbler is that addon. Inside Blender, Brain Tumbler presents a nice UI for setting and viewing this extra data. It can import/export the extra data set up by Oatmeal (in a JSON chunk called "lipo_data") to be used during conversion.

This split responsibility approach is ideal and far, far better for me personally than previous plans. If Brain Tumbler had been a full suite then it'd have presented the problem of me essentially needing to either find a way to use a .NET module in Blender (I tried, it's not good) or entirely re-implement PsychoPortal in Python (not happening). Now, with Oatmeal using PsychoPortal for conversion, all I need is some glue code in Brain Tumbler to translate specific things to Blender.