Documentation Gap
I haven't given a blog update in the last few weeks. Part of it was laziness - not in doing projects, but in documenting them. I regret not writing things down because now I have to piece together what happened from memory. The hardest part of documenting isn't writing, it's consistency.
The Hardware Reality Check
The biggest decision was putting anotherRUN on pause. The reason is hardware constraints. My machine is from 2012 and it's really starting to drag down productivity. I don't want the project to lose momentum just because I'm constantly fighting performance issues.
Long compile times, editor lag, constant performance bottlenecks—it's like trying to run a marathon with weights strapped to both ankles. I know forcing through would burn me out and possibly kill the project altogether. Rather than letting frustration bury it, I decided to pause and explore smaller projects that are both achievable and good learning opportunities.
Strategic Decision
Sometimes the best move is stepping back. Rather than letting hardware constraints kill a promising project, I'm keeping anotherRUN safe for when I have a stronger setup. This preserves the vision while allowing continued learning through smaller experiments.
Glow Playground: A Case Study in Frustration
One of the replacement projects was Glow Playground. The concept was to create customizable neon-style backgrounds for Wallpaper Engine, where text emits realistic light into the scene. I had a working vertical slice: neon text glowing against a brick wall.
But then I hit two critical problems:
The Frustration Reset
I only committed locally, not online. After hours of switching back and forth between URP and HDRP, constantly tweaking settings, and fighting with the limits of my old machine, I lost patience and deleted the project. It wasn't corruption or a crash—it was me pulling the plug out of frustration.
The regret is that without pushing online, everything was gone once I hit delete. That experience hammered home the need to push more often, even when I'm not confident about the state of the work.
URP vs HDRP Technical Dilemma
- Pros: Decent results using point lights, better performance
- Cons: Doesn't provide true global illumination
- Use case: Suitable for wallpaper performance requirements
- Pros: Could solve lighting delay and shadow sync issues
- Cons: Far too heavy for a wallpaper application
- Use case: Overkill for this project's requirements
I even tried creating "fidelity" and "performance" modes by varying the number of centroid lights (5 for fidelity, 1 for performance). It looked promising, but the core question remained: URP or HDRP?
Visual comparison of rendering quality vs performance overhead for wallpaper applications
Multi-Project Development Strategy
Another major development was starting my website build. I'm now on day three of development, and I've discovered something interesting about parallel AI assistance.
Claude Code is helping me with the website structure, while Codex assists with Voice Chain Reaction. It feels like I can split my workload and let different LLMs handle different projects simultaneously. That's actually been working out better than I expected.
AI Development Workflow
Using different AI partners for separate projects feels like managing a small studio, except my teammates are language models. This allows me to juggle multiple tracks at once in a way I couldn't before. Each AI seems to maintain better context when focused on a single project rather than switching between multiple codebases.
Voice Chain Reaction: New Direction
While Glow Playground stalled, I pivoted to a completely different idea: Voice Chain Reaction. The premise is silly but fresh—plushie enemies charge your base, and you stop them by making animal sounds into the microphone.
It's both chaotic and surprisingly hard to balance. This project is still in early brainstorming and technical setup, so there will be dedicated future blogs covering its development in detail.
Plushie enemies → Voice detection → Defense mechanics → Scoring system
Personal Challenges
On the personal side, I visited Victoria University in Melbourne today. The campus looked great and I honestly regret not choosing it for my Master's, though they didn't offer my course back then.
The main reason I was there was for my PTE exam. It was harder than expected, and I didn't study enough. Partly because I was busy with projects, partly because I was lazy. I really hope I pass since retaking it would be a serious financial strain.
Reflection on Priorities
I kept spending time on projects instead of PTE practice, which was partly laziness and partly misplaced priorities. It's a reminder that sometimes the urgent needs to take precedence over the interesting, even when the interesting feels more rewarding in the moment.
Lessons Learned
This gap in documentation taught me several important lessons:
- Commit and Push Frequently - Even incomplete work should be backed up online
- Document Consistently - Every skipped log means details get lost and progress threads unravel
- Hardware Constraints are Real - Sometimes the best technical decision is a strategic pause
- Frustration Management - Step away before making irreversible decisions
- Priority Balance - Important but boring tasks (like exam prep) can't be indefinitely deferred
Current Status
So that's where things stand:
- anotherRUN - Paused but not abandoned, waiting for better hardware
- Glow Playground - Shelved until I make clear rendering pipeline decisions
- Voice Chain Reaction - Taking shape as a new experimental direction
- Website - Actively developing with Claude Code assistance
It's messy, but it's progress. The key insight is that progress doesn't always look like forward motion—sometimes it looks like strategic pivots, technological compromises, and honest assessments of current constraints.
Looking Forward
The goal isn't to eliminate these challenges but to work within them more intelligently. Better documentation habits, smarter version control, and honest hardware planning will prevent similar frustrations in future projects.