KeplerBoy

Well the helmet is also a consumable (you should swap them every few years and of course after every crash). But having the visor integrated in the visor is awkward.

Kind of surprising this stuff still is little more than a concept. 12 years after google launched and scrapped it's glasses there are still no well established alternatives for cycling, which is such an obvious market. Everyone is wearing glasses, everyone has a computer mounted to their handlebar, let's integrate them together already.

I know quite lot of people who could fork vs code and write a passable agent harness. I don't know one single person who could build a humanoid robot and I majored in mechatronjcs.

Full seats don't mean much revenue in the cities I frequent. In Vienna for example the ticket for the whole year was 365€ until very recently. Now I live in a smaller city where it's 280€ per year for residents.

I love it and I am perfectly happy with my taxes subsidizing this for others but there's no way they're making money off this.

Yes and they do in many jurisdictations. In Austria gas tax is used for road maintenance, on top of that there are tolls for highways.

Public transport is not sustainable from ticket prices alone anywhere in Europe.

That's likely the datarate of the ADC chips. You would downsample them directly on the FPGA board and maybe perform an FFT or similar transform. 16 TB/s across a few dozen FPGA boards is nothing crazy. After some early stages in the signal processing you might transfer 1 or 2 TB/s over ethernet to the servers. Entirely feasible considering we have 800 gigabit/s ethernet.

There's no way you could feel even a millisecond in latency difference and that would be a lot.

True, it's a weird thing to say. I am in no position to rank them, I assume they are both excellent at their niches (granted bellard seems to be interested in a lot of niches) but it never hurt anybody to be humble in this position.

LLM output is just very predictable. If you spend 200$ on that /goal you will get some output. It may or may not work perfectly, but there will be a repo with some progress. If you specced the goal well and it is feasible, a decent model will most likely get you decent progress.

Also who would take on any of these projects for a meager 200$? Most of that stuff is borderline interesting, clearly not interesting enough for the people proposing the things to start working on them themselves.

Now Datel is a name I haven't heard in a while. Reminds me of those action replay devices which modified the console's memory to cheat. Ahh the nostalgia.

It's not just intended exclusively or limited to education. Many products ship with compute modules inside.

I bought the NDS version back in the day and let me tell you it was not worth the 40€ of my pocket money.

Do you really think Niantic didn't make sure their actual partners didn't agree to their business model?

Imo that's a big win. The LLM just gaslighting you into suboptimal approaches was insane.

It is cool, but still weird that it get's very basic stuff wrong like mapping the cursor coordinate to the canvas. There's some y-axis scaling issue.

That has never been true, unfortunately. The 2005 top500 was led by bluegene/L achieving 280 FP64 TFlop/s.

Apple is talking about 17.5 FP16 TFlop/s on the iphone 17 neural engine. So 20 years later we are still nowhere near, not even at reduced precision.

Isn't that just expected for modern macOS devices? They have been on arm64 for 6 years now.

I think a majority of old rich people is pretty much unavoidable at a return tour.

How does Darwin matter more than Windows?

That's literally the same chip. With pricing being whatever, who cares which one they revive.

Which problems would that be? Nasdaq crashing by a few percent and a major player to go under. Seems almost inevitable at some point.

At least it seems to not get worse.

That's true. You can just point the agent at a python function and tell it to slap some fastapi around that and spin up a frontend.

Thanks for the correction. I should have looked it up; I only remembered it being somewhat odd.

It's inevitable. What might be a prosumer device today priced at 4000$ will be a regular consumer device in 10 years and models only get better.

Local models today are fine for a lot of mundane tasks and will continue to be so. The use cases where paying for frontier models is worth it, will continue to shrink for folks not doing frontier work.

To be fair the connectx-7 in the spark can't even push 2x200 Gbps since it is connected via 4 pcie lanes.

It means nothing. The LLM thought it was edgy.

Also RAM was still quite a bit cheaper when the DGX was announced back in early '25.

they also announced a GB10/N1X windows desktop mini PC.