Hacker News: mistercow

New comment by mistercow in "GPT-5.4"

mistercow — Fri, 06 Mar 2026 00:10:16 +0000

My current hunch is that that benchmark captures most of the relevant gap between Anthropic and the rest. “Can’t distinguish truth from fiction” has long been one of the deeper complaints about LLMs, and the bullshit benchmark seems like a clever approach to testing at least some of that.

New comment by mistercow in "Claude Code LSP"

mistercow — Mon, 02 Mar 2026 13:13:04 +0000

I don't mind if people use AI to help them write, but when I see this kind of thing, it implies to me that they're barely even skimming it before posting. Surely people don't want this super cliche AI-hype-man tone in their blog posts, right? And if they haven't taken the time to at least skim through it and iterate on basic style, why should I assume it's worth my time to read it?

New comment by mistercow in "Stop generating, start thinking"

mistercow — Mon, 09 Feb 2026 07:01:28 +0000

> Maybe its because I'm using not-frontier models to do the coding

IMO it’s probably that. The difference between where this was a a year ago and now is night and day, and not using frontier models is roughly like stepping back in time 6-12 months.

New comment by mistercow in "Vitamin D and Omega-3 have a larger effect on depression than antidepressants"

mistercow — Thu, 29 Jan 2026 11:47:49 +0000

Flax seeds are a very tedious and inefficient way to get omega-3 as a vegan, particularly because they contain ALA, a short chain omega-3, which our bodies are extremely inefficient at turning into long chain fatty acids.

Just get an algae oil based DHA+EPA supplement.

New comment by mistercow in "Go Gray, Not Cray: Why You Should Grayscale Your Phone"

mistercow — Sun, 28 Dec 2025 15:29:23 +0000

Fine, luma, not luminance. But what you're describing is exactly that calculation. This does not change my point. Again: If you’re doing this correctly, the perceived brightness stays the same.

New comment by mistercow in "Go Gray, Not Cray: Why You Should Grayscale Your Phone"

mistercow — Sun, 28 Dec 2025 10:11:34 +0000

So specifically on a backlit LCD screen with dimming zones, and in the specific case where an entire zone would have originally been blue or red (which are perceptually dimmer), you could plausibly get a small amount of energy savings. But “twice the battery life” from this is not plausible.

New comment by mistercow in "Go Gray, Not Cray: Why You Should Grayscale Your Phone"

mistercow — Sun, 28 Dec 2025 05:02:25 +0000

Yeah, that’s the only explanation that makes sense. It’s just so strange to think that color pixels would use more energy.

New comment by mistercow in "Go Gray, Not Cray: Why You Should Grayscale Your Phone"

mistercow — Sun, 28 Dec 2025 04:49:38 +0000

> It seems obvious that a higher perceived brightness can be achieved for any given pixel if using greyscale instead of using colors, just because less of the backlight's area is occluded.

When converting to grayscale, you typically calculate the value of the pixel and then set all color components to that value. The point of this is to keep the luminance the same as it was in the original color pixel. If you’re doing this correctly, the perceived brightness stays the same.

And just as a smell test: have you ever converted an image to grayscale and flinched away because it seemed twice as bright? Of course not; it just loses its color.

The only way you would get more perceived brightness at lower backlight intensity would be if you physically removed the color gels that overlay the LCD matrix. Which is obviously not what they’ve done here.

I’m pretty sure the increase in battery life they observed is simply because they’re using their phone less, which is very much the main upshot of the other benefits they listed. The idea that color pixels drain more energy is just obviously nonsense.

New comment by mistercow in "Go Gray, Not Cray: Why You Should Grayscale Your Phone"

mistercow — Sun, 28 Dec 2025 04:04:43 +0000

> Color pixels drain more energy than grayscale ones. Personally found my phone lasting twice as long as before. Over time, a considerable extension of your phone’s lifespan.

What? Why? Why would you even entertain that as a hypothesis?

New comment by mistercow in "Why JPEG XL ignoring bit depth is genius (and why AVIF can't pull it off)"

mistercow — Mon, 27 Oct 2025 10:38:25 +0000

> and don’t think that the programmer more than the languages contribute to those problems

This sounds a lot like how I used to think about unit testing and type checking when I was younger and more naive. It also echoes the sentiments of countless craftspeople talking about safety protocols and features before they lost a body part.

Safety features can’t protect you from a bad programmer. But they can go a long way to protect you from the inevitable fallibility of a good programmer.

New comment by mistercow in "Why JPEG XL ignoring bit depth is genius (and why AVIF can't pull it off)"

mistercow — Mon, 27 Oct 2025 10:04:04 +0000

Multiple severe attacks on browsers over the years have targeted image decoders. Requiring an implementation in a memory safe language seems very reasonable to me, and makes me feel better about using FF.

New comment by mistercow in "Why JPEG XL ignoring bit depth is genius (and why AVIF can't pull it off)"

mistercow — Mon, 27 Oct 2025 10:00:48 +0000

Seems like the normal usage to me. The post above lists other criteria that have to be satisfied, beyond just being a Rust implementation. That would be the consideration.

New comment by mistercow in "High-resolution efficient image generation from WiFi Mapping"

mistercow — Wed, 01 Oct 2025 09:15:29 +0000

1. “High resolution” in this kind of context is generally relative to previous work.

2. “Postage stamp sized” is not a resolution. Zoom in on them and you’ll see that they’re quite crisp.

New comment by mistercow in "Brutalita Sans: An Experimental Font and Font Editor"

mistercow — Fri, 26 Sep 2025 00:09:29 +0000

Yes, and those glyphs don’t fit the grid. Try to redraw the n from the original font yourself. You can’t, because you can’t add points between the grid dots.

New comment by mistercow in "Brutalita Sans: An Experimental Font and Font Editor"

mistercow — Thu, 25 Sep 2025 13:46:02 +0000

I tried to do that, and now I’m just confused. The included glyph for the lower case n doesn’t actually fit the grid, so you can’t seem to replicate it. But also that grid doesn’t have enough resolution to do the tilde. Maybe I’m missing something?

New comment by mistercow in "Helium Browser"

mistercow — Thu, 25 Sep 2025 11:05:08 +0000

WebGPU is the only one of those I’ve really followed, but hasn’t that had a huge amount of input and changes due to other voices in the working group? That seems to contradict the simplistic picture painted above of Google just dictating standards to the industry.

New comment by mistercow in "Scream cipher"

mistercow — Sun, 21 Sep 2025 08:23:47 +0000

If you use base64 with the intention of hiding the encoded information, surely it’s as much a cipher as rot13 is, right?

New comment by mistercow in "Folks, we have the best π"

mistercow — Mon, 15 Sep 2025 09:22:33 +0000

> How to calculate π for n-metrics numerically. The general idea of "divide the circle into segments and calculate the length by the metric" is explained, but the exact algorithm or formulas are not shown.

I feel like that would have been a bit in the weeds for the general pacing of this post, but you just convert each angle to a slope, then solve for y/x = that slope, and the metric from (0,0) to (x,y) equal to 1, right? Now you have a bunch of points and you just add up the distances.

New comment by mistercow in "The math of shuffling cards almost brought down an online poker empire"

mistercow — Sun, 07 Sep 2025 08:44:40 +0000

There’s an almost mystical belief among certain tech and science journalists that computers are bad at randomness, and it’s really bizarre, and in my opinion, pretty harmful.

New comment by mistercow in "New interpretations suggest the "heat death" hypothesis might not hold (2023)"

mistercow — Sat, 30 Aug 2025 16:44:21 +0000

> ... all believe that the universe is not destined to grow more disorganized forever, but more complex and rich with information.

Maybe it's just a problem of being loose with terminology, but this seems to be contrasting entropy and information content, which is backwards?