yea, IDK how it works as I’ve never had a computer back then, but the quoted reply makes it sound like getting a sound card would take load off of the CPU.

Not in the same way, as you aren’t using the integrated gpu at all if you get an external one. I guess if you’re talking about shared ram this makes sense though.

I’m using a cheap one of those from amazon for my headphones on my laptop because the audio jack suddenly stopped recognizing when headphones were plugged in. (although I still get a dmesg error log when I stick a q-tip in to the jack? If anyone knows how to debug this, please tell me)

Yes, it can be pretty impressively good. It ‘just works’ and can print at amazing speeds. Note the small bed size and lack of ability to tinker though.

Ender 3 IMO is mostly obsolete now, nowadays you can easily get something for a similar price with better features across the board. Yes, an ender 3 will work, but why get one if you can get something better for the same price? (video on the topic)

Do not get a base ender 3. Get something direct drive with input shaping and auto bed leveling if you can. If you want a printer that “just works” where (after some setup) you can just press a button and get a print, auto bed leveling is a must-have IMO. I don’t have one myself, but I think the neptune 4 for ~$220 is a good option, anything else with similar specs would also work. Look at some reviews on youtube.

Yeah, it was like a boarding pass printing machine though, which seems like a weird use. You still had to get the pass scanned later.

I don’t care about hoverboards, but a great side effect of their mass production is that you can get a pretty decent brushless motor now for very cheap. I also saw a video about a hack you can do to make it run better at higher RPMs. You can get one of those hoverboard motors for like $30 on ebay and pair it with a $25 ODrive clone from aliexpress. Its good for probably 10 nm of torque at 36v 10a.

I think its largely the chip manufacturers, but ARM is still making money on licensing fees for Nvidia’s new ai chip (with an integrated 72 core arm cpu) for example

ARM is in the perfect place where, if a company using their architecture succeeds, they get tons of money, and if the company fails, they lose nothing.

Apple has published papers on small LLM models and multimodal models already. I would be surprised if they aren’t using them for on-device processing.

it seems like the physical limits in the strength of cubes are probably becoming a problem lol

Those are some pretty beefy motors. Its interesting that they don’t have a link to a product page for the motors on the video, as I assume that was the primary justification for the project.

No, I was wondering about the side of the guardrail facing the canal. If you look closely, there is a metal strip on that side too, which is not something I’ve seen here in the US. Maybe it’s just there to add extra strength? I guess traditional guardrails rely a lot on the guardrail deforming and acting like a net, which might cause a problem when the edge of the canal is so near, IDK

I doubt its ai - I don’t think AI would get that guardrail that consistent. It has that little hole in the exact same place over every bar, even the distant ones. Although, why is the guardrail double sided when there’s no road on the other side?

Nevermind, I found it on google maps.

IDK, for me it would just be another thing to worry about, adding extra weight and cost with almost no benefit.

Doesn’t a 3070 have less than 7k cores? A UHD 750 (relatively recent iGPU) only has 256.

And I don’t know the structure of JSON that well, but can’t tokens be made of multiple chars?

If I was lost in the woods, a human would be more likely to have cell phone service or know the way out. And realistically, it’s just going to be some hiker or birder, which from what I’ve seen are generally welcoming groups.

It really reads like it was written by AI. I’ve never been to linkedin, maybe everyone talks like that there but it really sounds like it was written by ChatGPT

At least meta AI actually releases most of their models, and publishes research papers about the ones they don’t.

The gap between the wheels in the trolley in the image is much smaller than the gap between the tracks in the image. Still, depending on the amount of friction of the road surface and the wheels, it could either stop safely, tip over and kill people inside, or keep going and run over everyone. I assume it isn’t going that fast though because that turn is very sharp and the trolley would derail itself on that turn if it was going at speed.

They would only die if it crashed/tipped over. The image says the train would stop safely, so I assume the friction and center of mass are low enough to prevent it from tipping.