For a lot of these changes, you need a motivated and engaged enforcer/system-owner who will work through these challenges. It takes time and effort to change the culture of a group. I can't imagine someone wanting to change unless they either gain a strong motivation (highly unlikely given that they have been in the current condition for a while themselves) or they are given the right framework/setup by someone else (govt. in this case). So, this is a hard problem and one shouldn't get disheartened due to no progress so far.
Maybe you were technically right for some small scope of the problem, no proper industry definition of GC was supported by C++ by any of the major compiler. So, the HPC scientists were right imo.
Being technically right for an aspect that doesn't matter is a common trope that a lot of software engineers fall for.
Finding one with a decent screen for anything other than gaming is the trick. I linked it up there, but this is IPS with 100% sRGB, more brightness than most people need, 16:10 (vertical space is nice!), 165Hz, and matte! Most gaming laptops are...not that. Maybe 50% sRGB on a dim glossy TN screen.
Then you haven't been looking close enough. Cheapo gaming laptops yes, but Lenovo Legion, Asus G14 plus most other more premium models from other manufacturers have very good screens.
Mostly Chinese companies have been very busy designing & manufacturing actual silicon. And prices for such products have come down a lot, too.
It wouldn't surprise me if at some point there's RISC-V based phones, tablets & other devices on Chinese market, that most people outside China simply don't know about.
Or that teardown of a cheap phone sold in say, India, turns out to have a RISC-V SoC inside without much public attention beforehand.
Probably sooner than later. RISC-V is bound to become a lowest common denominator for computing devices especially at the low end. Picked by default unless <insert specific requirement here>.
Android being ported is a logical consequence of this.
It's gonna be cheap risc-v Chinese phones for a while. And they'll get faster and faster every six months. But they'll stay cheap. Until even Samsung can't compete and stops making phones, like LG did.
A couple of weeks ago SiFive announced a new core, the P870, which is in ARM Cortex-X3 class. That's the biggest core in the latest Snapdragon Gen 8 chips.
It typically takes 2-3 years from core announcement to being in an SoC for a high end phone.
They are coming.
Arm will have something newer by then, of course, but a 2023 flagship Galaxy S23 is not going to look stupid in 2026.
Every hardware company makes some plans that far ahead because the pipeline is so long. Details of course change a lot up to about 18-24 months before release but general direction must be known way earlier.
Ironically, the EVs are currently priced higher due to Fed+State tax incentives. I bet the EV prices would be much lower if there were no such tax incentives (this is evidenced by the fact that EV prices are usually updated the day after the govt. changes these values).
Do you have any sources because that’s some backward thinking that needs some strong evidence? It would mean all the car manufacturers would have to collude to raise their prices by the amount of the incentives.
It’s funny how you assume they need to collude instead of each acting in its own independent selfish interest.
Why does one need to present strong evidence that prices will rise by $7500 when the demand curve suddenly shifts right by $7500 due to government intervention? Surely this is just Econ 101.
1. There’s an assumption here that all manufacturers will simultaneously raise prices by $7,500 in response to a demand curve shift. If a single manufacturer steps out of line and lowers prices, they will carry more volume, and therefore more revenue.
2. Manufacturing costs do not increase when subsidies are introduced. EVs, as a result, have higher profit margins. Therefore, this will result in enticing more firms to enter a more lucrative market, resulting in more EV sales, which really, isn’t that the goal of subsidies in the first place?
I don't know if you realize it, but you are making the assumption every EV manufacturer has the capacity to produce vastly more EVs than they currently do. What makes you believe in that assumption?
If you believe that firms don’t respond to incentives, then sure. If you believe that people do not innovate to optimize their utility, then sure. If you think America can’t build things, then sure; it’s an invalid assumption.
I don't know if you understand this, but factories have a maximum capacity they can produce. You can't just snap your fingers and double the capacity of a factory overnight if you realize that it is too small.
Look at how long it took Tesla to build the Shanghai gigafactory: About two years--and literally everyone who knew anything about auto manufacturing said it would be literally impossible to make that timeline when Tesla announced it in the beginning. People made jokes about it. Oh look at that! Elon Musk is lying again! Hurhurhur.
I mean, why do you think the Model S had zero competition for over 7 years? The first 2 years I understand--everyone thought EVs were a dead market. But why did Mercedes, BMW, Audi, et. al. wait the other 5 years (while Tesla was absolutely destroying their market share in the US) to bring out a competitive product?
