rerx

In defense of the comment you replied to: Research into treatments with Psilocybin or LSD was in quite a hiatus for decades after the substances were banned in the 1960s or 70s.

Don't know about the US. From what I can tell Samsung and LG appliances are certainly less common in Europe than the European brands (there's also cheaper brands like Beko or Gorenje).

The camera based object (and poop) avoidance actually works pretty well in my Roomba j7+, bought in 2022.

The cloud-based software for everything else has degraded in quality, tjough. I'll probably upgrade to a lidar-equipped competitor model if this continues to get worse after this bancruptcy.

BSH (Bosch, Siemens, Neff brands) have their main production sites in Europe, a lot in Germany still. But of course China as well. https://www.bsh-group.com/about-bsh/bsh-worldwide/

Miele (at a more premium price point) production is even more concentrated in Germany. https://m.miele.com/en/com/production-sites-2157.htm

(Edit: No replies after 8 hours, but of course they then came in quickly after Europe woke up..)

> There’s something about shoes that makes my legs tire faster.

Less weight on the parts of the body that move most during a run is an obvious benefit of not wearing shoes.

Do you have an opinion on the sleep tracking in Garmin Fenix watches?

I disagree on your 14th century building picture being "nothing special" to Europeans. All the major German cities that I've lived in were bombed to ashes in WW2. Most buildings are post war construction. Picturesque towns like Tübingen with conserved old buildings are an exception in Germany and attract local tourism for this reason. Split is a UNESCO world heritage site and pulls in tourists from all over the world.

Huge archive that's heavily biased towards text adventures / interactive fiction. There are only three titles published by Lucasfilm Games, and they predate Monkey Island.

> "Basically nobody writes CUDA," wrote Keller in a follow-up post. "If you do write CUDA, it is probably not fast. […] There is a good reason there is Triton, Tensor RT, Neon, and Mojo."

> Even Nvidia itself has tools that do not exclusively rely on CUDA. For example, Triton Inference Server is an open-source tool by Nvidia that simplifies deploying AI models at scale, supporting frameworks like TensorFlow, PyTorch, and ONNX. Triton also provides features like model versioning, multi-model serving, and concurrent model execution to optimize the utilization of GPU and CPU resources.

> Nvidia's TensorRT is a high-performance deep learning inference optimizer and runtime library that accelerates deep learning inference on Nvidia GPUs. [...]

Keller was speaking of OpenAI's Triton (https://openai.com/research/triton), a Python-like language that is compiled to code for Nvidia GPUs, but Tom's Hardware mixed this up with Nvidia's Triton Inference Server, a higher level tool that's really not a replacement for CUDA and not directly related to the Triton language. Easy to confuse these if you are a writer in a hurry.

Interesting, I guess this may have resurfaced here because a German translation was posted today, https://www.golem.de/news/arbeit-scrum-nervt-2401-180930.htm... ?

Freevee disgusts me. I don't think there's any paid option to watch Bosch: Legacy without ads.

The Marvelous Mrs. Maisel was fabulous. They had lot of good shows in the prestige comedy or dramedy (is that still a thing?) genre. Transparent, Red Oaks, Mozart in the Jungle, Fleabag (bought from BBC)

Earlier submission today where this is already being discussed: https://news.ycombinator.com/item?id=37610500

Absolutely a country thing!

In countries where the Amazon Prime membership (originally this was for fast free shipping only) is available, Prime Video is included with that: https://en.wikipedia.org/wiki/Amazon_Prime#Availability

Outside of those countries you can typically get Prime Video as a cheaper, separate subscription: https://en.wikipedia.org/wiki/Amazon_Prime_Video#Availabilit...

In Germany, for instance, I don't think you can get Prime Video without the full Amazon Prime subscription. In countries where the shipping benefit is more costly (like the US) they may offer different membership tiers.

I just recalled that Amazon increased the price slightly last year when they launched Rings of Power.

Typically I subscribe to these services for just a few months at a time, depending on what's interesting. Harder to do with accounts that are shared by a larger household, I guess.

While this is true, Netflix hiked prices massively over the years. Their 4K/HDR offering is 17.99 EUR per month here. Amazon Prime including Prime Video is still 8.99 EUR/month, cheaper if paid yearly, without any surcharge for 4K/HDR. So that's more than twice the price for a comparable streaming service.

Features a photograph of the lovely boxed set that's available for 35 euros.

I think that's a common myth. There's no separate category for frozen pizza in Nutriscore. The categories are: general food, drinks, cheeses, and oils/fats.

Pizza producers can game the system somewhat, however, by adding ingredients that are counted positive (like protein or fiber) to partially offset "negative" ingredients such as sugar or saturated fats.

Dairy lobbying has led to cheese as a separate category and that milk is counted as food, rather than drink.

I only drive carsharing and rental vehicles occasionally, so I probably have more limited exposure than most, but most German gas stations that I can think of have quite a lot of free space. There's usually room to park a couple of cars for a short while even while they are not taking gas (to go to the store, to use a coin operated vacuum, ...). Maybe it's regulations?

True. Memory bandwidth may be the most limiting factor.

GPUs will be better optimized for large matrix multiplications than CPUs, by design.

And you need to the inference again and again, not just a single time (like your training).

I don't think the article fully applies to large language models (LLMs).

> Inference will Dominate, not Training

This rings true. While LLMs will be fine-tuned by many, fewer companies will train their own independent foundation models from scratch (which doesn't require a "few GPUs", but hundreds with tight interconnect). The inference cost of running these in applications will dominate in these companies.

> CPUs are Competitive for Inference

I disagree for LLMs. Running the inference still takes a lot of the type of compute that GPUs are optimized for. If you want to respond to your customers' requests with acceptable latency (and achieve some throughput), you will want to use GPUs. For "medium-sized" LLMs you won't need NVLink-level interconnect speeds between your GPUs, though.