Sign in

Daniel P H Fox

@phfox.net
357 followers 10 following 49 posts

Recovering ex-Java programmer.

PostsRepliesMedia
Daniel P H Fox @phfox.net · 03/08/2026
My Windows desktop had its power supply unit crap out like a month ago, so I was forced to use my Linux laptop as my daily driver. I still haven't put my replacement power supply in my desktop. I love Linux now. it's just great
030
Daniel P H Fox @phfox.net · 05/07/2026
But also seriously. I just want a little local inference box that I can own and supports me in a human assisted workflow. I don't really want to cede all control to the robots.
010
Daniel P H Fox @phfox.net · 05/07/2026
I apologise in advance for not tokenmaxxing and shall repent for not glazing Claude
110
Daniel P H Fox @phfox.net · 05/07/2026
I'm not that excited about Claude Fable honestly. I don't want to spend way more money for what Sonnet 5 already does reasonably cromulently. Maybe as a thing that can break through on tougher issues only, perhaps. I want efficient cheap AI tools, not lumbering beasts that drain my wallet.
110
Daniel P H Fox @phfox.net · 27/06/2026
order of the sinking star is a pretty great puzzle game with some other confounding unfortunate circumstances
120
Daniel P H Fox @phfox.net · 21/06/2026
that said i would highly encourage people to go learn about arenas and pools and region based memory management, i found it useful as a tool and maybe you will too
010
Daniel P H Fox @phfox.net · 21/06/2026
it's perfectly valid to talk about things like whether RAII and lifetime annotations are _strictly_ necessary for acheiving safe programs or whatever, but altogether a different question when you're asking if it is better to write software with them. nobody knows! try it yourself
110
Daniel P H Fox @phfox.net · 21/06/2026
stop caring what other people think about what the right way to write software is if you're doing it wrong, you'll know, because it will suck more. make it suck less. that's all there is to it (this also means - don't foist your opinions onto others. what works for you is not what works for them)
120
Daniel P H Fox @phfox.net · 19/06/2026
1024x1024x1024 terrain generation distance on my desktop, in seconds it's more likely than you think
061
Daniel P H Fox @phfox.net · 19/06/2026
This was a beach to optimise
020
Daniel P H Fox @phfox.net · 13/06/2026
I, for one, know exactly whose software I will not ever be depending on...
020
Daniel P H Fox @phfox.net · 13/06/2026
You're free to embed your opinions and protests into your software. Make it catch on fire if it notices you have Java installed. Just don't be surprised when people think you're a terrible maintainer for doing that; freedom from liability isn't freedom from other people's ire
120
Daniel P H Fox @phfox.net · 13/06/2026
The jqwik incident really boils down to a few things: - open source projects are provided as-is and you can't hold the maintainer liable for breaking your stuff - however, a maintainer intentionally breaking your stuff when you're not looking is absolutely a violation of trust
130
Daniel P H Fox @phfox.net · 13/06/2026
fable: a short, fictitious story designed to teach a moral lesson.
020
Daniel P H Fox @phfox.net · 06/06/2026
Literally clankers vs wankers at this point
000
Daniel P H Fox @phfox.net · 06/06/2026
genuinely can't believe that people think it's OK to fantasise over threats of violence to the rsync maintainer over the most mild-ass LLM use. Whatever moral system you think you have, it's so incredibly fucked up, get help for the love of god please stop posting on the internet
210
Daniel P H Fox @phfox.net · 15/05/2026
officially announcing that i am bored of ai takes and that a lot of people need to go outside me included i need to go outside more
040
Daniel P H Fox @phfox.net · 11/05/2026
claude code is nice in moderation. helps me properly unwind in my free time while still being able to make a little progress on my projects, whether that's helping out with getting my plans into documents, adding small features, or doing well scoped tasks it's also nice to take a day to read code
010
Daniel P H Fox @phfox.net · 07/05/2026
(context: i was reading a leftist post thread i was really vibing with, but then the person got all puritan and elitist and now i'm sad about the state of online discourse again)
010
Daniel P H Fox @phfox.net · 07/05/2026
i can simply only apologise for the way the people i agree with choose to poorly behave themselves on the internet
110
Daniel P H Fox @phfox.net · 19/04/2026
in reality I'm spending a lot of time working on this renderer actually going back and making the _terrain generator_ more performant because I want to fill out the render distance faster. like just today I started precomputing noise fields across 4x4x4 chunk clusters to reduce double computations
010
Daniel P H Fox @phfox.net · 19/04/2026
don't mind the wood plank water i needed to test some UV calculations
110
Daniel P H Fox @phfox.net · 19/04/2026
honestly very glad i did the terrain generator before the renderers, because this is the best Lorem Ipsum content you could ask for
150
Daniel P H Fox @phfox.net · 16/04/2026
