First, I want to thank everyone who has supported GDSim with your downloads, reviews, and comments. It has now been downloaded over 4600 times!

A year has gone by, and it is past time to check in, as well as to clarify my intentions somewhat.

GDSIM
Unhappy news out of the way first: GDSim, in its form, is not going to be updated again. Some of you may be disappointed, and I apologize, but it is the right call.

I want to be crystal clear about this point: GDSim is my first-ever project of its kind, and it contains messes left over from everything I had learned while working on it, making it prohibitively challenging to expand upon, or to improve its poor performance. I am a novice in training, so I am grateful for your patience as I stubbornly do this the hard way, by myself.

Following wise advice, I have taken what I've learned with GDSim back to the drawing board. Although it is a setback in terms of content/features, it has enabled forward progress on core systems, benefiting from new ideas.

The main purpose for this public alpha was to seek feedback on the handling physics and control. From your kind compliments and reviews ♥️ I seem to have accomplished my goals there, close to how things should be. I now feel confident that players will enjoy the physics of a full game based on this handling model!

PROJECTED SCOPE
I have received requests for certain features or content, so I think I should settle some expectations in light of my inexperience and my plans.

Intentions:
- Singleplayer "career" mode (not just a sandbox)
- "Quick race" mode for one player or splitscreen multiplayer on the side
- Sim-oriented handling, with details hand-picked based on my own ideas about sim gameplay design
- Cars are likely to be street-legal vehicles inspired by models built before 2000
- Maps are likely to be rural/natural/"rally", not very "civilized"
- Something akin to the autocross course importer
- More customization and upgrades, eventually

Not interested / lacking experience:
- High-level AI (racing on difficult courses or in a large pack) -- This will be a steep challenge, and I cannot have a game depend upon it if I can't make it work well
- Police chases, NPC traffic -- Streetracers evading cops is not what I have in mind; it's also more AI programming
- Collision/visual damage -- I want to encourage risk-taking, and discourage recklessness in other ways
- Network multiplayer -- Maybe in the future; I have to learn to walk before I can fly

CHANGES AND PROGRESS
The next thing I release is going to look different from GDSim. I am still brainstorming and experimenting with elements of how I want to render things, but I have determined that my efforts call for more stylized graphics, cutting down on fidelity to reduce the workload and reduce overhead.

I have spent this past year studying 3D rendering, shader programming, and Godot Engine to correct mistakes I made in GDSim, to improve performance, to accommodate larger environments, and to come up with solutions to make creating & adding content easier.

The new unnamed prototype already boasts significant improvements to the wheel/tire model. The new model can derive a surface normal from any arbitrary direction around the wheel, enabling a hardcore 4x4-style vehicle to apply traction up the face of a wall, for example, and fixing edge case scenarios in GDSim where a wheel could make contact but traction could not be applied. The new model also enables wheels to interact more accurately with dynamic objects such as a moving platform or loose obstacles.

The new tire model also greatly improves the traction behavior of stopped wheels and therefore reduces the necessary physics tick rate, reducing CPU cost for physics.

Meanwhile, the lion's share of development work this year has been invested into a new custom solution for a dynamic 24-hour background. It will have a greater base overhead than GDSim's sky with zero time progression, but it updates smoothly on the GPU and looks much nicer. I plan to continue offering some manner of resolution scaling that will improve performance for slower devices; an exponential benefit for the raycasting I'm using to draw the sky.

THE FUTURE
Even if I avoid biting off more than I can chew, this project will still take a long time, so I will see about shifting to more regular devlogs instead of pushing for a schedule of updates I can't keep.

I am also considering taking on a much smaller project (or two) to gain more experience before slogging all the way through a big one. If I do that, I would confine myself to 2.5D with SNES-Mode-7-ish graphics. Please comment below if you love that idea based on your impression of my design sense with GDSim!