Many people, including myself, don't write code everyday, and it's not something we have to be good at. Any assistance is appreciated since the alternative is to bang away for hours, copy stuff from the Internet and try to make it fit the problem I'm trying to solve, resolve bugs etc until somehow the code works. The new tools are not a perfect time saver, but they still save much more time than they "waste".

Perplexity.ai uses several different models (automatic or manual selection) and gives sources.

Similar experience. One of my children had been doing Duolingo for 2 years in a single language. Had no idea how to say the number 7,8, or 9, by themselves or in any other context.

A subsea power cable that long is a bit unusual, as in it just isn't done unless there is no better alternative. It would be very expensive. The other options might also be expensive but you can buy an awful lot of power plant for 32 billion dollars and you wouldn't have the risk of an anchor incident. The insurance alone on a cable that long and expensive would be eyebrow-raising.

And fiber in my area. An upstart fiber company did the hard work of putting lines in my whole neighborhood about 2 years ago. This prompted AT&T to finally wake up and they upgraded their copper lines to fiber last year. The new fiber company offered no caps and a flat monthly fee and so Comcast quickly lost at least half the neighborhood.

There are benefits to higher refresh rates beyond just reducing latency or reaction times. It improves motion clarity, that is, it reduces motion blur. There are other ways to do that, like black frame insertion or backlight strobing, but those have severe negative impacts on brightness, and framegen is a way to improve motion clarity without reducing brightness. Framegen shouldn't be about improving low framerates. It should be about taking something like 60 FPS and turning it into 240 FPS for better motion clarity. Or higher, 120 FPS into 480 FPS, where the latency penalty will be minimal. Taking advantage of those high refresh rate monitors to improve motion clarity.

4K is probably the limit of what we'll ever need on TVs, because that's well past the "retina" resolution even for TVs much larger than we have now at couch distances. There's a reason why 8K televisions failed and were mostly abandoned. There's a case to be made for monitors going higher than 4K, though. You're much closer to those. They fill much more of your field of view.

I'd dispute that. First because the first GPU to support full bandwidth DisplayPort 2.X didn't come out until 2025, and second because DisplayPort 2.X can't support uncompressed 4K 240 Hz with a 12-bit colour depth. It can do it with compression, or it can do 10-bit colour depth, but not uncompressed 12bpc. That's not a big loss, I'd bet you that nobody in practice could tell the difference between 10bpc with temporal (or even spatial) dithering (which GPUs do automatically) and 12bpc. And the consumer HDR formats that use 12bpc (like Dolby Vision) use chroma subsampling.

So your argument is that technological advancement is pointless because only high-end products can benefit today?

Multi-monitor is not a feature that HDMI has ever offered. That's always been a DisplayPort thing, and I don't see anything in this article about adding multi-monitor support. However, we do need more bandwidth for higher refresh rates. Many monitors on the market today exceed the 48 Gbps that HDMI 2.1 provides, and fall back on DSC or DisplayPort to do it.

It will become useful yesterday. Monitors that exceed the bandwidth of HDMI 2.1 have been on the market for some time now. They currently either rely on DSC or DisplayPort (or both).

Current 4K240 monitors require around 129% of the available bandwidth that HDMI currently provides. When operating in 10bpp HDR, they require 161%.

Considering that lower resolution monitors on the market today go up to 540 Hz, the appetite for increased connection bandwidth is insatiable.

In every car except Teslas. In a Tesla, the lane assist will not allow deviations from its chosen path. If you try to correct it, it will fight you until you do it strongly enough, at which point it will turn off entirely.

There is no "encourage" in a Tesla.

I'm sure it uses some very cherry-picked numbers.

Let's say a query uses 3000 watts for 10 seconds. That's probably within an order of magnitude because a 300w consumer GPU can do quite a lot in 10s, but the bigger models will be using a bit more hardware.

If every bit of that 0.00833 kWh of that went into boiling some water, you could vaporize about 13 grams or about 1/40th of a bottle of water.

We then need to consider that the data center cooling system will add about 1/3 more, and that to produce that electricity in a 40% efficient (coal) power station entirely cooled by evaporative cooling would a little more than double it again.

But we should also consider that not all thermal power stations are that inefficient, that even in evaporative cooling setups significant heat is removed by conductive b means, most thermal power stations in the US West are air cooled, and a lot of power will be coming from renewables. On a cowboy calculus basis you're back to 1/40th of a bottle or less.

One could argue that this doesn't include the energy to train the model, but amortizing that over all the queries made during the economic lifetime of the model probably doesn't add much more than 1x the power used by a single query. I would say that shouldn't be included in what a single query costs anyway.

In any case it isn't anywhere near using a bottle of water.

It is a big deal though. Pressure vessel and piping integrity are not typically major issues for a rocket this far along in development. Their QC program must be in a sad state for something like this to happen. It points to a culture that emphasizes speed and not quality, which completely aligns with how the CEO works and how the company is run. And that won't get better quickly.

Steeeve has been spouting nonsense, blaming the pilots without having any concrete evidence. Then his next video suddenly comes up with "oh, new evidence, the RAT deployed" even though there had been reports of this within hours of the crash.

He has zero credibility at this point.

It's very common for pipes to freeze in the winter. Most policies will cover the consequential damages of this even if the damaged pipe itself isn't for some reason. Frozen pipes are the justification for recent steep insurance increases in Houston of all places.