voidspark

Gemini refers specifically to a family of multimodal LLMs, which is exactly what they are using here.

They have other models with different names used for different purposes.

https://ai.google/get-started/our-models/

300% single core performance from Haswell (2015) to today.

Single core performance has improved about 10x in 20 years

As the other guy said, top of the line CPUs today are roughly ~100x faster than 20 years ago. A single core is ~10x faster (in terms of instructions per second) and we have ~10x the number of cores.

128 MB was almost unheard of back then.

Pentium 66 - 1993

Pentium 90 - 1994

Pentium 166 - 1996

More than doubled the performance in 3 years. Two orders of magnitude from 1990 - 2000.

There was no multi-CPU Pentium. Not until the Pentium Pro in 1996.

We don't know if it is losing money. It's a private company.

He reduced the headcount to roughly what it was in 2017. At the time of the acquisition, many of the employees were in non technical roles, contributing nothing of value, posting videos about their empty work day on TikTok. Jack Dorsey admitted that he made a mistake by over hiring - more than doubled the headcount from 2017 to 2021.

https://www.statista.com/statistics/272140/employees-of-twit...

> slightly beefed up (say, dual or quad CPU) typical early ‘90s gear.

Early 90's Intel was the 486 33 Mhz. It barely had enough performance to run the TCP/IP stack at a few hundred KB/sec, using all of the CPU just for that task. I think you forgot how slow it was. Pentium II is where it starts to get reasonably modern in the late 90's. Pentium Pro (1995) was their first with multiprocessor support. It was moving so fast back then that early/mid/late 90's was like comparing decades apart at todays pace of improvement.

Because the comment you replied to never mentioned GitHub. I thought you didn’t know the difference. Linus created Git.

GitHub is not Git.

Git is by far the most widely used VCS. The majority of code hosting services use it.

All of their photolithography equipment is manufactured in the Netherlands by ASML

https://www.asml.com/en

Other companies are using the same equipment (Samsung and Intel) but TSMC has deeper expertise and got more out of the same equipment so far.

TSMC factories use ASML hardware (designed and built in the Netherlands), that actually produces the chips.

https://www.asml.com/en

TSMC is running a successful business but they're not the only customers of ASML.

In this case it means research, as it pertains to universities.

The article explains that the lab would support universities by providing infrastructure.

"LLM" is not mentioned anywhere in the article.

It's a mundane corporate feature request and quite reasonable and valid.

This is interesting

https://chatgpt.com/share/68219da9-1e78-8007-b083-8a81bfbea2...

"Dendrites can implement non‑linear sub‑units and even logic‑gate‑like behavior before the soma integrates them, whereas the standard artificial neuron uses a plain weighted sum."

"Neurotransmitter diversity (e.g., glutamate, GABA, dopamine) allows different semantics on each connection. An artificial edge conveys only a signed scalar."

That is your assumption and it is wrong.

https://grok.com/share/bGVnYWN5_ab498084-58c4-4345-9140-07b5...

Biological Neuron: Processes information through complex, nonlinear integration of thousands of excitatory and inhibitory inputs across dendritic trees, producing spiking outputs with rich temporal patterns. It adapts dynamically via synaptic plasticity, neuromodulation, and structural changes, operating in a probabilistic, energy-efficient manner within oscillatory networks.

Artificial Neuron: Performs simple, linear summation of weighted inputs, applies a static activation function, and produces a single scalar output. It lacks temporal dynamics, local plasticity, or neuromodulation, operating deterministically with high computational cost and fixed connectivity.

Neurons are surprisingly not simple. Vastly more complex than the ultra simplified model in artificial neural networks.

It was due to many reasons, including taxes.

I live on the opposite side of the world from the US. I’m not American. No freedom of speech in my country.

I think you got the facts mixed up. The person who wrote the op ed was not an American citizen and they were not jailed, they were ordered to return to their home country. They were on a student visa which is conditional and it is a privilege, not a right. They raised some flags and the student visa was revoked.

> How is making a country have hate speech laws somehow causing wars to emerge?

http://en.wikipedia.org/wiki/American_Revolution

Yes. “Hate speech” has a specific legal definition in the US.

Holocaust denial is not hate speech. People should be allowed to question a historic event. People should be allowed to think.

In the US that is free speech which is allowed under the First Amendment.

You have no idea. Infosec in my company would ask for this.

It helps to prevent accidental sharing of confidential information.

> Not a museum

That is two museums. It's in the title: "Hands-On Vintage Technology Museums" and mentioned all over the front page.

> Index funds will sell if the valuation drop

No. The big index funds fully replicate the index.

They will drop Tesla only if Tesla is dropped from the index.

Tesla is currently up by 1150% since 2019.

Archive, library, and museum are three different things.

The majority of Tesla shares are held by institutional investors, and most of those are passive investors, holding it long term. Tesla is included in multiple index funds, including those tracking the S&P 500 and Nasdaq-100.