magiccupcake

Don't forget the fundamental scaling properties of llms, that openai even used as the basis for strategy to make chat gpt 3.5.

But basically llm performance is logarithmic. It's easier to get rapid improvements early on. But at later points like we are now require exponentially more compute, training data, and model sizes to get now small level of improvements.

Even if we get a 10x in compute, model size, and training data (which is fundamentally finite), the improvements aren't going to be groundbreaking or solve any of the inherent limitations of the technology.

Even accounting for inflation arcades should be cheaper.

The compute hardware costs much less and is much more power efficient.

Other power hungry features like lights and displays are both cheaper and more power efficient.

The argument that they still need to be expensive makes so little sense, other than the physical space they occupy.

As someone who learned way more about pans than I really want to know, let me say that a good cook can make good food in any pan, however some pans are more suited to tasks than others.

First off, searing meat in a non-stick pan (traditionally Teflon) is a bad idea, the pan can reach temperatures that produce toxic gases, and are known to kill birds that are more sensitive to them than we are. The coating that makes them nonstick isn't very durable and will at most last a few years before being useless. While other kinds of pans are likely to outlive you.

Other common pans include cast iron, stainless steel, carbon steel, and ceramic non-stick (non-toxic, but are delicate)

Specifically for searing meets, my favorite is stainless steel. It holds heat similar to cast iron, but is slightly more conductive and can transfer a lot of heat to sear meat. Meat also literally bonds to pan and can be used to make great flavorful sauces with deglazing. Cleanup is easy, if anything is really stuck just boil water in it to loosen. Alternatively stainless steel holds up decent in a dishwasher. Cleanup can't be easier than automatic. However, stainless steel is still quite heavy.

For general purpose cooking my personal favorite is carbon steel. It's seasoned like cast iron and can be quite nonstick, but is much lighter making it feel very similar to nonstick pans, which are made with aluminum.
I won't lie, seasoning has a learning curve. Seasoning is very tough under some circumstances, and very delicate under others. Notably acid will eat the seasoning away.

Cast iron is great, but it is so heavy that it is inconvenient to use.

All will work with induction, except for cheap aluminum nonstick pans

We already have superweapons In the form of nukes, this is a lot harder

They're human like everyone else, and try to use language that is specific and descriptive. In this case the word direct observation has become to mean something very specific In the field of astrophysics. It's not out of malice or anything, just results from the difficulty of scientific writing, so you use words that already have established meaning.

Direct observation ≠ direct detection

I would argue that dark matter is much more based on indirect observation, things like rotation curves and baryonic acoustic oscillations.

Confusingly, direct observation does not mean the same thing as direct detection.

This study "directly observes" a hypothetical dark matter signal. However this is distinct from direct detection experiments, where a dark matter particle is found in a collider.

I am a physicist, studying dark matter.

Firstly, It would be nearly impossible to prove that dark matter definitely does not exist.

And secondly, there are no alternatives to dark matter that come even close to explain our universe as successfully as dark matter.

That doesn't mean it's right, but any explanation without dark matter is not favored IMO.

Alternatively, if you can't remove the modem, find and remove the antenna. And if you can't remove the antenna try and surround it with a metal, like aluminum foil.

Well I'm probably wrong then, framework said they couldn't get good performance and maintain signal integrity with upgradable memory for the Ryzen Max cpus, so this is likely discrete Cpu and GPU. Probably all soldered in the same mainboard though.

This seems to blur the lines between desktop and mobile APU's, but I would bet that's it's closer to a higher clocked mobile chip, than it is to desktop. The only reason I think this is the case is due to the similarity spec wise with the Max 385, and that it's semi-custom.

If it was just a 7600x CPU + 7600 GPU I think they would have just said so. It could be separate CPU+GPU, but I think it might be possible that it is built more like a SOC, where the GPU is just given its own dedicated VRAM.

Looking at the hardware of say a PS5, it has 16 GB of GDRR6, the same as the Steam Machine's VRAM.

If everything is soldered anyway, there is no reason to have separate chips for CPU+GPU, especially if that hardware already exists like the AMD Ryzen AI Max line.