Makes sense. I guess the UI experience, including the terminal was so alien I assumed routerOS was an actual low level OS but, in retrospect, I was an idiot and of course it is Linux under the hood!
Their software It's incredibly powerful, but also quite opaque to someone not into hardcore networking. I setup a small private WISP using their gear and configuring it was pretty rough for a non networking nerd. It's not running Linux, but a custom OS. The quickset is handy, but as soon as you want to do something slightly different, you're up to your neck in low level config. Still great HW and if you know how it can do everything you need. Ubiquity gear has a friedlier interface.
I agree, using LLMs the right way and SO the wrong way is definitely possible, but it's easier to use LLMs the wrong way and just blindly without understanding. A diligent person would use it as a learning tool for sure.
One concern Ive heard is that Llama can be a great accelerator for senior engineers who know what they want. However for junior engineers it could hold back their learning as they just use whatever it gives without really understanding what or why something was done that way. It seems plausible to me. At least using stack overflow people get explanations of why you do x to get y, it's a bit copy paste, sure, but the pace brought learning. Now if you type a comment and get code, a junior engineer may not even read what's generated as 'theu don't need to, it usually works' .... Any ideas how we can ensure junior engineers do effectively learn and understand what is going on, while still getting the LLM benefit?
I usually make it Monday 18 or 19 and then I loose the will as the time taken is excessive, and obsessing about it is not good. I've finished a 5 years, but all after the fact. Some I am not sure I'll ever finish....
The cost of standards is very high, probably second to the cost of no standards!
Joking aside, I've seen this first thang when using things like ethernet/tcp to transfer huge amounts of data in hardware. The final encapsulation of the data is simple, but there are so many layers on top, and it adds huge overhead. Then stanrds have modes, and even if you use a subset the hardware must usually support all to be compliant, adding much most cost in hardware. A clean room design could save a lot of hardware power and area, but the loss of compatibility and interop would cost later in software.. hard problem to solve for sure.
The transition to RISC-V depends entirely on how much of the cpu core susbstem is from ARM. The ISA itself is one part, there are branch predictors, l1,l2 and 3 caches, MMUs, virtualization, vector extensions, the pipeline architecture, etc etc. So moving away from ARM means they need performant replacements for all that.
I'm sure there are folks like SiFive that have much of this, but how is it competitively I don't know, and how the next snapdragon would compete if even one of those areas is lacking... Interesting times.
Go garmin. I have a vivoactive 5, it has 11 days of battery and loads of stats. I love it. My wife liked mine so much she dumped her Fitbit and now has the same watch. It's great.
I agree largely. I have a p30 pro, and it's a great phone despite being 4 or 5 years old . It's so good, recently it got water damaged and I as faced with replacing it. I couldnt find a phone for under 500 that could beat it, and especially the 2 day battery life. So I didn't, I instead paintakingly dried it out over 3 weeks and it's pretty much fully functional. That's how much I didn't want to change phone, and how the new sparkly features don't really matter as much as battery life, and utility.
That's obscene. Less than 10k of materials (and they can probably source it cheaper than a regular joe like me), 5k labour. 35k profit is a bit much....
Exactly, it makes no financial sense, you'd never get payback on it at that price. The company offering it had payment plans and all, with 0% interest, to make it more attractive, but that's just part of the grift.
The price to get solar installed in the US is insanely high compared to the cost of the materials. I've seen my extended family get quotes north of 40k for modest installs. I got a modest system with batter for around 9k in the EU, and id say I'd install the same for under 5k at today's prices.
Agree. It took a truly superior product at lower cost to make a dent in Intel's dominance in server, all the while Intel tried their best to flex their lock in muscle.
That happened well after 2008, with the advent of Zen and chiplet bases tech and better perf/W
Really nice writeup. For the test, could Fermat's method be used to find candidates and then trial division to weed out the false positives? Given they are rare this should not cost much extra time as the trial division (with some precomputed divisiors) is rarely used.
I charge at home and almost certainly would not have an EV if I could not. I know someone who had an EV and could not charge at home who got rid of it as it was really inconvenient to have to regularly wait 30 to 40 mins to charge as a charger was not on his commute.
Long story short, I think EVs with current battery tech is practically limited to those who can charge where they live (be it apartment parking garage or at their house)
I just got 2 new monitors in work, one has a built in usb-c hub, and that monitor can use display port out to drive a 2nd monitor.... It can, the MacBook hardware can, but MacOS cannot. I was thrilled to no longer need multiple dongles to drive my 2 displays in a clean way, but ended up having to get an external thunderbolt hub, which can drive 2 monitors, but only if one is driven from thunderbolt and the other from display port. It's pretty shocking that this is al standard stuff, but MacOS limits what you can do.
I'm well aware it is Intel, but it was still sold at the time as a 10 hour battery life, and I'm sure it can be, when nothing is running but one tab of safari.
So my point was, and in agreement with you, is when using similar sw (buggy or not) a modern zen4 will have similar battery life to a M1 / M2
Modern zen4 based laptops can get 15 hours. My 2019 MBP was quoted as 10 hours but gets under 2.... Why? because it's more software than hardware that is the issue. My MBP runs out of battery quickly because outlook (or chrome) uses too much CPU and it drains quickly and runs hot. You get the 10 hours only if then SW lightly loads the CPU. This is the same for Zen and appple Mx CPUs. If you run exactly the same apps you will get similar performance.... The modern zen are very efficient CPUs.
Usually for me the culprit for poor battery life is MS office apps, but under MacOS I've seen window manager consume 90% CPU regularly.
Long story short, buggy SW dictates battery life on modern laptops, with very little between apple and zen (can't speak to modern Intel CPUs, i dont have a laptop with one)
Essentially averaging. To put it simply, noise is random while the signal is not. So if you average 2x the noise reduces but sqrt(2) but the signal remains. Keep doing that and you have the signal appear out of the noise.
GPS operates at a negative SNR. That also uses code division multiple access to allow multiple transmitters to operate on the same frequency and the signals do not interfere.
I've often thought about this myself. I'm sure scientists involved are aware of the compounding errors with each step and build that in, but I'd love to see an analysis that breaks that down. When I first saw it I thought the errors due to cephids must be a large component of uncertainty, but really I've no idea how well contained that is.