How does one distinguish between all the tiny devils in the details between all possible chargers to have the maximum efficiency in ones iPhone charging?
There really aren’t any nuances to be aware of here. If the charger can provide 45W or more, it will charge your iPhone 17 Pro as fast as it can go. If you have an older iPhone, you don’t even need that much.
The Apple charger is mildly interesting because they made a slightly smaller than usual one that displays a dynamic behavior where it boosts up to 60W. That’s it.
(If you actually mean efficiency of energy delivered, and not time efficiency… the energy efficiency of charging your own personal phone is completely irrelevant to anything. You could save many times that amount of energy per year by switching to a heat pump water heater, or a heat pump clothes dryer, but most people haven’t.)
There was some speculation that chargers would need the AVS spec to full speed charge the new iPhones, rather than just any USB-PD. Has that been tested yet?
I could definitely see maximum sustained charging speeds only being sustained with optimal supply voltages, which minimizes converter loss (and thereby heat, which is the actual limiting factor for fast charging these days) in the phone.
I didn’t say the charger doesn’t support AVS. AVS is fine, but the practical impact on charging time is going to be difficult to notice in this specific case, if it exists at all. Apple does not even hint that AVS is needed to meet their charging time claim in their official documentation: https://support.apple.com/en-us/102574
Apple simply says “For iPhone 17 models, you can charge up to 50% battery in around 20 minutes with a 40W adapter or higher.”
Sure, it would be fun to see someone do a thorough test on this. No, I don’t agree with all the posts that are fear mongering people into believing they need some super fancy charger to reach the charging speeds Apple claimed, when Apple themselves doesn’t claim that, and the original 9to5Mac article no longer claims this true. There is simply no reason to believe that AVS is required, and anyone saying that has the burden of proof here.
It could of course also be included only for power efficiency without being strictly necessary for optimal charging performance, but cooling performance can vary between cases and with environmental conditions, so I suspect it might be at least partially a functional requirement.
More fearmongering. Provide proof, or please stop confusing people. This thread was asking for recommendations to understand how to charge their phone. The answer is extremely simple. This is not the thread to dig into irrelevant details. Read the room.
If you want to get technical, the difference in conversion losses between AVS at 13.3V@3A and a non-AVS 15V PDO @ 2.667A is going to be rounding error. Both need to be regulated down to about 3.5V to 4.2V by the phone's internal circuitry. A hypothetical buck converter that is just as efficient at 15V -> 3.5V as it is at 13.3V -> 3.5V would see no difference at all in heat generation. In the real world, such a small difference in input voltages at >1A would likely yield a less than 1% efficiency difference. 1% of 40W is 0.4W, but I repeat: I am saying less than 1%, not 1%. Based on tests I've seen in the past, previous iPhones could dissipate at least 4W of power continuously, forever, in normal ambient conditions. For a 95% efficient buck converter (which is probably conservative here), the total thermal load from charging at 40W is 2W. Unless the thermal load exceeds 4W, there should be no difference in charging speed potential. 2W + 0.4W would be 2.4W, which is well below the 4W threshold, and the buck converter on a high end smartphone is probably more than 95% efficient, while the difference in input voltages probably yields less than 1% difference in efficiency, so 2.4W is a very conservative number here. Keep in mind that phones can dissipate significantly more than 4W for brief bursts, and the charge curve on the battery isn't going to allow 40W charging for very long regardless of the thermals, so the phone likely has even more thermal headroom here in the real world. Even a 20V PDO @ 2A -> 3.5V should be perfectly fine here. I see no evidence that there is even a possibility of AVS making a difference here unless you wrap the phone in real insulation (not a phone case) or stick it in a toaster oven while charging.
Talking about how there might be an imperceptible difference under very specific extreme conditions is exactly what I would call fearmongering in a thread where someone was specifically asking for recommendations. It won't matter to anyone, ever, in any real world situation. This is not the thread to be contrarian for the sake of digging into minutiae. You're welcome to start a thread where that is the goal.
In summary: OP does not need to buy a fancy AVS-enabled charger. They shouldn't avoid such chargers, but there is no reason to get one. Unless you have material proof to the contrary that you want to present, but it seems like you don't.
I don't have access to, nor a need for, the 40W charger, yet I'll continue speculating about it (which is why I prefixed my comments with "I could see", "it could", and "I suspect"), thank you very much.
Mainly, I'm trying to understand why AVS is supported by both the phone and the adapter if supposedly it makes effectively no difference.
It could be as simple as being a requirement for PD 3.2 wall adapters, of course, but that raises the question for why it's required as well, if it's supposedly that irrelevant.
There is nothing to speculate about here, other than to create confusion for others. The physics is immensely clear. AVS is not needed here, and OP can save money by buying whatever is the cheapest option that provides >40W.
Since I have shown through basic math that it isn’t related to the ability to charge the phone any faster, that possibility is obviously eliminated.
AVS is likely supported because there’s no reason not to support it. It is part of the newer USB-C PD specs, which Apple wants to support, and it continues to iteratively refine the energy efficiency of the charging process, which matters a small amount at a scale of billions of devices, but doesn’t matter at all for a single individual’s purchasing decision.
There is a group policy to "turn off windows consumer experience". To edit local group policies win 10 pro is needed. It is also possible to change the register value. See the link for more info.
Are they really though? A serverless event driven architecture system I’m working on literally costs less than £10 a month on Azure. Running full blown VMs instead of cheaper more appropriate technologies like containers or functions will always cost more.
As per our calculation for CloudAlarm [0], as we reach a few hundred users, it'd be cheaper to use a dedicated instance than serverless (Azure Functions) design. So it may vary from system to system depending the amount of work you perform for each user.
0: https://cloudalarm.in/ – btw, you may wish to have daily budgeted pace based alerts using it – to inform you when the usage spikes up (much faster than Azure's consumption threshold based alerts).
I was playing with the Azure "free" tier. Even I tried to be extremely careful with it, after a while noticed that I had left a storage blob for a VM hanging around and some external IPv4 address. I will continue to use Hetzner online for my own stuff instead running this on "public cloud".
The nice thing about using Cloudflare in front of a real host is that you can still do dynamic pages. You can instantly purge any file from the Cloudflare cache, so all you have to do is purge anything that changes and your users see the update instantly, while your backend only sees a couple of extra requests.
My site updates some data every 10 minutes or so, I don't think that would work with GitHub Pages. Maybe you could do something with Cloudflare Pages combined with Workers, but Workers have a limited free tier. The normal Cloudflare CDN scales to infinity for free.
I used to backup with Arq to S3 and the restores were very fast. With Glacier, Arq initiates the data retrieval and waits for file availability in the UI. Glacier can take a day or so, laptop needs to be open and connected to internet during the time for the transfer to complete. Given I'm primarily backing up external drives, I find this UX not an issue, but if you wanted backups with instant accessibility, backing up to S3 really isn't that pricy.
