Hacker News: benchloftbrunch

New comment by benchloftbrunch in "The 1987 game “The Last Ninja” was 40 kilobytes"

benchloftbrunch — Mon, 06 Apr 2026 12:39:42 +0000

27 unique levels. 40KB minus a handful of spare bytes and some unused code. The max the NES can support without mappers. Modern NES homebrew and demoscene can do fancier stuff with this budget given the extra decades of learned tricks, but for the state of console gaming in 1985, SMB1 is damn impressive.

Also remember all of that was ROM, the NES had a mere 2 kilobytes of RAM for all your variables and buffers.

New comment by benchloftbrunch in "Tin Can, a 'landline' for kids"

benchloftbrunch — Mon, 23 Mar 2026 11:10:46 +0000

Best (for older kids) would be a dumb cell phone like we had in the 2000s. Good for phone calls, texting, and simple offline apps like casual games, camera and music player. Maybe email. Definitely no web browser, youtube, or social media crap.

I don't know the extent to which such devices are still manufactured today.

New comment by benchloftbrunch in "The gold standard of optimization: A look under the hood of RollerCoaster Tycoon"

benchloftbrunch — Mon, 23 Mar 2026 10:16:33 +0000

Problem with f16 is that hardware support is still "new" and can't be relied on in consumer grade CPUs yet.

New comment by benchloftbrunch in "Australian author's erotic novel is child sex abuse material, judge finds"

benchloftbrunch — Tue, 10 Feb 2026 13:33:11 +0000

GP is asking about the text prompt itself, not the generated image. If pure text can qualify as CSAM in Australia then it's a logical question.

New comment by benchloftbrunch in "AMD64 Bit Matrix Multiply and Bit Reversal Instructions"

benchloftbrunch — Sat, 31 Jan 2026 20:59:44 +0000

I believe this means "VBMACXOR16X16X16" is now officially the longest x86 mnemonic

New comment by benchloftbrunch in "Understanding the bin, sbin, usr/bin, usr/sbin split (2010)"

benchloftbrunch — Mon, 05 Jan 2026 07:46:21 +0000

Question: why did they decide to make /usr/bin the "primary" and /bin the symlink? Methinks it should have been the other way around as was the original Unix design before the split.

Also the first URL is serving me scam popup ads that do a crap job at pretending to be android system alerts. Next time please try to choose a more reputable source.

New comment by benchloftbrunch in "OneDrive just deleted all of my files"

benchloftbrunch — Mon, 05 Jan 2026 07:36:11 +0000

Thankfully it's trivially easy to disable OneDrive via the task manager startup tab. Never had any issues with MSFT sneakily turning it back on either.

This super aggressive OneDrive shit is also why I've stopped putting most things in the standard folders and now just have my own alternative hierarchy in %USERPROFILE% instead.

New comment by benchloftbrunch in "Go away Python"

benchloftbrunch — Tue, 30 Dec 2025 19:33:01 +0000

And on macOS if you need bash > 3.2

New comment by benchloftbrunch in "Gpg.fail"

benchloftbrunch — Sun, 28 Dec 2025 16:12:29 +0000

What is the alternative to PGP for the specific use case of secure email? That doesn't mandate dealing with the X509 certificate bureaucracy?

New comment by benchloftbrunch in "Package managers keep using Git as a database, it never works out"

benchloftbrunch — Fri, 26 Dec 2025 17:03:08 +0000

As long as you don't have any security compliance requirements and/or can afford the cost of self hosting your LLM, sure.

Anyone working in government, banking, or healthcare is still out of luck since the likes of Claude and GPT are (should be) off limits.

New comment by benchloftbrunch in "Time-Traveling to 1979: Advice for Designing 'C with Classes"

benchloftbrunch — Wed, 24 Dec 2025 15:15:08 +0000

I would add to that, replace #include with a proper module system that fixes the encapsulation and redundant parsing problems once and for all.

It's 2025 and C++ modules still aren't suitable for real world use yet despite being standardized 5 years ago.

Additionally standardize the ABI up front so that different compilers can interoperate. Make namespaces native to the object file format.

Also, explicitly standardize a compiler optimization mode that does not try to exploit UB in eldritch ways that break basic assumptions about how the machine works for 1% performance gain. I get that's an undecidable problem so it's ok if some extra annotations (call them "attributes" and write them [[like this]]) are needed here for explicit optimizer hints.

New comment by benchloftbrunch in "Japan law opening phone app stores to go into effect"

benchloftbrunch — Sat, 13 Dec 2025 13:39:58 +0000

No doubt Nintendo was involved in the lobbying effort for this. Back in the 80s they successfully pushed to amend Japanese copyright law to ban game rentals.

New comment by benchloftbrunch in "NaN, the not-a-number number that isn't NaN"

benchloftbrunch — Fri, 31 Oct 2025 05:56:27 +0000

Per IEEE 754, yes, but JS the language doesn't distinguish between NaN representations.

New comment by benchloftbrunch in "NaN, the not-a-number number that isn't NaN"

benchloftbrunch — Fri, 31 Oct 2025 05:38:18 +0000

Also, NaN is the only value in JS that isn't === to itself, so if for some reason you want to test for strict value identity with the value NaN of type number, that's one way to do it:

if(x !== x) ... // x is NaN

New comment by benchloftbrunch in "Second Chances on YouTube"

benchloftbrunch — Sat, 18 Oct 2025 14:40:37 +0000

In other words, certain people banned for political reasons are now being unbanned for political reasons.

New comment by benchloftbrunch in "Show HN: Firm, a text-based work management system"

benchloftbrunch — Thu, 16 Oct 2025 01:30:36 +0000

Looks like an early stage, immature project, but it's a neat concept.

It seems Windows Defender flags the zip download as a trojan. Likely false positive since scans on the unzipped exe (Defender and virustotal) come back clean.

I'd suggest providing a way to disable the builtin schemas in case someone wants to use it for more tech things as opposed to business things.

New comment by benchloftbrunch in "Keyhive – Local-first access control"

benchloftbrunch — Thu, 02 Oct 2025 11:48:56 +0000

Note: this has absolutely nothing to do with the Windows registry, despite the name

New comment by benchloftbrunch in "Kmart's use of facial recognition to tackle refund fraud unlawful"

benchloftbrunch — Mon, 22 Sep 2025 13:48:56 +0000

In Australia, apparently

New comment by benchloftbrunch in "NPM debug and chalk packages compromised"

benchloftbrunch — Tue, 09 Sep 2025 14:28:45 +0000

> X509 certificate

It should be a PGP or SSH key, absolutely not an X509 certificate (unless you allow self signed).

Personal identity keys should be fully autonomous and not contingent on the formal recognition of any external authority.

New comment by benchloftbrunch in "NPM debug and chalk packages compromised"

benchloftbrunch — Tue, 09 Sep 2025 13:48:29 +0000

As long as you're OK with self signed certificates or PGP keys, I'd be on board with this.

I really, really dislike the idea of using TLS certificates as we know them for this purpose, because the certificate authority system is too centralized, hierarchical, and bureaucratic, tightly coupled to the DNS.

That system is great for the centralized, hierarchical, bureaucratic enterprises who designed it in the 90s, but would be a pain in the ass for a solo developer, especially with the upcoming change to 45 day lifetimes.