They already picked out a interconnect standard it is called 100 Gigabit Ethernet. They can interoperate with multiple different vendors using the sfp28 ports on the switch. I bet they are even working on new programmable hardware for a next gen 200 Gigabit Ethernet fabric with oxide controlling the entire stack from hardware to software

Ethernet would add way too much latency to this application.

ever heard of RoCE? what latency are you thinking of in the Spectrum-X 800Gbit RoCE era?

Not sure Tenstorrent agrees.

First half life one in browser now we have half life 2! I guess it’s that time again Mr Freeman

No? The website has units available here[0] in the US. I think this article is old. There must have been a very small first shipment that got sold out before the next batch came in

[0] https://store.steampowered.com/steamdeck

the article was posted on 5/28, three days ago. both steam decks are showing as available for me too.

If you just install CUPS in a virtual machine (emulated in wasm on the web) what patches do you need to share?

See above, that's my mistake.

The same you have to share if you don't use a virtual machine, this isn't hard.

If they reverse engineered the drivers then why do they need a virtual cpu and a Linux kernel to run them. Is this reverse engineering or just installing software in a weird environment?

Speaking of not just gluing stuff together with usb/ip could one make a virtual WebUSB host kernel module that could be used by the Linux kernel USB stack? They most likely would not want to do that because then all of the code would be GPL and would have to be shared with the public.

I don't think a usb host driver is necessarily tainted into being GPL? But if it is, plenty of non-gpl oses that can run SANE.

A another take on this problem is zoo.dev . They wrote a brand new from scratch cad engine that is driven a custom openscad style language called kcl.

Then then have a trained llm that has can generate kcl to either create new parts or act as a llm assistant for changes to existing parts.

It’s neat that llms can do 3-D but I wonder how much of the problem is integration.

It’s ddr2 so it is not really affected by the ram crunch

Sometimes these demos enable caching on the reverse proxy. So then for these tiny demo html pages you request, you may not even reach the fun tiny computer it is supposed to demonstrate.

Manned-Unmanned teaming is not a new concept created in the last couple months to placate fighter pilots in the age of ai. With 5th generation fighter using datalink they to use the active radar in far away AWAC planes for targeting so the stealth fighter can get closer to the enemy without breaking cover by turning on active radar.

If you can outsource the radar on a jet it is not a huge leap in logic to put the very hot missiles onto a unmanned aircraft. All of these concepts where written up 20 years ago by both china and the US

> With 5th generation fighter using datalink they to use the active radar in far away AWAC planes for targeting so the stealth fighter can get closer to the enemy without breaking cover by turning on active radar.

What benefit does a human pilot offer in this case? Are going to be using their eyes to track their location or see a Chinese fighter jet launching a standoff antiair missile at them? Drones can do AWACS and deep sensor roles, with a pilot and sensor operators far away from the planes.

> All of these concepts where written up 20 years ago by both china and the US

Defense contractors and the gov planners are often the same group of people, the same small community, there's not that many vendors. They show up at defence trade shows and see what industry is offering them. They tend to stick with the same big safe companies that change slowly. Bold ideas are infrequent. The smaller countries can take those risks easier than America.

Don't you have that backwards? If AI gets so good that it can replace all human labor, will capital like money and data centers be the only moat left?

> If AI gets so good that it can replace all human labor, will capital like money and data centers be the only moat left?

If AI gets good enough to replace all human labor then actual physical moats to keep the hungry, rioting replaced humans away will be the most important moats.

Did you see those Chinese robots from last week? I’m pretty sure they’ve got their moats covered

Which is bought by money in the first place, see billionaire doomsday bunkers. The poor will not have such a bunker.

Unless they intend on generating their own oxygen to breathe, I don't see how these bunkers stand a chance.

Fortunately they do.

For how many weeks? Or months? Or years? Then what?

Money is useful mostly for hiring human labor to outcompete others, e.g. Satya Nadella has 100K employees under his command, you don't, so you can't realistically compete with MS today - this is their main moat.

If AI renders human labor a cheap commodity (say you can orchestrate a bunch of agents to develop + market a Windows competitor for $1000 of compute), what used to be "Satya + his army vs. you" now becomes mostly a 1:1 fair fight, which favors the startup.

Frankly, you have a pretty good chance of displacing windows right now. You should go for it.

How powerful is the device you wrote this comment from? On prem or self hosted affordable inference is inevitable.