Introducing a new wireless protocol is incredibly difficult. You basically have to have all the countries in the world to give you a chunk of their spectrum.
You have 2 mainstream protocols now, one for low energy, slow data transfers (Bluetooth) and one for fast, but more power hungry devices.
This is an UX problem, not a technical problem. You could easily use Wifi to transfer files between devices quite fast, there's just no agreed upon open protocol for it. Afaik that's how AirDrop works.
This is an $27k car, with $7.5k rebate, so much for the unfairly competing Chinese.
The MG4 is £22500 in the UK before tax, which is about $30k in USD, and that's a full-featured 5 door car, with double the range.
I do applaud the philosphy of cheap and barebones, and easily moddable, but my two cents is that trim is not the thing that's making cars expensive to make.
BYD's allowed to sell in Europe. They're not crushing the market here. They're not substantially cheaper, or better for what they offer for the price compared to other manufacturers.
No way I'd trust them. When you crash them or they have a battery fault, the doors lock you inside before the battery catches fire. Many videos of this happening inside China with one recent event in the West.
Only cybertrucks I've heard about catching on fire where the ones purposely set on fire. While I'm sure it happens I doubt it's any higher than any other vehicle on the road
Fail-safe designs are more expensive because they require redundancies, fully manual linkages, or just non-centralized control.
The Cybertruck went with daisy chained PoE automotive Ethernet variant. The same cables delivering power to subcomponents handle data. Damage/problems in a single component can not only bring down the network but kill power to all the car's subsystems. It means less wiring in the Cybertruck (and lower production expense) at the cost of durability and fail-safety. Someone looked at TokenRing Ethernet and said "yes that is best".
I think it's a well intentioned safety feature that was never fully thought through. Locking the doors in a crash can prevent a passenger from being ejected from a vehicle. However, if there is no reliable way to unlock the door once the acceleration forces have subsided, you've created a death trap.
There's a mechanical latch release handle integrated into the doors, but they are very much not meant to be used during normal operation and are designed to be inconspicuous. This seems to cause at least some people to fail to operate them during a fast-paced emergency situation.
China has one of the least free trade regimes in the world, their currency controls alone amount to potentially more than Euro tariffs on cars and that’s just one part of their governmental stacking of the deck for their manufacturers.
I think it’s easy to look at the outputs of their industries and compare them extremely favorably to the outputs elsewhere, especially in EV.
But once you start comparing tariff adjusted pricing it gets much trickier much faster.
BYD could slash european prices by quite a bit. They price them competitively to take advantage of the margin. The increase in price compared to their domestic MSRP is pretty wild, 2x in some cases. In a race to the bottom, they will win.
You're right, but comparing Switzerland to America... You need a car to live in 90% of the USA. That said, talking only about specs or prices is pretty reductionist. If anyone on this forum could forecast car sales based on pre-delivery marketing, you know, become a billionaire investor.
I have said this and will reiterate - building an 'afforable' EV is impossible with the current level of technology - by which I mean a vehicle that competes on price with affordable ICE vehicles, and doesn't make compromises that would make it impractical to own as the only car.
There are $20k cars with infotainment, bodypaint and probably a lot more creature comforts than this thing. Also this thing has a 150 mile range (less probably IRL), which is not practical.
Looking at the basic shape, the drag looks horrible, and probably the efficiencys bad too, considering they only manage 150k with an 52kWh battery.
Euros have already tried this, they put out abominable shitboxes where they tried to save money everywhere but the battery and charger, and the result were poverty cars which barely cost less than a Model 3.
Once you spend the money on a 400 mile battery and a fast enough charger to be practical, you're most of the way in terms of BOM to a 300HP electric upmarket road monster. Tesla understood this, and are dominating the market.
BYD also knows this, and there's a reason their C-segment EVs cost more than their D-segment plug-ins, despite the latter having tons of electric range.
Also doesn't cost $20k from the factory, it costs $20k with tax credit.
Why is this? We've been told that the cost of batteries is in freefall for years now. Why aren't they competitive yet with ICE, even if they are heavier?
At this point he's just saying what he thinks is expedient in order to avoid the government breaking up his companies.
It's why the whole Meta thing exists - they wanted to be seen as a VR company who has a side hustle in social media to avoid being classified as a monopoly. That argument has failed so now he's asserting that social media doesn't matter.
Because real humans have their own wants and preferences and may not like talking to you or may not like talking about what you like to talk about. And even if they do, they might not be available when is convenient for you and the constraints of the platform might prevent you from forming a long term relationship, which is really what loneliness is about, not having friends that know you. An LLM will always find your discussion engaging, will remember almost everything you share with it, and won't ever make you feel unwanted. I don't think it's good or healthy that people engage with these tools in this way, but there's a lot of ways in which they are obviously superior to people.
I wonder if FreeDOS has anyl relevance in our modern cloud-based VM world, where a real-mode operating system could serve as a razor thin wrapper for running applications on virtual machines with minimum overhead.
In these scenarios, having another layer of memory protection and pre-emptive multi-process multitasking might be completely unnecessary.
Well, just yesterday I had to run FreeDOS from a USB in order to run a sas2flash utility that didn't run on Linux or directly via UEFI, so I could flash some firmware to a SAS controller. Seems at least some of us still need it :)
Btw, avoid getting a SAS controller if you can, and get a motherboard that directly supports as many SATA drives as you need. Few things have been as frustrating as dealing with this SAS controller thing.
When I have to deal with a SAS controller my go to is always to see if it's possible to flash it with IT (initiator/target) firmware so that it just becomes basically pass through disks for the OS to handle however it pleases.
Off-topic, but what SAS controllers are you having trouble with? Was going to get some for my NAS, would be good to get some insight on what to get and expect!
As mentioned in the other comment, the controller shipped with IR firmware, while I wanted IT firmware on it, so all drives shows up in the OS, instead of doing hardware RAID. I don't think this specific controller is more/less problematic than others, just that all the tooling around SAS seems to suck. At first I couldn't get sas2flash to work on Linux, so ended up trying to boot an UEFI shell with sas2flash.efi instead, which refused to work seemingly because of some motherboard/UEFI version/controller incompatibility.
So ultimately I used FreeDOS to finally be able to run sas2flash, so I could flash the IT firmware. Maybe I'm spoiled, but overall it's been a somewhat confusing journey.
And today I also started looking into getting LTO8 for long-term backups, probably will be even worse, judging by the docs I've gone through so far...
No, the technology serving what you're describing is microVMs, or taken even further, unikernels.
FreeDOS has no technical relevance other than its intended purpose of continued support for legacy DOS applications on modern hardware - which is perfectly fine.
Not anymore, it's either signed EFI binaries or handled live from user space (with LVFS covering the Linux side). Sometimes vendors still ship boot images, but nowadays it's likely a small Linux live ISO.
Real mode would not be an efficiency gain at all on modern CPUs.
Even if they were not the case I think it would only have any use between containers and full VMs when you want more separation than the former but less weight than the latter, and there are various existing microVM options that seem to have a handle on that space.
I was shat on multiple times by throbbing-brain architecture guys for this, but what I've done multiple times that worked in prod, was to start with a single monolith with dependency injection, and run all services in the same process.
Once the compute and robustness demands mandated we should scale the thing across multiple machines, both horizontally and vertically, I would replace most DI services with proxies that wrapped the original functionality and called out to a remote host.
So for this scaled up version, I would only need to change how the DI container got started, which could either be in 'Manager' mode (meaning a high level functionality, with the submodules injected as proxies calling out to services on client machines), 'Worker' mode (meaning it served the proxy requests on said worker machines) or 'Standalone' mode (meaning the DI container actually injected the actual fat versions of the services, that allowed the whole thing to run in a single process, very useful for local testing and debugging).
The was only a single executable which could be run in multiple 'modes' selected via a command line switch, which made versioning and deployment trivial.
It was C# with ASP.NET containers. Not the sexiest tech stack in existence, but it's really versatile. I kind of simplified the architecture in my post, there were like almost a dozen different 'roles', and not all of them ran off the same executable (since some roles had specialist requirements, with some running off of Windows, the others off Linux, so there was some real world messiness to account for), but the Standalone 'god' mode was there and pretty much worked well on dev machines.
You have 2 mainstream protocols now, one for low energy, slow data transfers (Bluetooth) and one for fast, but more power hungry devices.
I don't see the usecase for UWB.
reply