Hacker Newsnew | past | comments | ask | show | jobs | submit | larodi's commentslogin

As a side note - Claude Code is making the CLI attractive in a renewed fashion - more than anything else did it last 20years.

Doesn’t seem to work on mobile for me on ios18 and iphone12

Claude Code does a very decent UI, somehow the text mode is much more attractive. As if there is once in a lifetime opportunity to make the console great again.

Really, inst there anything which comes Slug-level of capabilities and is not super expensive?

Vello [0] might suit you although it's not production grade yet.

[0] https://github.com/linebender/vello


Just use blend2d - it is CPU only but it is plenty fast enough. Cache the rasterization to a texture if needed. Alternatively, see blaze by the same author as this article: https://gasiulis.name/parallel-rasterization-on-cpu/


ThorVG might be worth a look - open source (MIT), ~150KB core, GPU backends (WebGPU, OpenGL).

We are using it as official dotLottie runtimes, now a Linux Foundation project. Handles SVG, Lottie, fonts, effects.

https://github.com/thorvg/thorvg/


In terms of performance, it's quite far from something like Blend2D or Vello though.

Blend2D is a CPU-only rendering engine, so I don't think it's a fair comparison to ThorVG. If we're talking about CPU rendering, ThorVG is faster than Skia. (no idea about Blend2d) But at high resolutions, CPU rendering has serious limitations anyway. Blend2D is still more of an experimental project that JIT kills the compatiblity and Vello is not yet production-ready and webgpu only. No point of arguing fast today if it's not usable in real-world scenarios.

How JIT kills compatibility if it's only enabled on x86 and aaarch64? You can compile Blend2D without it and it would just work.

So no, it doesn't kill any compatibility - it only shows a different approach.

BTW GPU-only renderers suck, and many renderers that have GPU and CPU engines suck when GPU is not available or have bugs. Strong CPU rendering performance is just necessary for any kind of library if you want true compatibility across various platforms.

I have seen many many times broken rendering on GPU without any ability to switch to CPU. And the biggest problem is that more exotic HW you run it on, less chance that somebody would be able to fix it (talking about GPUs).


Good ! Finally someone does it, sadly I didn’t have the time been planning for 1 year now.

Was this posted automatically or why it reads Vienam? Without the T! And the title also reads so?

I posted it with the original article title. I'm not sure who changed it, but yeah, there is a typo which also exists in the linked article.

Indeed it first had no T, and s.o. changed it. Also raises questions reg the original title.

everyone understood overnight what vibe-coding was, but only dared to go through the looking mirror and try to grok what the mirror is made of.

Alex Zhang just published the repo alongside already published paper here https://arxiv.org/abs/2512.24601v1

how about the track was 123 and they squeezed it to be within certain length, which defnitely would bring the BPM a little higher?

And they accidentally made it 1.003713731707 times faster? They chose a number that contains only the digits 0, 1, 3, and 7? C'mon, this cannot be a coincidence!!!

It's a good thing we're jumping to conclusions instead of exhaustively evaluating all of the places values exactly like this one appear when dealing with swing and quantization on, and especially when mixing 8, 12, 16 bit samplers and sequencers. Nevermind all of the little nudges from byte window mismatches when reading, playing back, or manipulating samples at varying bit depths and sample rates.

> It’s a good thing we’re all jumping to conclusions

I wholeheartedly agree. This thread would have been a lot less fun to create if I’d had to apply rigorous methodology and proper hypothesis evaluation practices. I’m really glad someone else appreciated that too :D


is super ineffective, indeed. if you need to pay 20$ to get s.o. to pay you 50$ for a service/product, well in all honesty calling people one by one and giving them 10$ is more likely to result in sale.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: