Not sure about jellyfin, but I assume it uses ffmpeg? The M1 is fast enough that ffmpeg can re-encode raw video footage from a high end camera (talking file sizes in the 10s of gigabyte range) an order of magnitude faster than realtime.

That would be about 20W. Apparently it uses 5W while idle — which is low compared to an Intel CPU but actually surprisingly high.

Power consumption on my M1 laptop averages at about 2.5 watts with active use based on the battery size and how long it lasts on a charge and that includes the screen. Apple hasn’t optimised the Mac Mini for energy efficiency (though it is naturally pretty efficient).

TLDR if you really want the most energy efficient Mac, get a secondhand M1 MacBook Air. Or even better, consider an iPhone with Linux in a virtual machine - https://getutm.app/ - though I’m not sure how optimsied ffmpeg will be in that environment… the processor is certainly capable of encoding video quickly, it’s a camera so it has to be able to encode video well.

This. My Mac has 16GB but I use half of it with a Linux virtual machine, since I use my Mac to write Linux (server) software.

I don’t need to do that - I could totally run that software directly on my Mac, but I like having a dev environment where I can just delete it all and start over without affecting my main OS. I could totally work effectively with 8GB. Also I don’t need to give the Linux VM less memory, all my production servers have way less than that. But I don’t need to - because 8GB for the host is more than enough.

Obviously it depends what software you’re running, but editing text, compiling code, and browsing the web… it doesn’t use that much. And the AI code completion system I use needs terabytes of RAM. Hard to believe Apple’s one that runs locally will be anywhere near as good.

You don’t need metaphors. It’s pretty simple.

The Spotify app should have a button that takes you to their website, where you can sign up for a premium subscription.

It doesn’t have one because Apple would kick Spotify out of the App Store.

Also - all other links to the Spotify website (support, terms of service, privacy policy, etc) take you to pages where the main navigation of the website has been removed so that you can’t find the signup page. Because again, Apple bans that. For the longest time apps have not allowed to have any way for users to find a signup form on a website.

That policy is now illegal in the EU (and a growing list of other countries) and Apple’s attempt at compliance is a new API - only available in Europe - that informs the user that they might be a victim of theft, fraud, etc before they get taken to a website that is deliberately sandboxed… supposedly to prevent theft/fraud/etc but more likely because it makes it really difficult for Spotify to link that signup with an existing free account.

Oh and if Spotify opts to expose users to see that horror show… they’d have to pay tens of millions of dollars per year to Apple. They have so far refused to do so, meaning the new regulations have failed (well, they were failing, until the EU declared Apple’s compliance efforts insufficient).

near ChatGPT4 performance locally, on an iPhone

Last I checked, iPhones don’t have terabytes of RAM. Nothing that runs on a small battery powered device is ever going to be in the ballpark of ChatGPT. At least not in the foreseeable future.

First of all, you’re implying it runs latest Windows - but Windows 11 shipped a few years ago.

Second - not really a fair comparison. 18 years ago the iPhone didn’t even exist. And the oldest model (17 years old) had really weak hardware. 4GB of storage, 128MB of RAM, and the CPU was an order of magnitude slower than current spec CPUs (it was also 32 bit - and 64 bit ARM is a completely new architecture - similar to the failed Itanium).

Even if it was supported, it would be a horrible experience.

Apple was advertising that these devices would be refurbished when they were sending them out to be destroyed

When did Apple claim that? Sure if you send them a one year old phone, they will refurbish it (they will also pay you several hundred dollars to take the device off your hands). But they’ve never refurbished several year old models. Those have always been recycled (destroyed) regardless of what condition they are in.

does anyone know if there’s any actual data that shows personal disability information being recorded/collected?

That’s basically the crux of the case right? The law is pretty clear, Google can’t collect that data (or at least, if they do, then they’d have to comply with a long list of privacy regulations that I’m pretty sure they don’t comply with).

Um, did they actually do something impressive or is it just a really big gas tank?

I’m struggling to imagine why anyone would even want 1,300 miles of range in a PHEV. Surely it’d be better to have a smaller tank and more space inside the vehicle.

CCP has been claiming it for a while

“A while” as in about 400 years — that’s when China took over Taiwan.

After World War II, there was a power struggle between the Republic of China (backed by the USA) and the Chinese Communist Party (backed by Russia).

The ROC/US controlled pretty much all of China, but then the US withdrew support and simultaneously granted concessions to Japan (as part of the peace deal between Japan and the US) and the CCP/Russia took advantage. The resulting civil war “ended” with the ROC having control of Taiwan, and the CCP controlling all of the rest of China.

But that civil war never really ended - it merely cooled down and became non-military conflict.

AFAIK the optics have to be regularly cleaned, calibrated and replaced. And by regularly, I mean daily/weekly for some of that.

The process is a carefully guarded trade secret and intentionally difficult. The companies that own the machines are not allowed to have employees who are trained in the process. When you buy those machines it comes with a service contract from the manufacturer. And the manufacturer is ASML - a Dutch company.

Sure, but the vast majority of Mac software at the time, including loads of first applications from Apple, couldn’t run on Tiger. You had to run it in the “Classic” environment - and they never ported that to Intel.

Tiger shipped just 4 years after the MacOS 9.2 and plenty of people hadn’t switched to MacOS X yet.

The reality is Apple only brings things forward when they can do it easily.

Apple has done eight major CPU transitions in the last 40 years (mix of architecture and bit length changes) and a single team worked on every single transition. Also, Apple co-founded the ARM processor before they did the first transition. It’s safe to assume the team that did all those transitions was also well aware of and involved in ARM for as long as the architecture has existed.

Apple has the target disk mode, but doesn’t the laptop need to be shut down for it to work?

Modern Macs can’t do Target Disk Mode. If you had the right cables (thunderbolt or firewire) it was really fast, just as quick as a high end internal PCIe SSD.

And yes, you did need to reboot - because the other computer had full arbitrary read/write access to the raw sectors on the drive with no safety checks or security. If you did that while the computer was running normally, you’d corrupt the data on the disk as soon as they both tried to do a write operation at the same time — and also TDM needed to be used with caution - the other computer could easily install a rootkit or steal all your saved passwords.

It’s been replaced with “Mac Sharing Mode” which operates while the Mac is running normally, does have all the necessary algorithms in place to avoid corrupting the disk, full security to authenticate each read/write operation and block attempts to mess with system files, and therefore is orders of magnitude slower than TDM.

I know iPads (and I assume Android tablets) can be a second screen over wireless using third party software but it’s not uncompressed video with disk access last I checked.

The video is compressed (how much depends on your network speed, it’s not always noticable). And it’s far more than just video - you can copy files over the connection. Keyboard/mouse/touchscreen/stylus inputs are sent over it, and video camera/microphone data can be streamed in real time as well. There’s also a control protocol to temporarily switch from sending the entire screen to sending just a URL (and auth cookies) to a HLS video stream such as a YouTube video - which will cause the other computer to directly access the content over the internet instead of one computer downloading it, decompressing it, then recompressing it and sending it to the other computer.

And it’s not just iPads. Macs, iPhones, Apple TV… they all have that capability. It’s the core underlying system behind AirPlay, AirDrop, Continuity Display, Universal Control, Clipboard Sharing, Continuity Camera, etc etc.

I do it all day every day between my desktop and laptop Mac — I effectively use this as a KVM so I can control my laptop using the nice mechanical keyboard and mouse attached to my desktop (also, it’s a handy way to avoid having to keep data in sync over the cloud… I tend to do all my note taking on the laptop and just never access them from the desktop - eliminating any risk that one of them might not be fully synced up with the latest data).

It works best over thunderbolt but it’s usually done with wifi — always a direct wifi connection that bypasses your router because the amount of bandwidth required is so high that if you sent it to a router and then to a computer… your wifi would almost certainly collapse under the load.

Target Disk Mode doesn’t exist on modern Macs. It has been replaced with a new “Mac Sharing Mode” which is technically completely different. The new system is basically just a regular network fileshare (I think it uses SMB), while I think the the old system was PCIe connection if you had thunderbolt/firewire (fast) or something much worse if you were using USB (that never worked well).

Solar, wind, hydro can do it, but the amount of CO2 produced by manufacturing the generators is still massive

That’s FUD.

Sure - the concrete in a large hydro dam requires a staggering amount of electricity to produce (because the chemical reaction to produce cement needs insane amounts of heat), but there’s no reason any CO2 needs to be emitted. You can absolutely use zero emission power to high temperatures needed to produce cement.

And not all hydro needs a massive concrete wall. There’s a hydro station near my city that doesn’t have a dam at all - it’s just a series of pipes that run from the top of a mountain to the bottom of a mountain. There’s a permanent medium sized river that never stops flowing that comes down off the mountain - with an elevation change of several hundred metres. It provides more power than the entire city’s consumption and does so while only diverting a tiny percentage of the river’s water. As the city grows, the power plant can easily be upgraded to divert more of the water though pipes instead of flowing uselessly down towards the sea.

Covid and Russia’s war created massive fluctuations recently but if you look through that noise global CO2 emissions are pretty much flat and have been for a few years now. They are almost certainly going to trend downwards going forward (a lot of countries already are seeing downward movement).

The simple reality is fossil fuels are now too expensive to be competitive. Why would anyone power an AI (or mine crypto) with coal power that costs $4,074/kW when you could use Solar at $1,300/kW (during the day. At night it’s more like $1,700 to $2,000 with the best storage options, such as batteries or pumped storage). Or wind at around $1,700.

Nuclear is $8,000/kW unless you live in Russia, where safety is largely ignored.

Hydro can be cheap if you happen to be near an ideal river - but for most locations it’s not competitive with Solar/Wind. So hydro is safe as a long term power generation method into the future, but it’s never going to be the dominant form of power unless (like my city) you happen to have ideal geology.

Unless you pay for expensive tags (like $20 per tag) or use really short range scanners (e.g. a hotel key), RFID tags don’t work reliably enough.

Antitheft RFID tags for example won’t catch every single thief who walks out the door with a product. But if a thief comes back again and again stealing something… eventually one of them will work.

But even unreliable tags are a bit expensive, which is why they are only used on high margin and frequently stolen products (like clothing).

All the self serve stores in my country just use barcodes. They are dirt cheap and work reliably at longer range than a cheap RFID tag. Those stores use AI to flag potential thieves but never for purchases (for example recently I wasn’t allowed to pay for my groceries until a staff member checked my backpack, which the AI had flagged as suspicious).

I’m betting the owners of the NYT would LOVE to have an AI that would simply re-phrase “news” (ahem) “borrowed” from other sources

No way. NYT depends on their ability to produce high quality exclusive content that you can’t access anywhere else.

In your hypothetical future, NYT’s content would be mediocre and no better than a million other news services. There’s no profit in that future.

If the EU finds apple guilty of “systematic” non-compliance, which will happen if they “continue their shenanigans for every judgement”, then the DMA doesn’t call for a fine. It calls for a TikTok style forced sale. Apple could be ordered to sell the iPhone to another company or face an outright ban in the EU.

Of course that’s assuming the EU has the balls to actually enforce their own laws knowing full well the transatlantic political consequences.

we can be sure a megacorp like apple is pulling every lever at their disposal

Sure. But as far as I can tell the only lever at their disposal is to drag their feet and try to delay compliance as long as possible.

It’s an effective strategy, but it’s stupid. They’re making a lot of people angry and that is never a good long term strategy.

Where will they get their info from with no one to scrape?

It’s not like there’s a shortage of human generated content. And the content that has already been generated isn’t going anywhere. It will be available effectively forever.

just “standing on the shoulders of giants”.

So? If you ask an LLM a question, you often get a very useful response. That’s ultimately all that matters.

I disagree. The real news is the free model will now search the internet for up to date answers, and for calculations it will write and execute a python script, then show you the result.

Paid users of ChatGPT have had those features for months, and they were a massive step forward in terms of how often the AI provides accurate answers.