Finally set up Claude to run in a super-locked-down container so we can skip the permission fatigue (which inevitably would've caused me to accept bad commands). aka: I gave my Claude guy their own room. (A basement room, padded cell, locked door, and I feed them through a slit in the door).
030
Daniel P H Fox @phfox.net · 13/04/2026
goodnight world
010
Daniel P H Fox @phfox.net · 12/04/2026
bevels + new terrain generation = win
0100
Daniel P H Fox @phfox.net · 12/04/2026
(for transparency; these are the reference renderer shots! realtime will probably not look exactly the same)
010
Daniel P H Fox @phfox.net · 12/04/2026
and some atmospheric fog
120
Daniel P H Fox @phfox.net · 12/04/2026
higher render distance
140
Daniel P H Fox @phfox.net · 10/04/2026
Pair programming with Claude on laying out chunks in a more cache-friendly way in memory & reducing memory overhead/bandwidth of terrain generation. Now boasting fast startup time and very smooth & consistent frame pacing.
270
Daniel P H Fox @phfox.net · 06/04/2026
Auto exposure, courtesy of Claude running in the background while I do other things - still needs a lot of tuning, but I'll do that later when I can give it more attention.
041
Daniel P H Fox @phfox.net · 06/04/2026
the new reference renderer is so good
010
Daniel P H Fox @phfox.net · 06/04/2026
goodnight
111
Daniel P H Fox @phfox.net · 05/04/2026
Accumulating multiple samples per pixel in the ray traced path now. (Not TAA, no reprojection - just to get cleaner reference stills.)
010
Daniel P H Fox @phfox.net · 05/04/2026
In short; this is not a skill worth training. Instead, you should do the thing our parents always told us to do (and then promptly forgot about): Seeing something on the internet is not enough conviction to say or think something. You need to do the work to get at the truth.
000
Daniel P H Fox @phfox.net · 05/04/2026
2) Even if you've somehow figured out a way to, that knowledge isn't going to be transferable to the next generation. And the next. And the next. Eventually, it will reach the point where you're reading tea leaves, and then you'll be even more destructive.
110
Daniel P H Fox @phfox.net · 05/04/2026
If you think you can tell AI generated stuff from human stuff: 1) No, you can't. I've seen so many people lash out at actual artists for using AI when they _haven't_. You're hurting the people you're trying to evangelise for. None of us want it. Stop.
110
Daniel P H Fox @phfox.net · 05/04/2026
As an aside; since switching from rust_gpu to Slang, shader development has been going a ton smoother. I suppose it really does help to use something that is both reasonably complete and reasonably documented after all. (I never got to the point of sharing any Rust code because of the toolchain.)
010
Daniel P H Fox @phfox.net · 05/04/2026
Now rendering albedo, normal and combined AO/emissive maps in the ray tracing path! Still no water pass as that's more complex to pull off. Next up: going to start setting up a proper tonemapping pipeline and get some antialiasing in there.
020
Daniel P H Fox @phfox.net · 04/04/2026
Building the ray tracing acceleration structure on the fly from actual voxel data.
0110
Daniel P H Fox @phfox.net · 03/04/2026
One year ago; the original ray-traced lighting solution I was tinkering with for Cavey. Funny that exactly one year later, I'm still playing with rays.
040
Daniel P H Fox @phfox.net · 03/04/2026
Runs at 90fps on Deck, pretty goooood
010
Daniel P H Fox @phfox.net · 03/04/2026
youtu.be/PBa2MQ3WkQc?...
youtu.be
Voxel 64-tree DDA test w/ 2048x512x2048 world
YouTube video by Cavey / Daydream
100
Daniel P H Fox @phfox.net · 03/04/2026
2048x512x2048 GPU voxel ray tracing renderer test.
1110
Daniel P H Fox @phfox.net · 31/03/2026
Oh shoot I guess it’s time to follow it up with something more profound…. lets see…. WGSL is terrible and badly equipped for anything GPU driven. But I still haven’t found a shader language I like, so maybe that makes it relatively less bad. Subscribe for more insightful video essays or whatever.
020
Daniel P H Fox @phfox.net · 31/03/2026
(The irony of this being an AI post is not lost on me.)
000
Daniel P H Fox @phfox.net · 31/03/2026
Whatever the answers are… there has to be something more interesting to do with your time than flood Lobsters and HN with the same rehashed bullish/bearish/moralistic blog posts everyone has read two hundred times. I am pro-human. Can we just go back to sharing what you’re doing?
110
Daniel P H Fox @phfox.net · 31/03/2026
There are multiple orthogonal questions about multiple orthogonal things. What are the ethics of using / profiting off crawled data sets? Is code generation actually useful for focusing on design decisions, or a distraction from them? What about the interns & juniors? Is human provenance important?
110
Daniel P H Fox @phfox.net · 31/03/2026
Anyhow, for my first post… Having an opinion about AI is not a replacement for a personality trait. It’s a complex, nuanced, multifaceted issue with real world weight. I don’t care *what* your opinion is - even in my field, reasonable people don’t agree on fundamentals here - I care about *why*.
110