Adam Byrd

Engine and systems programmer. Previously Epic, Phoenix Labs, Blizzard, Limit Theory, Pixel Dash. Games, physics, C++ replacements, and dark beer.

[Adam Byrd @abyrd89]

Seems like a great time to start investing in in-house tech again. The biggest barrier has always been raw hours, but it's faster to build than ever before. The productivity boost of full control can be immense.

[Adam Byrd @abyrd89]

@marcsh Yep. It's still (tied for) the best project I've ever worked on. I had so much fun. Wish we had managed to pull it off.

Tomorrow Corporation's engine is a big inspiration. So much of it is possible in a native language and large project, with enough effort to lay the foundation.

[Adam Byrd @abyrd89]

What I want as a game dev: an engine that's both high productivity and high performance (instead of just one or the other); extremely fast workflows; smaller, higher skill teams; smaller scope; more novelty.

What I don't want: the metaverse; Roblox envy; UGC; tools for amateurs.

[Adam Byrd @abyrd89]

@OlexGameDev That's not what dictates company direction. Leadership decides what the goal is and hires people to make it happen.

[Adam Byrd @abyrd89]

@grigsbysworld I do understand the concern. There wasn't enough space to clarify "at the cost of tools for experts".

Though, even with that caveat I'm skeptical. You don't get a "table saw for people who don't understand table saw". You get a full fat tool and professional respect.

[Adam Byrd @abyrd89]

Could use VEH and handle any EXCEPTION_ACCESS_VIOLATION but that feels heavy handed. Probably better to just put the auto-commit in the arena instead of the allocator. It was a neat idea though.

[Adam Byrd @abyrd89]

I tried making a virtual allocator that auto-commits memory only as it is touched. Unfortunately on Windows you can only guard a page (so a vectored exception handler is run) if it has already been committed. Bummer.

[Omar 🍋 @ocornut]

An unexpected thing happened: somehow I received a special award at the Pégases (french game industry awards), partly recognizing the continued work on Dear ImGui. Thank you🙏 https://t.co/FmvHQiPBH6

[Adam Byrd @abyrd89]

I love the irony of Perforce, a version control system, giving me warnings that file operations cannot be undone. That's...literally it's job.

[Adam Byrd @abyrd89]

Why is it { lvalue, prvalue, xvalue } + { glvalue, rvalue } instead of { lvalue, rvalue, xvalue } + { glvalue, grvalue }?

[Adam Byrd @abyrd89]

Many random issues I've been having with my PC were suddenly resolved when I realized several USB ports are malfunctioning. I had a Wooting keyboard (analog, Hall effect sensors) plugged in and noticed every key was registering a non-zero input all the time. That's a new one.

[Eoin Higgins @EoinHiggins_]

Doesn’t get much clearer than this

[Jason Smith - 上官杰文 @ShangguanJiewen]

I am an American living in China, and I believe—deeply and without apology—that telling the truth about China’s successes will make us stronger.

In fact, the real danger to America is not that some of us are willing to speak honestly about what China is doing well, but that so many voices at home insist on hiding, minimizing, or distorting that reality. A nation that refuses to see the world as it is cannot compete in it, cannot reform itself intelligently, and cannot secure a prosperous future for its people.

Those lying to us about China’s advancements in affordable healthcare, public transportation, 90% homeownership, public safety, and so forth are holding us back. They are preventing the USA from excelling.

Patriotism is not cheerleading. Patriotism is not repeating comfortable myths. Patriotism is the courage to confront facts, especially when those facts are inconvenient! Patriotism also does not mean hating other nations. China and the US are perfectly capable of meeting as peers and cooperation to build a shared future for all our peoples.

Every major American success story—from industrialization to victory in World War II to the space race—was built on brutally honest assessments of where we stood relative to others. We studied our peers, learned from them, and adapted. We did not weaken ourselves by pretending they were incompetent or morally inferior. We strengthened ourselves by learning from others.

Living in China has forced me to confront realities that many Americans are shielded from. I see functioning high-speed rail that connects entire regions. I see cities built around public transit rather than endless sprawl. I see long-term infrastructure planning, aggressive investment in energy, and a population that broadly understands where the country is trying to go over the next twenty or thirty years. None of this means China is perfect. It is not. But pretending these achievements do not exist, or dismissing them as illusion or propaganda, does not make the United States stronger. It makes us blind.

Those spreading lies about China are not making America safer, they are betraying what the USA stands for. We are not cowards. We should know if other nations are able to create the conditions for economic progress by newly minted means. We also can and should work with China to build that future.

What is truly unpatriotic is deceiving the American people about the scale, speed, and seriousness of China’s development. When Americans are told that China is “collapsing,” “incapable,” or “doomed any day now,” it encourages complacency. It tells us we don’t need to invest, reform, or rethink our assumptions. It reassures us that our systems are automatically superior and self-correcting. History is unforgiving to nations that believe that.

I speak honestly about China because I care about America’s future. If China is building energy capacity at a pace we are not matching, Americans deserve to know that. If China is training engineers, deploying infrastructure, and planning industrial policy more coherently than we are, Americans deserve to confront that reality.

The path to American renewal does not run through denial. It runs through clarity. The truth matters. Facts matter.

If we want the United States to thrive in the decades ahead, we must stop confusing propaganda with patriotism. Real patriotism demands courage—the courage to look at the world as it is, learn from it, and act accordingly. Anything less is not loyalty. It is negligence.

[Omar 🍋 @ocornut]

Converging toward a principal Unreal Engine backend/binding for Dear ImGui? https://t.co/QWqRiZRoud (Help wanted!)

[Adam Byrd @abyrd89]

Despite being an expert in the language, I recognize it for what it is: a mess of unnecessary complexity and wasted productivity.

[Adam Byrd @abyrd89]

Clickbait title aside, the first half of this is a well-informed, thorough, and accurate critique. Some points are tooling issues, and some are probably intrinsic to any natively compiled language, but the remainder are more than enough to make the point.

https://t.co/WYFZO3ZieU

[Adam Byrd @abyrd89]

@nicbarkeragain Though, upon reflection, I suppose that's not what you're looking for. Data generated and durable knowledge aren't the same. Data generated must be several orders of magnitude larger.

[Adam Byrd @abyrd89]

@nicbarkeragain There's an estimate for all of human knowledge at something like 5 exabytes as of somewhere around 2015. Might be able to use that for a Fermi estimate.

[Adam Byrd @abyrd89]

@OlexGameDev @molecularmusing @epyoncf @PompiPompi @mgrviper @_nakedeyes @liveplusplus O3 doesn't exist on msvc. This example is inlined with O2.

[Adam Byrd @abyrd89]

@awesomekling If it solves a problem with acceptable trade-offs, I find it really hard to make an argument that people should do it "the hard way". Hard work and discipline are in themselves desirable, of course, but arbitrarily picking a less effective place to apply them is counterproductive

[Adam Byrd @abyrd89]

@awesomekling I don't have a dog in this fight, but my loose understanding at this point is that it substantially reduces visceral fat and dieting doesn't. That in itself means it performs a useful function that doesn't have a reasonable alternative.