But can’t you also buy 5 American cars for the average price of American car? Just buy used 🤷‍♂️

So what? Gen AI can build a new database and populate it with data 🤷‍♂️ /s

Coding sounds trivial. And it’s a gateway term to “vibe coding”.

In The Martian, the hermes ship used to go between Mars and the Earth used this method. In the hail Mary Project the ship split in two halves joined by cables to replicate earth gravity, by separating the two halves by large distance.

It’s a Sci fi concept so far, there must be a reson it’s not implemented.

In my experience it works across circuits, but speeds aren’t that great. Just connect them to the same outlet first for pairing. I’ve heard about this method working between apartments too.

Repeaters can add fuel to the fire if there’s already a lot of access points around. Especially on 2.4ghz which has limited number of channels. It can be helped by reducing the transmitting power on all devices.

Id rather recommend the wall plugs (they connect to wall power socket and transmit Internet over power wires via RF), if you can’t run Ethernet.

Define celebrate. Do they go to church? Do anything religious on those days?

If it were proven, and you could identify your previous identity, would you be able to claim your wealth and possessions from previous life?

Good point. What even is reincarnation if you can’t remember a thing?

Peak skeuomorphism.

Smartphones and mobile phones used to be inventive too.

PCs aren’t faster, they have more cores, so they can do more at a time, but it takes effort to optimize for parallel work. Also the form factor keeps getting smaller, more people use laptops now and you can’t cheat thermal efficiency.

Everyone. One person is too lazy to write a message, the others can’t be bothered to listen to the whole thing 🤷‍♂️

The transcription should be attached to the audio recording so if the sender cares about it being correct they should be able to comment or add correction.

This is where AI could be useful. Transcribing and summarising the voice recordings.

I like the size of the display

Fiat Punto.

Exam cars in my area were puntos so schools tended to teach the same model.

I think we can assume it’s nvidia H200 which peaks at 700W what what I saw on Google. Multiply that by the turnaround time from your prompt to full response and you have a ceiling value. There’s probably some queueing and other delays so in reality the time GPU spends on your query will be much less. If you use the API, it may include the timing information.

There’s a huge difference between a model you can run locally and a chatgpt model.

The question is about per-prompt, so number of users is not relevant. What may be more relevant is number of tokens in and out.

If anything, number of users will decrease power use per prompt due to economy of scale.

I hear there’s significant space left

I like this. I also like yaml, I’ve had very few issues with it and it’s nicer to work with than json.

Json’s lack of support for trailing commas and comments makes it very annoying for everyday use.