Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
kristianp
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
31.
▲
by
kristianp
2mo ago
Their statement about sharing an Ultra for an office LLM server is a little misleading. The tokens/second would be too slow for any business use case. You'd want to buy multiple GPUs for that purpose, probably Nvidia, despite th
32.
▲
by
kristianp
2mo ago
Thanks. I see but 2 RAM chips on one photo.
33.
▲
by
kristianp
2mo ago
https://xcancel.com/IntCyberDigest/status/209195400829849193... I wonder if these are M5 Ultra machines?
34.
▲
by
kristianp
2mo ago
This is by the author of "Designing Data-Intensive Applications", one of HNs favourite books. https://dataintensive.net/
35.
▲
by
kristianp
2mo ago
Agreed. The AI news cycle is constant and promises the eventual end to human usefulness. I feel like one day we'll be but pets to the machines (like those in Iain M. Banks' culture novels), but real life won't be a communist
36.
▲
by
kristianp
2mo ago
There are some posts on https://www.cpu-world.com/forum/viewtopic.php?p=330329 , however the images are not visible. I tried registering, but it says "bot registration detected".
37.
▲
by
kristianp
2mo ago
The "incredibly expensive Intel Pentium" image link 404s.
38.
▲
Nvidia backing $105B in financing for OpenAI data center in Ohio
(cnbc.com)
3 points
by
kristianp
2mo ago
|
0 comments
39.
▲
by
kristianp
2mo ago
Enjoying the "DOS" font of the code examples. The wikipedia page on the Pentium has multiple references to an Intel publication called "Solutions", May/June 1993. It would be interesting to see that, but can't
40.
▲
by
kristianp
2mo ago
Does it get loud and hot with a smaller MOE model like this?
41.
▲
by
kristianp
2mo ago
I assumed Chrome is memory hungry due to the JavaScript VM, but there's probably many optimisations that trade memory for speed. (Unlike this one).
42.
▲
by
kristianp
2mo ago
Curious what kind of application you used wasm and Tinygo for? Obviously not the kind of app that an LLM would usually spit out a bunch of react for.
43.
▲
by
kristianp
2mo ago
The original post: https://mastodon.social/@bagder/117061983160992793
44.
▲
OpenAI wraps $7B share sale ahead of potential IPO
(cnbc.com)
22 points
by
kristianp
2mo ago
|
3 comments
45.
▲
by
kristianp
2mo ago
Why not both?
46.
▲
.NET on AWS Open Source Software Fund
(github.com)
1 points
by
kristianp
2mo ago
|
1 comments
47.
▲
by
kristianp
2mo ago
No embeddings, just a bag of words vectorisor.
48.
▲
by
kristianp
2mo ago
Great comment. China is still building Coal and Gas power too, I wonder how they compare.
49.
▲
by
kristianp
2mo ago
> you can run data centres on grid electricity The reason they're not doing that is because grid providers can't grow their supply fast enough to support new AI data centres.
50.
▲
by
kristianp
2mo ago
It's because it's a novel solution to "I lost my phone".
51.
▲
by
kristianp
2mo ago
No headphone jack! Why Dell? The processor is a Intel Core 5 320 Wildcat lake, with 4 "low power efficient cores" and 2 non-hyperthreaded Performance-cores. It's variant of Panther Lake which has some models with Arc GPUs: h
52.
▲
by
kristianp
2mo ago
Pleasant to hear about the Object-Relational Impedance Mismatch [1], haven't heard about that concept for a long time - maybe a decade!. The reason I like Dapper [2] is that it makes you use your own sql. Edit "lets" -> &
53.
▲
by
kristianp
2mo ago
https://news.ycombinator.com/item?id=49187575 1 day ago | 256 comments
54.
▲
Chinese memory maker CXMT enters mainstream with Corsair Vengeance DDR5 kit
(tomshardware.com)
6 points
by
kristianp
2mo ago
|
2 comments
55.
▲
$1B of iPhone 18 Pro chips 'awaiting packaging' due to DRAM shortages
(tomshardware.com)
1 points
by
kristianp
2mo ago
|
0 comments
56.
▲
by
kristianp
2mo ago
If it's about the size of a Maverick as the article suggests, that's not small at all.
57.
▲
by
kristianp
2mo ago
I've been eagerly awaiting their 2nd gen HC2, which uses multiple chips to host a "mid sized reasoning" [1] model. Its due in summer according to the article, I wonder if it will ever be released in that form now. [1] https
58.
▲
by
kristianp
2mo ago
Just a note about the price of the M3 Ultra with 256GB RAM with 8TB disk, it's $16999. The 96GB model with 8TB ssd on apple.com costs $10299. So you're getting 2.6x the ram for 1.65x the price, which is an interesting comparison
59.
▲
by
kristianp
3mo ago
I'm badly remembering, but this idea was in a Sean Williams novel, perhaps "Echoes of Earth", and maybe its actually units of distance, it's probably been used by other SF writers too.
60.
▲
by
kristianp
3mo ago
Obviously that should be 10 hours a day, 10kiloseconds an hour. A hectosecond would be the equivalent of a minute. That would require different length of a second to match the Earth's rotation, so we'll have to give it an annoyi
More ›