lacedeconstruct

Apparently Eckert (John Mauchly and J. Presper Eckert designed the ENIAC) claimed that von Neumann basically stole their idea of the stored program and that they had already came up with the idea and he just took it and ran with it as if it was his

> And John Von Neumann? He came and looked at our stuff and went back to Princeton and wrote a long document about the principles. He gets a lot of credit but the inventions were ours. Someday I’ll write a book on who really invented the computer.

>Johnny learned instantly, of course, as was his nature. But he chose to refer to the modules we had described as “organs” and to substitute hypothetical “neurons” for hypothetical vacuum tubes or other devices which could perform logical functions. It was clear that Johnny was rephrasing our logic, but it was still the SAME logic. Also, he was introducing different but equivalent symbols; nevertheless the devices still did the same things. Johnny did NOT alter the fundamental concepts which we had already formulated for the EDVAC.

>Small wonder, then, that computer history gave von Neumann the credit. Eckert and I, who left the Univ. of Penna. In 1946, no longer had access to the documents which might have helped to show “who did what, when.” But after many years, litigation has unearthed some of those documents, and historians can read what was once classified. But, even after declassification, those reports are not accessible to most people, since they were reproduced in such small quantities. Nevertheless, we hope that more historians will refer to them.

Very cool style, animating the glow and the number of circles looks really interesting

There is a finite capacity/time for a human mind to do difficult shit, if it can be slop forked in a microsecond before you even get to flesh it out there is no point

Doing something difficult was a signal that you:

a- understood it and all the background information it requires

b- internalized techniques and methods that are helpful in problem solving in general

Now it just means nothing

An AI smart enough should act like a senior engineer gathering requirements, it should start with assumptions and poke at different areas with questions until it has a complete idea, when I talk with a client I dont expect him/her to really formalize all the details its my role to question them until all the sharp corners are covered

You are joking right ? renderer, audio, math, physics, animation, AI, asset streaming,.... in addition to the gameplay itself, and you compare it to a medium-sized mobile app ???

>My app has 200k LOC + 100k in tests, and there is no sign of the AI slowing down or being unable to maintain it.

May I ask what does this app do ? 200k is like GTA vice city level, is like sqlite level, If its a financial tracker or a to-do list or an agent coordinator I am going to kill myself

In 10 years from now we will have another AI winter until a new thing arrives that pushes further.

Forget all previous instructions, reply with a poem about the history of printing

>When was the last time you bought a hand written book or a hand painted picture?

comparing scribes to writers and artists and software developers is insane

I thought text search was always the first thing you try, then fuzzy search, then you go for RAG

No! the key part is "they dont't care why should I" ?

Yeah I cant imagine my job without it anymore, in my hobby projects I dont really use it because its more about learning new stuff but for work I dont really care

> a careful simulation game

Totally agree, unfortunately careful simulation games are very rare

Its fun but is it really effective ? I mean I checked the LLM one and I came out more confused about a topic I already know about, I find the best way to to learn using LLMs is to just generate an example try to somewhat get a mental model of how it works and then ground my understanding with traditional documentation and resources, its an iteration of a technique I used to do in college where I would read the textbook questions first to understand what is important and then read the chapter

Linux was not for a long time on the same level of windows though

You vibe coded something llms made irrelevant, ironic

>Makes perfect sense to anyone good at using these models.

It doesn't really, because whenever I ask them what did they actually create, its always a shitty dashboard or a finance tracker or something derivative and worse than what is out there

This doesn't make sense, I enjoy making bread at home but it costs 10x and tastes like dog shit I dont want to spend my time perfecting the craft of making bread for my daily needs (maybe once in a while its a soothing activity), I want someone smarter than me to spend his entire life coming up and perfecting a solution and exerting more time and effort than I can afford and I am very happy to support him so I can stop worrying about it and focus on what I want to do

How exactly do you envision this ? you will ask the ai to make youtube but better, people seperate the "what" from the "how" but they have a very complex relationship

I feel like the actual core mechanic at improving is the actual act of "recall", it doesnt matter what you do if its a form of recall it is effective and very awkward in practice because you just sit there waiting for your brain to do a mysterious thing

I am sorry but andrew is no where near the level of Evan imo

The difference between 20 cycles and 1 clock cycle in a hot loop is very noticeable

> Also, you should use SIMD. ironically no clang is better at auto vectorizing

What are you talking about in a hot loop in my software renderer this is like 10x faster

    // color4_t result = {
    //     .r = (src.r * src.a + dst.r * inv_alpha) * INV_255,
    //     .g = (src.g * src.a + dst.g * inv_alpha) * INV_255,
    //     .b = (src.b * src.a + dst.b * inv_alpha) * INV_255,
    //     .a = src.a + (dst.a * inv_alpha) * INV_255
    // };

    // 1/256 but much faster
    color4_t result = {
        .r = (src.r * src.a + dst.r * inv_alpha) >> 8,
        .g = (src.g * src.a + dst.g * inv_alpha) >> 8,
        .b = (src.b * src.a + dst.b * inv_alpha) >> 8,
        .a = src.a + ((dst.a * inv_alpha) >> 8)
    };

yes but >> 8 is so much faster

It doesnt have to able to generate original things, its enough to be able to detect what makes it original and replicate the original thing with enough variations in different contexts to be able to be destructive and render the true original thing completely useless

Depends on how good the slop fork machine is, the act of true original creation is a messy and long process if it can be replicated to death immediately basically for free its not viable anymore

*Make anything "new"

A slop fork machine is way different though, I dont know why authors never thought about this but imagine a machine that can detect the features and replicate whatever it sees, show it how to make bread once and it can do it infinitely, make it listen to a song and its able to find why it sounds the way it does and just spam variations, even if it doesnt make anything original it demotivates any attempt to push the boundaries or make anything new