I think the reserves have more complex dynamics. The law requires 250 million barrels to be kept but the pumping rate decreases as more is pumped. Also, there's various estimates that indicate that the storage areas will lose structural integrity and start to collapse at levels higher than 250M barrels. I'm not sure anyone really knows about the levels for sure and the best estimates are probably kept secret for security reasons.
I don't think that's accurate. If tokenization takes say 10ms and the rest of the inference steps take 50 ms then, improving tokenization will improve the time to first token but won't affect throughput much. After the first token, the inference steps effectively hide the tokenization time.
It's hard to get more ECC guaranteed than the mobo manufacturer having ECC ram on their QVL (qualified vendor list) for the motherboard and indicating that ECC is supported in the specs. I suppose you can just get a complete workstation with ECC from HP or DELL and that'll be guaranteed to work but you'll certainly pay a hefty premium for that.
Plastic cases might actually be a bit better than steel depending on the radiation type. Neutron radiation is more effectively shielded by interacting with light atoms so stuff with lots of hydrogen atoms like water or polymers have a better chance of interacting with neutrons and slowing it down so that it can be stopped. Steel is probably better for gamma radiation or charged particles like muons.
But overall, pc cases are probably too thin to appreciably affect the radiation levels. Cosmic rays and their secondary radiation is fine with going through buildings so a 1mm thick case isn't going to do much.
Laptops are pretty much unavailable with ECC especially with soldered memory becoming common. On the desktop side, there's multiple motherboards that explicitly note that they support ECC memory. AsRock and Asus both include ECC memory on their QVL lists. I'm not sure how many ITX boards support ECC though so getting it on something that's not a tower might be tough.
But it's definitely possible, I have an ECC setup on my desktop and it's caught a memory error or two over the last few years.
It's preferred but not having it still works if you're willing to leave some potential avenues for errors to get in. However, it's not like not having ECC is a recipe for a corrupted FS.
AMD has historically supported ECC on all it's chips. That's changed in the last gen or two where some of the lower end APUs don't support ECC but the Pro APUs do and all the ryzen stuff (5, 7, 9) support ECC if you get a motherboard that supports it.
It also varies depending on where in a time zone you are. The eastern and western sides of a timezone could see sunrise and sunset at significantly different times.
I think that glass flowing thing is a myth. The manufacturing method back then resulted in uneven panes and they were installed with the thicker side at the bottom.
At upper undergrad and grad levels, it probably would have improved a lot. The issue is that a lot of the why requires quantum mechanics to really explain and even that becomes intractable extremely quickly. Like you can probably do the analytic solutions for hydrogen atoms and electrons but once you get to helium or past that, you basically need to use a computer to do numeric calculations and even there, you are very quickly using approximations instead of solving the quantum equations directly.
I don't think that's actually the case. If you get an tax attorney to document your treatment and indicate that they believe it's legal (e.g. a tax opinion letter), then you're off the hook for penalties and criminal liability (assuming the tax attorney did everything properly). You still have to pay the difference though.
An accountant may not save you from financial penalties but I believe that they are liable for them so you can recover it from their insurance.
I think consumers do care when they have to throw away a device or pay for a repair because the battery health has significantly degraded. This is really similar to the argument against having a standardized charging port or requiring usb-c charge ports for cellphones.
Given the prices and shortages, I'd think people would keep and use the current gen stuff till it drops dead. It may not be as good but it's paid for and given the prices for next gen stuff, it's probably worth using for another cycle or two.
It's been in the planning stages for at least 5-8 years at this point. The instruments and upgrades have been tested and validated so it's just putting stuff together and then validating/calibrating. Stuff can still happen like the splice issue and repairs.
There's at least 3 copies of all the data and possibly more spread around the world at different Tier1 and Tier2 centers. The T1 and T2 centers are very likely using different filesystems and everything is checksummed pretty much at every major step (data generation, before and after any network transfers, before analysis, etc).
A scissor lift or some other thing seems a lot simpler than trying to allow for a bipedal robot to knee to reach things close to the ground. Or if legs are the best option, why not a quadruped? Much more stability with a lot of the same advantages.
Sure, but I'm not sure humanoid robots is the best form factor for this. E.g. something with a wheeled base makes movement calculations a lot easier since you don't need to deal with balance while moving.