Scene_Cast2

8 ms·

61.
by Scene_Cast2 7mo ago
Some more love for the updates page. E.g. select a subset of updates to install, be more clear that the last update time could be different if you installed updates via CLI, that kind of thing.
62.
by Scene_Cast2 7mo ago
There's also Moth-class sailing or windfoiling for mere mortals.
63.
by Scene_Cast2 7mo ago
Are there any good static (i.e. not runtime) type checkers for arrays and tensors? E.g. "16x64x256 fp16" in numpy, pytorch, jax, cupy, or whatever framework. Would be pretty useful for ML work.
64.
by Scene_Cast2 7mo ago
Of typical homelabs that are posted and discussed. The online activity of the homelab community leans towards those who treat it as an enjoyable hobby as opposed to a pragmatic solution. I'm on the other side of the spectrum. Devops is
65.
by Scene_Cast2 7mo ago
It still seems to have the same pitfalls as all the other image generation models. I ran it through my test prompt (wary of posting it here, lest it gets trained on) - it still cannot generate something along the lines of "object A, bu
66.
by Scene_Cast2 8mo ago
It's a shame that it's not on OpenRouter. I hate platform lock-in, but the top-tier "deep think" models have been increasingly requiring the use of their own platform.
67.
by Scene_Cast2 8mo ago
Sweet! I love clever information theory things like that. It goes the other way too. Given that LLMs are just lossless compression machines, I do sometimes wonder how much better they are at compressing plain text compared to zstd or simila
68.
by Scene_Cast2 8mo ago
I just ran this with Gemini 3 Pro, Opus 4.6, and Grok 4 (the models I personally find the smartest for my work). All three answered correctly.
69.
by Scene_Cast2 8mo ago
WizTree uses a similar idea - load the file system indices and works almost instantly.
70.
by Scene_Cast2 8mo ago
Not the OP, but the new LLMs together with harnesses (OpenCode in my case) can handle larger scopes of work - so the workflow moves away from pair programming (single-file changes, small scope diffs) to full-feature PR reviewing.
71.
by Scene_Cast2 8mo ago
Just to nitpick - compilers (and, to some extent, processors) weren't deterministic a few decades ago. Getting them to be deterministic has been a monumental effort - see build reproducibility.
72.
by Scene_Cast2 8mo ago
Try looking into videography COBs. I would recommend something like the Zhiyun Molus G300, SmallRig RC 220B or similar. The absolute cheapest lumens per dollar COB would be the GVM SD300D, although I highly question the reliability and ligh
73.
by Scene_Cast2 8mo ago
Oh hey, I have one of these! I really like it. It's quite a unique design. People (especially where it snows and gets gloomy) should have more lumens, and I'd recommend this lamp for others. One downside is that the active fan coo
74.
by Scene_Cast2 8mo ago
I bought a vintage (pre-WW1) circular slide rule. But it's too much of an "artifact" to use - I'm afraid I'd damage it.
75.
by Scene_Cast2 8mo ago
Then why not geometric / Clifford algebra?
76.
by Scene_Cast2 8mo ago
As someone who's worked with population data, I found that there is an enormous rift between reported opinion (and HN and reddit opinion) vs revealed (through experimentation) population preferences.
77.
by Scene_Cast2 8mo ago
You're typically issued a corporate phone, it's the only phone that can open work email. You have a choice between an Android (something like a Pixel and a Samsung) and an iPhone, with some companies incentivising Androids with th
78.
by Scene_Cast2 8mo ago
I discovered it a few weeks ago. It solved my problem of having a hundred arxiv tabs open. Highly recommended. I'm also looking forward to trying out the new annotations feature.
79.
by Scene_Cast2 8mo ago
It also percolates into reviews, too. When a nontrivial fraction of the community is buying dreams and is about collecting as opposed to using whatever it is, some reviewers style their content towards that crowd and overlook issues or bene
80.
by Scene_Cast2 9mo ago
I used to use Doxygen to create caller and callee graphs to understand code flow. Unfortunately, the tool hasn't really changed in more than a decade.
81.
by Scene_Cast2 9mo ago
My baseline was non-HC "vanilla" residuals; I didn't do a meaningful HC run to compare. My application has some particularities (important and easy to identify per-token signals) that result in values growing (about 3x to 10x
82.
by Scene_Cast2 9mo ago
I implemented this for a toy 8M ViT-style model. Got neutral results. This is just an anecdote and is not representative - I think mHC will help with larger parameter sizes and larger token counts.
83.
by Scene_Cast2 9mo ago
Some types of software aren't too far from that paradigm.
84.
by Scene_Cast2 9mo ago
Tangentially related - I was doing some highly sensitive analog electronics work. It's surprisingly hard to find capacitors that don't exhibit microphonic behavior - i.e. they don't produce (miniature) voltage swings when you
85.
by Scene_Cast2 9mo ago
There was previous discussion a few days ago on "Solving The Problems of HBM-on-Logic" that I think is relevant. https://news.ycombinator.com/item?id=46302002 https://morethanmoore.substack.com/p&#
86.
by Scene_Cast2 10mo ago
I've heard from a friend who works in the embedded space that Tensorflow Lite is still the only realistic (supported by vendors) game in town for running ML models on microcontrollers such as ESP32, nRF, etc. The hardware support liste
87.
by Scene_Cast2 10mo ago
If you're into carrier landings and have a VR headset, I highly recommend checking out VTOL VR.
88.
by Scene_Cast2 10mo ago
I'd guess that the datacenter "GPUs" lack all the fixed-function graphics hardware (texture samplers, etc) that's still there in modern consumer GPUs.
89.
by Scene_Cast2 10mo ago
There is friction to asking AI yourself. And a comment typically means that "I found the AI answer insightful enough to share".
90.
by Scene_Cast2 10mo ago
Ah, I think I agree. There could be a potential unrelated handicap, so there is a lack of a guarantee or a proof.

More ›