Well, I am not sure if it's easier in a glider but the approach and landing of an aircraft under various conditions is about 50% of PPL training.
(For the non pilots here, getting the private pilot license is equivalent to learning your first programming language/shipping your first app). All other accreditations and ratings are gated behind it.
Delaware is widely regarded as having strong protections against personal liability for corporations. Some advisors say Nevada has better protection against personal liability. This is arguably true, but the differences are very unlikely to be relevant to founders trying to build a legitimate business. Some people have also observed that Nevada has an adverse selection problem in that their unusually strong protections attract bad actors. As a result, it's possible that if you incorporate in Nevada, you'll be inviting closer scrutiny.
Nevada protects founders from shareholder lawsuits. So if someone defrauds or intends to defraud shareholders - they are more likely to prefer Nevada. To be fair, a lot of things can turn into a shareholder lawsuit in Delaware.
Playing around with assembly language on a RISC CPU is pretty fun, because there's only a few instructions to keep track of. Assembly language is like a puzzle game, with discrete values for the size and runtime of every instruction, making it easy to compare execution strategies and choose the best one. Something like AMD64 has so many instructions that it's difficult to figure out when to use each specific variation, let alone what resources they use. RISC, on the other hand makes everything straightforward.
Treating every programming task like a speed run challenge isn't particularly productive, so playing around with it theoretically doesn't provide a useful skill, but for the tasks where resource usage does matter that much, hand written Assembly language really does shine.
> making it easy to compare execution strategies and choose the best one
Once your CPU has pipelining and out-of-order execution (which they will, of speed is an issue), that easy gets out, no matter what instruction set your CPU uses.
Fast code also often needs to be adjusted depending on the size and timing of various caches. That’s no different with RISC-V.
I also expect that RISC-V will have warts in its instruction set, if it doesn’t already have them.
For example, there’s the idea of instruction fusion, where the CPU treats a pair of instructions as a single one. If you’re writing assembler, whether your specific CPU fuses an instruction pair can affect what the fastest code is.
Pretty much this, in superscalar, OoO, uOP era with branch predictors the whole RISC vs CISC debate is just matter of amount of silicon used for decoders on die.
Playing around with 68k assembly is actually much more fun. These days all the logic is absolutely dwarfed by caches in terms of chip area. This means using RISC does not really make as much sense today as it did in the 80s. That's why the most popular architectures are still CISC (assuming ARM64 can not really be called RISC).
Personally I would be more interested in a fully orthogonal instruction set like 68k but without the insane addressing modes and a better binary format.
Playing around with assembly on a RISC machine is fun but that's only because RISC CPUs are the only ones where you do actually know the runtime of every instruction. The problem with amd64 is not that there are so many instructions (…just don't use the weird ones) but that the normal ones are optimized in ways that make writing assembly with them difficult if you want to understand why something performs the way it does.
Hmm, can we make a MTG style card game with Assembly instructions? RISC is a faster deck with smaller but quicker cards, CISC has heavy hitters but warms up late etc. AVX-512 comes into game and vector arithmetics the bazinga out of the enemies?
Every once in a while, eBay emails me out of the blue and asks me to update my personal details, with a link to a web page.
I always assumed they were phishing scams, but I looked closer at one, and it is a real link too a real page on their site. It's like they're training people to fall for phishing scams. One of them even displayed the name of a variable, instead of my user name.
I have had two banks (Lloyds and Barclays) phone me and ask for personal details to verify me over the phone. Again, training people to fall for scams.
Then there are all the links that go to other sites to track clicks or because they have a separate domain for some reason. Again, training people to fall for phishing.
The ones I have used do not accept photos, they require real-time video with the front-facing camera and they prompt you to move your head to face different directions on command. Not impossible to attack, I'm certain, but it's tougher than simply uploading a photo.
on desktops you can have virtual camera, if you can generate video fast enough wen AI you can ask to edit it according to instructions. Definitely tougher but I'm sure someone will offer services or software like that.
I have a kitchenaid dryer from the 80's with multiple selections for dryness levels and it works great every time. I can leave the clothes a little moist if the air is dry and I'm going to hang them immediately or set them to completely dry, in case I'm going to be away when they are ready.
My parents' modern dryer is awful, just like yours. The craziest part is that it starts a countdown timer when there's tens of minutes left, as though the designers new the sensor was awful and decided to add some extra drying time to cover it up.
I have a moisture-sensing dryer from the 80's that lets me select between multiple dryness levels, and it is extremely repeatable, as opposed to my parent's modern moisture-sensing dryer that that adds a fixed amount of drying time after the sensor trips, in hopes that the clothes will be dry enough. Sometimes they are and sometimes they aren't.
Ha. I modified my Whirlpool dryer with an Arduino-based automation because its own internal sensor is not precise. I used this nice sensor: https://www.adafruit.com/product/4099
Mine has metal contacts in the back of the drum, which I presume measures resistance, similar to how this soil moisture sensor works: https://www.adafruit.com/product/6362
I'll never understand how modern designs work worse, despite access to better sensors.
Where I am in California, there's a $30+/mo charge to connect to the grid, and the largest savings from a battery was being able to disconnect from the grid. There's lots of time I have excess power generation when I could give to the power grid, if I were connected, but I would have to pay extra to do so, so the potential goes unused.
Is delivering back to the grid economical in California? Where I'm from people disconnect solar panels on sunny days because it costs them money to return to the grid.
I'm on NEM2.0 so I can generate more than I use at peak hours and push into the grid for higher credit value than I consume overnight to charge my car.
Still, I don't see the value proposition for batteries on NEM2.
If I wasn't using _any_ electricity at my house, and I could 100% charge the batteries off-peak and push the power back to the grid at peak, I'd only be arbitraging like 5-10c/kWh * 15kWh per pack.
So, $1.50 per day, per pack. Unless I'm totally thinking about this wrong. The spread between peak and off-peak rates is just too small.
My provider wasn't PG&E, and I don't know if they do the same, but there's a fixed $30/mo service fee for connecting to the grid, regardless of how little power is used, even if its negative.
But is that rate always positive? Where I'm from during peak sun hours, the rate is negative and you end up paying money to deliver money to the grid. They do this to incentivise you to decouple your solar installation during peak sun hours so the net doesn't get flooded with too much energy.
This is for solar only installations. People are advised to flip the breaker of their solar installation when energy prices are negative as it would cost them money if they deliver electricity to the grid. This helps to reduce strain on the grid.
The reason is that California has made their grid extremely vulnerable. The grid already heavily overproduces solar so it is reasonable to have negative prices. There is no sink available.
I'd love to see an LLM equivalent, but I don't think that's enough data to train from scratch. Could a LoRA or similar be used in a way to get speech style to strictly follow a few megabytes worth of training data?
Yup that'd be very interesting. Notably missing from this project's list is the KJV (1611 was in use at the time.) The first random newspaper that I pulled up from a search for "london newspaper 1950" has sermon references on the front page so it seems like an important missing piece.
Somewhat missing the cutoff of 1875 is the revised NT of the KJV. Work on it started in 1870 but likely wasn't used widely before 1881.
reply