In most cases, efficiency and performance are pretty synonymous for CPUs. The faster you can get work done (and turn off the silicon, which is admittedly a higher design priority for mobile CPUs) the more efficient you are.
The level of talent Apple has cannot be understated, they have some true CPU design wizards. This level of efficiency cannot be achieved without making every aspect of the CPU as fast as possible; their implementation of the ARM ISA is incredible. Lots of companies make ARM chips, but none of them are Apple level performance.
As a gross simplification, where the energy/performance tradeoff actually happens is after the design is basically baked. You crank up the voltage and clock speed to get more perf at the cost of efficiency.
> In most cases, efficiency and performance are pretty synonymous for CPUs. The faster you can get work done (and turn off the silicon, which is admittedly a higher design priority for mobile CPUs) the more efficient you are.
Somewhat yes, hurry up and wait can be more efficient than running slow the whole time. But at the top end of Intel/AMD performance, you pay a lot of watts to get a little performance. Apple doesn't offer that on their processors, and when they were using Intel processors, they didn't provide thermal support to run in that mode for very long either.
The M series bakes in a lower clockspeed cap than contemperary intel/amd chips; you can't run in the clock regime where you spend a lot of watts and get a little bit more performance.
Control theory was also one of my favorite classes that a low of software people should learn (at least the very basics). So many hand rolled heuristically driven if/else type systems that can simply be replaced more reliably with a PID.
Absolutely the worst control systems of all time have been written by software engineers that don’t understand control theory. The second worst control systems are designed by those who only know the PID heuristic, and can’t be bothered to model a little non-linearity from motor drives saturating.
The biggest problem with PID control is that the integral term performs double duty as both a signal that accumulates small errors to minimize the steady state error, but also as a signal that shows deviation from the target due to unmodeled constraints.
It should be pretty obvious that you cannot overcome constraints by moving even harder in the direction of the constraint, which is what the integral term does.
I've played around with this over the years in my career but have found that tuning PID loops is very tricky, much trickier than creating a soup of if/else clauses and much less auditable to those who don't understand the math.
PID is standard in the industry, but the reality is it is infinitely easier to model in the discrete domain. The z-plane if math, but you don’t really need much math. Just model like a games developer. Simulate with a bit of JS or python. Add the motor saturation! Play with feedback and disturbances.
I just think this gives much better results. The model can be as simple or complex as you need, and we aren’t trapped in the linear response range. PID is good enough for many tasks, but it’s never good.
If you can model your problem with linear differential equations then control theory replaces the need for tuning. The coefficients you need just pop directly out of the analysis.
Maybe I should add more context. I have specifically tried applying PID style feedback systems to computational problems, not controllers that interface with hardware, circuits, etc. My undergrad was in math and electrical engineering, I "pivoted" to software as a grad student (though I was always very involved in the software side of my department; I was a coder from when I was a kid.) The place I found it to work the best is with designing a homegrown autoscaler years before k8s ever became a viable thing for a company to play with [1]. Most of the problem domains I applied it to do not have linear models that can effectively model the theory. Yes I know that a PID is only proven to be stable when working with linear systems, but this is the reality of the problems I've worked with.
Eventually when if statements stop working I found that decision trees work great and XGBoost continues to be a great iteration of a decision tree.
[1]: I was an early hire at a tech unicorn and we built an autoscaler pretty early into the company's tenure. While it was a great success for a long time once k8s became established in the industry we had a really hard time training new talent to it and I left as we began a massive company-wide effort to move our workloads onto k8s.
I have a friend that went through 3 remarkables due to failures, I didnt dig into the root cause but I suspect you might be right on the USB-C port since they all "stopped charging".
They also have the USB C charging quirk/cheap-out that if they're completely drained they will only charge with a low powered trickle charger until the device gets to some minimal level of charge, and then you can use a higher power source.
the video in there, via microscope, you can see how the contacts peeled off. Larger photos show the PCB and how the connector is "affixed". Final photo is after it was fixed, but before the epoxy
Its unfortunate that the supply chain for eink/epaper displays all seem to be centering on typical mobile device aspect ratios (like 16:9 for this device) particularly because remarkables are marketed as productivity oriented replacements for notebooks.
I would much rather have a A6 or A5 sized display or any other standard size for paper notebooks.