Was it because they're a bunch of dumb-dumbs?
Or was it maybe because it takes many years to design a complex product like an EV and build a factory to build the car?
> If a single manufacturer steps out of line and lowers prices, they will carry more volume, and therefore more revenue.
How will they do that? Every one of them is production-constrained right now.
> Manufacturing costs do not increase when subsidies are introduced. EVs, as a result, have higher profit margins.
Now you're getting it!
> Therefore, this will result in enticing more firms to enter a more lucrative market, resulting in more EV sales, which really, isn’t that the goal of subsidies in the first place?
Sure, but it takes 5-10 years to design a new EV and spin up a plant. Will the subsidies still exist by then? (Probably not, because we are in the exponential phase of the EV adoption curve, so the subsidy will quickly become unsustainable).
This seems like a good test to determine the amount of competition in an economic sector. With suficient competition, the price will race to the slimmest sustainable profit margin. I mean you could try to raise your price up by $7500, but if your competitors don't, who will be getting sales?
In any case, EV prices have been dropping over recent years, and I haven't seen a $7500 bump in prices, just in more people being willing to go for a nicer vehicle instead.
That only works if the total capacity to produce products for that market is greater than the total demand for products in that market.
If people want 100 units of something, and 5 competing manufacturers can only produce 90 units total (and can only expand to, say, 110 units in a "short" timeframe), offering subsidies doesn't reduce the price.
If the government creates a demand shock, it doesn't matter how much competition exists in the market. They don't need to collude for prices to go up.
Look at what happened to toilet paper during Covid. Were Walmart, Amazon, McMaster and Grainger all colluding on toilet paper prices? Or was there a demand shock?
> In any case, EV prices have been dropping over recent years, and I haven't seen a $7500 bump in prices, just in more people being willing to go for a nicer vehicle instead.
All car prices have been dropping because we're getting past the Covid supply shocks. And if you go back farther, EV price drops are mostly the result of introducing new less-luxurious models (the current base Model 3 still costs more than the first RWD Model 3, and it has been decontented--missing features like parking sensors).
They don't have to collude to react to a change in the market simultaneously.
It has precedent too, solar panel incentives, home building incentives etc. When government grants show up, vendors can take it into account if they want. If enough of them do, the market price has changed and so everyone can put their prices up. They are in a market that is not racing their prices to the bottom to drive demand so it works out.
Most manufacturers (besides Tesla) lose money on every EV they sell, such as ford losing 32k per EV (other OEMs don't break out reporting for EVs), they would not lower cost by the tax incentive if it went away.
I think Tesla may start losing money soon. Sales are down, new quality issues are found every day and the company is involved in 100+ lawsuits.
I don't know about Ford, but most other major companies are only "losing" money because they have started a new vehicle platform and need a year or two to get the development costs back. Compared to previous CE platforms they are actually doing much much better.
I am still using my Pixel 3 (now without upgrades) without major issues. So, happy to see the new phones have longer promised update cycles. Hopefully Google doesn't clawback this promise in the future.
I've owned a few Pixels over the years, but after my Pixel 3 bricked itself (and my kid's did the same a few months later), Google did zilch to remedy it, and I have stuck with Samsung Galaxy phones since
Same here (well, a 3a)! This phone does everything I need, and then some. Granted, I'm relatively light user who mainly uses the browser, a chat app, and not much else aside from snapping a pic here or there.
I'm honestly confused about the lack of updates (I really only care about security updates). I run Xubuntu on a 13 year old computer, and I get updates. Is this just a cash grab from Google, or is there more to it?
Until recently, Qualcomm provided BSPs (binary support packages) including the kernel for Google phones. For whatever reason -- possibly that their one and only corporate purpose is to sell as many chips as possible -- Qualcomm only briefly updated their packages for chips they no longer sold.
Google updates as much of the Android ecosystem as it can. First-party Play Store apps, system webview... if you look at the normally hidden system apps on your phone you'll see that the Android team has "unbundled" many parts of the formerly monolithic system to allow updates to as much of it as possible even if the kernel is marooned at an older version.
Unfortunately, some bugs are in the kernel or drivers, so there's nothing any Android OEM (including Google) can do if their chipset vendor won't do the (admittedly non-revenue-generating) engineering to update that firmware. And eventually the system itself requires newer kernel features, so there's a limit to how far back Google or other OEMs can reasonably backport a newer version of Android.
