trzy

I’m not really sure what the state of ARKit learning materials is but there are great tutorials online by e.g., Ray Wenderlich. Getting started with Xcode and iOS can be a bit of a struggle. Follow some tutorials and take it step by step. Apple has some pretty good samples, too.

Unfortunately one big problem is the quality of Apple’s documentation. It’s nowhere near as good as e.g., the Unity engine documentation or Microsoft’s API docs. At best, it’s useful for discovering what exists and then you either figure it out or look for specific examples online.

There are many cases where it does not work or does something insane. However, the number of hooks exposed are very low and I’m impressed with how often it works. When it goes off the rails, it is usually because it 1) tries to use functions that don’t exist (the distance() function is one that I had to relent and implement; but I’ve also seen euclideanDistance() occasionally), 2) defines functions of its own after invoking them in code, or 3) misunderstands the relationship between the user (whose position is exposed through “cameraPosition”) and planes. I’ve put very little effort into mitigating these but have found that simply adding functions for it to use in common scenarios helps a lot.

I think the window size of ChatGPT is quite limited so I can’t use prompts that are arbitrarily long but it’s not hard to see how increasing this window size and fine tuning on custom APIs could produce a model that would work more often than not.

This is very cool! And I think it's a perfectly good approach. My understanding is that the usual way of doing things with LLMs is to train them on a series of specialized tokens that represent actions in your environment. E.g., this sequence of words results in this completion of action tokens.

The code approach is intriguing and I'd like to explore further but controllability is a real problem and bullet-proofing it would require a lot of effort, if it's even possible at all. I do think that a hybrid environment where one speaks what they want and then sees the code and can interact with it in a friendly way would be very intriguing for a sandbox experience.

Well, I feel less bad about myself for getting stuck there :D The semantics of how properties work in JS are so obnoxious. A read-only property can still be replaced entirely, which means e.g.:

entity.position = [0,0,0];

Is ok but:

entity.position[0] = 1; entity.position[1] = 1; entity.position[3] = 1;

Fails (and ChatGPT occasionally tries to do the latter).

Question: I couldn’t figure out how to export properties that were writable in JS. I describe the problem in Entity.js. Any pointers would be much appreciated! :)

I would love for this to become robust enough to give to kids and see what they dream up. Unlike adults, kids are far less inhibited at coming up with fun prompts, but that also means they would break this admittedly brittle prototype :D

Thanks for sharing that! I hadn't heard of LangChain. That looks super cool. I really hope ChatGPT gets turned into a proper API because fighting OpenAI's bot countermeasures is a real time sink.

These expat enclaves are a bit of a fantasy world. They attract like-minded outgoing people eager for adventure and making new friends who all happen to have one big thing in common: they’re Westerners in a foreign land. You’re playing on easy mode ;)

Back when I wrote a lot of x86 assembly in the late 90's/early 2000's, I used NASM (the Netwide Assembler). It was the most natural syntax. Very similar to MASM and TASM but without the unnecessary cruft (no "dword ptr", etc.) It was a pleasure to work with.

If you took all the assembly language ever written, I'd wager a negligible amount of it was written for GNU as. So yes, they are the odd ones out.

You want Tilt5 then.

All excellent points but Windows is a very widely used platform (perhaps the most widely used, by some metrics) and it’s actually quite nice for development. It’s hard to take tools seriously that don’t work on it.

Unpopular opinion but I think the Unix approach of lots of little tools with arcane configuration files all blaming each other sucks. I get why it exists, why it became popular, and why it remains popular in some scenarios, but I don’t feel beholden to it.

Last year I tried doing a side project with a talented ex-Google buddy who insisted we set up Bazel to replace my simple Makefile. Three weeks later it still wasn’t working on my Windows box. We had a mixed Python and C++ code base and I like to use MinGW64 gcc on Windows. He blamed Windows and tried to get me to switch to Mac (no thanks, lol) and eventually he lost interest and gave up. The project went on to win an OpenCV funded competition and became the basis of a startup — good job, GNU Make!

So the answer IMHO to “when to use Bazel” is “never” :)

The time has never been better for a new “game engine” or 3D app framework. Unity is trying to be everything for everyone and hasn’t done a great job of taking care of its core audience: indie game developers.

Unity is the de facto choice for XR developers but it’s a poor application framework. Something like StereoKit that caters to non-game devs and can provide rendering, a flexible interaction model, UI, and a good multi-user app model could really steal Unity’s thunder for multi-platform development.

This is super cool. When I was in college I used a breadboarded Z80 with an FPGA to build a really simple video game system with VGA output and a Genesis controller for input. The schematics, source code, and project write up are still on my home page (http://trzy.org — near the bottom). I should probably throw it up on GitHub.

Sounds like you are trying to rationalize a taste for terrible movies with frenetic pacing.

The non FMV library on the Sega CD is surprisingly good! Give it another look. Back in the day the FMV games had all the spotlight.

Good point. This is a problem with Anglo culture specifically (note the poster who pointed out that Canada’s gun crime per capita closely tracks the US). Anglo culture is not sufficiently developed to support widespread gun ownership and therefore access to guns here must be curtailed.

There was someone on my last team who loved using this term in pointless nitpicking critiques. We were a prototyping team. One minute he’d wax philosophical about how prototyping should be done simple and fast, using throwaway code and cobbling together any preexisting bits we can find to move fast, and the next he’d claim that the name “NetworkManager” for the net code client component was “code smell”. Very eye roll inducing.

I've heard that they did order soldiers to advance through mine fields (effectively de-mining a path through them using abundant human bodies) or face the NKVD. The calculation was simple: you might survive the minefield but you had a 100% chance of being executed for refusing the order.

Suicide torpedoes were proposed by the Polish military in 1939 and the program appears to have gotten as far as signing up recruits but to my knowledge none were ever deployed.

Why? The labels are given in absolute units but they are correct.

Question: where did you learn 3D rendering in the 1980’s? How was this knowledge disseminated? How much was “solved” at the time?

This is a “debate” that exists only on HN. Laptops are not what propelled Apple to become the most valuable company in the world and their importance inside the company reflected that until fairly recently.

Research on finding a legitimate use case ;)

You’re right, they’ve moved on from periodicals and discussion groups to laptop usage for hours at a time :)

Historically, cafes were social forums where people gathered for hours at a time to read and converse. Maximizing absolute profit by optimizing customer throughput is the modern take. But that’s not what cafes are for and so you should not feel bad for forcing owners to understand their business and optimize for both profitability and the cafe experience.

True but the thorny question is not “what happened” but “what happened that was important?” What do we focus on investigating and how do we connect it to a broader narrative (and when writing a paper, a book, or a lecture, there is always a narrative because that’s how we communicate).

I don’t understand the point about the difficulty of grayboxing with Unity. Can you elaborate? All my projects start with programmer art — often just the built-in primitive (cubes, spheres, cylinders, planes).

Put a bunch of smart, pragmatic people in a room and you’ll only end up with the current state of the software industry. The stricter the rules, the more durable the movement.

He could definitely benefit from coming across as more pleasant and affable, though.