I have a Supernote Nomad and love the size of it, which is A6 sized. I struggle to see how making one more narrow than an A6 pad is useful. This Remarkable kind of looks more like a long post it note or grocery list instead of a notebook.
> I struggle to see how making one more narrow than an A6 pad is useful.
The trick to something narrower is to get comfortable with doodling in landscape mode, e. g. for classroom notes, and scroll (and orientation-switch) accordingly when neccessary. Ideally you'd have physical complementary buttons present, but a good touchscreen with palm rejection works as well. To-do lists and the like can be done vertical mode. In other words, a digital notepad.
Now you only have to built a corresponding smartphone-sized, pen-focused, modular and connectable open-standards general-purpose computer. :) ... :(
The reMarkable looks too underpowered and maybe too enshittified (subscriptions, lock-ins) to be used for anything else but a digital notepad.
Landscape mode would make sense, if scrolling on eInk wasn't completely awful. I can sometimes deal with it on really light websites like Wikipedia, but I just prefer not to.
I will say that my supernote really is just a digital notepad. I keep all my work to do lists organized on it. But since it's Android and supports side loading apps, I have the Kobo app and read a ton on it even without a backlight.
> Landscape mode would make sense, if scrolling on eInk wasn't completely awful.
I don't see how that can't be improved.
This smells more of a fundamental problem to me where vendors enshittify their devices to try wooing complete non-users, e. g. people who are too incompetent to work their head around using a penabled smartphone-sized machine and its notetaking/sketching applications exactly like they would use an analog pendant (the classic pocket notepad/pencil combo), or people who bore others with tedious litanies about how "the screen is too small". The latter crowd is well-cared for options-wise, so this is virtually a non-problem. I want to carry around and use a digital pocket notepad on steroids, and not a bloody whiteboard.
> I will say that my supernote really is just a digital notepad.
I only tested the Nomad's pen functionality (as well as ergonomics resulting from its size) and was very pleased. It's weaknesses lie elsewhere.
Is it the case that these devices are converging on 16:9? I don't know about the supply chain, but there seems to be no lack of e-ink tablets at A5/A6 sizes and/or with better ratios than 16:9.
Remarkable has the roughly A6-sized Paper Pro, Kobo has three e-ink devices with styli and good screen ratios, and Supernote has models named A5 (and A5 x2) and A6 after the paper sizes. I think the options are quite good.
eInk devices are very much not converging to 16:9 or wider aspect ratios. This device is intentionally the size and shape of a reporter's notebook, but there are virtually no other eInk tablets which diverge significantly from more common paper aspect ratios – they all (ReMarkable, Supernote, Boox, Kindle, etc.) are and continue to be exactly what you say you want.
I had the same thoughts, there are clearly indicators that the weakness in the labor market started happening before LLMs and AI took over popular discourse.
All the more reason to believe that while correlated, LLMs are certainly not the largest contributor, or even the cause of the job market weakness for young people. The more likely and simple explanation is that there are cracks forming in the economy not just in the US but globally; youth employment is struggling virtually everywhere. Can only speculate on the reasons, but delayed effects from questionable monetary and fiscal policy choices, increasing wealth gaps, tariffs, geopolitics, etc. have certainly not helped.
The Boomers are largely gone now from tech. One former manager of mine was 65 when I left, she retired a year later at 66, gone. Unfortunately, most of the roles those people held were not back-filled. The roles vanished, too.
Anecdata: I spent quite a bit of time driving around office parks in Eagan, MN. Most of them are dead--really, really dead. Vacant offices everywhere, the hotel that used to cater to business travelers is shuttered and the parking lot looks like a jungle. I can't peg exactly when all this took place because I haven't worked in that area in several years, but probably the effects of 2020, the remote work culture, and now the layoff hangover. I see a lot of people in their 50s and 60s working retail jobs right now. They often look like folks who would have been working in an office someplace.
No, because they hire someone to massage their feet and do the books. /s
I'm sorry, I just find the whole concept silly. Makes it sound a silly business decision to do some SRE inhouse instead of “cloud everything” (prob. with the “help” of “consultants”) even at the $1bn scale.
Also unfortunate that SK seems to have gone overnight from a generational family unit to the western style nuclear family. All of those elderly living in poverty despite modern SK built on their backs.. I am sure they would love to contribute to raising the next generation of children.