This is part of why Google's recent phones are based on Google-designed, non-Qualcomm chipsets. It was a truly Herculean effort to scrub the Pixel line of Qualcomm, and especially of Qualcomm's incentives to abandon still-good phone hardware in order to sell more chipsets.
Your PC's OS distribution is nearly totally open-source, and the economic incentives for the Linux ecosystem are completely different from Qualcomm's. That contributes to any given general-purpose computer's longevity if it runs Linux.
Yeah I use this for very limited functionality like maps, whatsapp, uber eats. I am also mostly interacting with Google playstore for apps/apks etc. This does still leave the phone open for any day0s etc. but we will cross the bridge when we get there. I am not important enough for anyone to target me specifically and I also keep a low profile.
I was thinking about upgrading this year but I am now thinking of waiting another year since there are no immediate problems.
Which applications you use is almost irrelevant -- over the past several years, there has been a steady stream of "zero-click" exploits that allow an attacker to compromise phone with no user interaction. The remote code execution vulnerabilities discovered last year and this year in the modems used in Samsung devices require nothing more than knowing the victim's phone number.[1] And you don't need to be particularly important to be caught in a wide net, cast by criminals looking to build up a bot-net or harvest data from as many devices as possible.
All of this makes me think I should just switch to a flip phone. It's exhausting having to constantly drop hundreds of dollars every 3 years just to stay safe.
I have a Pixel 5 that does everything I want. Google will stop supporting it within the next year. It doesn't make sense to me that this device already needs to be recycled. Yes, I know about custom ROMs, but even those end support for perfectly OK phones (GrapheneOS for example no longer supports Pixel 3a).
I completely agree. My phone lost official LineageOS support last year but it still works fine and I cannot justify throwing it away to replace with a new expensive device full of features I don't give a damn about.
Probably I'm just stupid but I'm going to keep using it until it breaks.
My understanding might be limited but I don't see this being a big enough risk to warrant spending couple hundred dollars every few years for a new phone when the old one still works.
> My understanding might be limited but I don't see this being a big enough risk to warrant spending couple hundred dollars every few years for a new phone when the old one still works.
There are enough zero-day RCE exploits on both Android and iOS devices at this point that, if you're running phones that are that far out of date from security updates, you should basically just assume your device is fully compromised.
As stated above, many of the RCE exploits don't even involve any user interaction, so it's not like you can argue "well, I don't visit sketchy websites so I'm fine".
So basically, stop using smartphones, because it's fucking ridiculous to drop hundreds of dollars every 4 years on a device that is virtually the same thing as your old one. This is a huge joke, and Google and Apple need to do better or stop milking us. I was safer using a Nokia 3310 in 2007.
> So basically, stop using smartphones, because it's fucking ridiculous to drop hundreds of dollars every 4 years on a device that is virtually the same thing as your old one. This is a huge joke, and Google and Apple need to do better or stop milking us.
I mean, you're literally posting this complaint on a thread about a phone that is now legally bound to receive seven years of updates.
While there have been flagship phones that have received support for seven years, no other phone has been released with a legally binding up-front commitment to provide support for seven years.
It feels very misplaced to complain about obsolescence on a thread in that context.
As someone in a similar situation to other poster, I'm still annoyed at the choice between buying a new phone (with the financial and environmental consequences) or having to deal with the maybe I'm invisibly hacked maybe I'm not when all I want is security updates and/or a software that is built with enough safety to avoid zero-click exploits.
But yeah, this is a good news thread, thank you Google.
Could you link one at-the-time zero-day RCE that is really without user interaction and will hit any user with an old phone regardless of the user applications (like the browser) used?
It's not O(N^2) is it? It can be a continuous line of ice being pushed up. Depending on the weight bearing ability of the lift and digging capacity, you would figure out a fixed distance after which you would place the buckets to carry up the ice.
Its an interesting interview question at the very least. (More complications arise as and how you get deeper into the ice).
You can't have a continuous line of ice coming up, unless you're digging for slush. Each intact X-meter core must be hauled up on its own, and then the drill has to go back down. The deeper you are, the longer it takes to haul up one core and send the drill back down. So, retrieving the cores is clearly O(N^2).
Drilling the core itself is O(N), but as you go deeper the core retrieval dominates. Not to mention everything getting more complex the deeper you go.