There are so many cultural factors in SK that might take generations to reverse without being "forced" by the government (which is also wary to enforce). Of course, by then its too late per the video.
The nuclear family worked when you could easily have a single household provider, it doesn’t work anymore and the recent trend of having multi generational households seems to be completely driven by people not being able to afford to move out.
I always pondered if child baring should be done as a generational leap.
As in how would society look like if people have kids in their 20’s with the grandparents who are in their 40’s being the primary caretakers and rinse and repeat.
Seems that this combines the best outcomes in terms of biology and still being able pursue educational and career goals.
But this is a very major shift from where we are today. It’s going to be far more likely that more and more people will start having children in their late 30’s and even 40’s and 50’s. If we are going that way then freezing sprem and eggs at a young age should be much cheaper than it is now and people should really start considering it.
Presume that a majority of women of reproductive age per generation al cohort do not want children, and intend to exit those fertility years childfree. What then?
I see no crisis, only total fertility rates reaching a neutral rate based on women empowered to make the best choice for themselves.
Society will have to correct itself somehow, either through social change or technological advancement that or we will all go extinct…
I actually wonder why this isn’t a bigger talking point, we are probably not at the point of no return yet but many countries are getting there and people will be caught by surprise as whilst the effect is delayed human life expectancy isn’t that long and it doesn’t take more than a couple of generations like ours until we are going to be facing a major crisis.
I really don’t know where we went wrong, and I’m not sure it’s purely financial either (tho it is for many), at least from my anecdotal experience.
> I really don’t know where we went wrong, and I’m not sure it’s purely financial either (tho it is for many), at least from my anecdotal experience.
The problem is welly explained in the video: the society in East Asia (South Korea, Japan, China, Taiwan, etc.) is too toxic for the youth and it is too hard for them to live a decent life.
Young people there are constantly competing with others of their generation. Households are unaffordable for the majority of the population, and for the lucky ones, it takes approximately 30 to 50 years to buy one. Traditional culture encourages people to work extremely hard, to the extent that they don't even have time for social activities or to form relationships. Meanwhile, promotion is often difficult because management-level positions are occupied by older individuals.
People are stopping having children not only because they cannot afford it, both financially and in terms of time, but also because they do not want their children to suffer. It is expected that the situation will persist for generations to come.
The population ballooned because women were not educated and empowered. Now that they are, and have robust access to family planning, TFR is coming down rapidly and population will eventually follow.
Where we went wrong? Women not being empowered in the first place. This is the fix, not a problem. This is a success story.
I don’t know what’s right or wrong but can we agree that TFR below replacement isn’t sustainable in the long term?
Even if you don’t see shrinking population as a massive problem which it will be, if the TFR remains below ~2.1 humanity won’t be here for much longer.
I disagree. The world has ~8.2B people, and has blown past 6 out of 9 planetary boundaries while headed to 9-10B people by 2100 (due to population momentum). Humanity will successfully continue on with an order of magnitude reduction in that number 150-200 years from now, based on a median global TFR of ~0.5-1. TFR isn’t going to 0. We can plan accordingly, if we choose to. We are currently on the unsustainable path; a lower TFR puts us closer to sustainability.
You can disagree all you want but if the TFR of the world becomes as low as the one of Chile we will get to below 1 billion people world wide within less than a century and go extinct within a millennia and the latter is based on that life expectancy won’t change and if there will be that big of a reduction in population life expectancy will plummet.
I’m also not sure how much empowerment anyone will have once we are forced back to living as agrarian subsistence farmers within a few generations.
So I don’t know if you are trolling at this point or not…
Not trolling at all. Actually bootstrapping a non profit to buy unwanted fertility from people who don’t want it, to sell into carbon markets to spin up a flywheel to help everyone who doesn’t want kids to be empowered to not have them. So perhaps we just see the future and individual agency and empowerment differently.
The level of talent Apple has cannot be understated, they have some true CPU design wizards. This level of efficiency cannot be achieved without making every aspect of the CPU as fast as possible; their implementation of the ARM ISA is incredible. Lots of companies make ARM chips, but none of them are Apple level performance.
As a gross simplification, where the energy/performance tradeoff actually happens is after the design is basically baked. You crank up the voltage and clock speed to get more perf at the cost of efficiency.