Live data from Hacker News

Newest MacBook Pro is the first MacBook not recommended by Consumer Reports

businessinsider.com

431–440 of 440 posts

Re: Newest MacBook Pro is the first MacBook not recommended by Consumer Reports

#431
post #410
post #372

Earlier quoted context omitted.

> Sorry, but you're wrong about iOS not supporting security fixes, and you're having to create an artificial situation to justify your argument: the situation where a user doesn't stay updated to the latest iOS, when in fact, the user has every motivation to do so. Read, comprehend, and then post. I said the current iOS version. The latest iOS versions vastly decrease the performance, especially on the older iPhones.…

>Any Google Pixel. That's hilarious. That device was just released. > Read, comprehend, and then post. I said the current iOS version. The latest iOS versions vastly decrease the performance, especially on the older iPhones. So the user has a choice: * No security updates. * The latest and the greatest with decreased performance. -- You shouldn't snark when you're not even right. You don't know what you're talking ab…

> That's hilarious. That device was just released.

Pixel/Nexus then. The reason Pixel is valid is because Google has been giving these devices (since you're not very good at reading: 'these devices' refers to 'Pixel/Nexus') good support for a long time.

> You're also wrong about "vastly decreased" performance.

I am not; it is widely documented.

> Apple is kicking Android's ass in this department, that's universally understood.

Did you read what I wrote? Did you comprehend? I haven't been comparing to all Android devices. I never did, never have, in none of my posts. Why would I? I know there are Android vendors who deliver shit support. What I am not saying is that all Android vendors do this. Motorola, for example, when still part of Google delivered good support, and all the examples I previously gave also still count.

> Another distorted lie-response to a claim I didn't make; I didn't say "with security updates". I said full compatibility with the fully-updated software.

Yes, I know you didn't claim that because you're unable to see I am the one who claimed that because your claim was, well, I'll be friendly: "inaccurate". I don't want full software support, I don't give a shit about full software support. I want security and reliability fixes, and, quite frankly it should fall under warranty. Nobody tosses their fridge, car, oven, PC after 2 or 3 years yet in handheld (smartphone/tablet) market this is somewhat normal. We agree there is a problem here, and on average compared to Android, yes Apple does better. At a premium price though.

> Apple is able to push out vital security updates WITHOUT doing a full software update

Yeah, with magic fairy tale patches. Get real. I don't have proof? All the vulnerabilities found and patched lately (good part of 2015, and all of 2016) in iOS 10 are not available in iOS 8: https://en.wikipedia.org/wiki/IOS_8

Nevermind iOS 7 and lower.

Even iOS 9 seems not so good supported anymore: https://en.wikipedia.org/wiki/IOS_9

"Support status Third-party application support only"

Great that means my mother's iPad from 2012 isn't supported anymore. Yeah, great software support...

> Yes, your bias is evident

Pot, meet kettle.

> Tragically what you're missing here is that locked-down nature also makes the devices secure.

My Android phone is not rooted, and I'm talking about locked down in the sense that I am not able to replace the hardware. In the sense that I'm more or less leasing the device since I am not allowed or easily able to get root if I desire. In the sense that an American company is putting their values on me as an European which means: violence is good, sex is bad. Fuck those values, I decide what is good and bad for me, my wife, and my children.

I'd also argue that every phone is fundamentally insecure to nation states if the attacker has physical access.

> You probably have no clue what it actually costs.

I do, I bought an iPhone SE for my mother and had the shop install a screen protector for her. While they were attaching it, I informed how much it'd cost to repair. Like I said, 160 EUR at minimum.

Re: Newest MacBook Pro is the first MacBook not recommended by Consumer Reports

#432
post #359
post #155

Earlier quoted context omitted.

I've gotten myself into a stupid situation where I have purchased something like 4 adapters that don't accomplish what I thought they would when I bought them. USB-C to DisplayPort -> DisplayPort to HDMI -> Monitor - doesn't work USB-C to DisplayPort -> DisplayPort to DVI-D -> Monitor - doesn't work USB-C to Ethernet - doesn't work I'm actually planning to just go to Best Buy today and try to find anything that will…

To be fair, I'm facing the same issue with a Dell XPS 13 9350 i bought instead of the new MacBook Pro: I want to connect it to a 4k monitor at 60Hz via DP or HDMI 2.0. You have to find out how DisplayPort is done on the TB3 port (with USB-C plug) to know which adapter will work. The safest way seems to be to read Amazon reviews. Thanks Amazon reviewers!

I have the same problem, but since I don't have Amazon here I'm testing different cheap adapters from aliexpress. No luck so far.

Re: Newest MacBook Pro is the first MacBook not recommended by Consumer Reports

#433

Earlier quoted context omitted.

> Or try out Thinkpads or XPSses with some flavour of Linux but I don't have high hopes for their touchpads and screens / HiDPI software support. Touchpads generally work perfectly out of the box, with any Linux distribution that uses libinput instead of one of the older drivers. HiDPI screens also tend to work perfectly out of the box, though you may wish to tweak them to your tastes. (For instance, I don't want to…

I'm curious why Apple's approach of drawing things at 2x or 3x and then downscaling the whole screen hasn't caught on in the Linux world. It seems like scaling the result (a single buffer) should be extremely easy? Laptops and screens targeted at Windows 10 mostly have crappy resolutions for both running at 2x or 1x.

Linux desktop environments do use the 2x/3x scale approach for high-DPI by default. If GNOME detects a high-DPI display, it'll render everything scaled larger by an integer factor.

1440p represents the one case that integer scaling doesn't quite handle right, because you don't typically want it to scale 2x (showing the same amount of information as a 720p screen), you want it to scale roughly 1.33x (showing the same amount of information as a 1080p screen). So, on a 1440p screen, I disable the automatic high-DPI scaling (which does make the display readable, it just shows less information than I'd like), and use other mechanisms to scale by 1.3x.

Re: Newest MacBook Pro is the first MacBook not recommended by Consumer Reports

#434
post #353

Earlier quoted context omitted.

The Razer Blade Pro also has a laughable battery life: 2 hrs 45 min for WEB SURFING. That's hilarious. What that means is, probably less than an hour of battery life if you are actually gaming, which is the raison d'etre of this laptop. Laughable. What would people say if Apple made a high-end professional machine with half an hour of battery life? We'd never hear the end of it. Perfectly acceptable in the PC world,…

If the ram's power use is the main problem, then that disproportionately affects low-power uses. There's no reason to think gaming would have a similarly crippled battery life. Any idea how long the MBP lasts if you peg the GPU?

I wonder if it would be possible for the OS to consolidate "hot" memory into as few modules as possible and power down the ones holding relatively cold data when memory pressure is low.

Re: Newest MacBook Pro is the first MacBook not recommended by Consumer Reports

#435

Earlier quoted context omitted.

What a clusterfuck...

The moment where the Hacker News crowd can't even figure this out anymore ... is the moment I have to stop recommending macs for anyone in my family or friends. There were a few mac sales due to my recommendations, but that has to stop now. Good luck, Apple.

Well this seems more due to the USB-C to rule them all paradigm that will actually deploy in every PC as well very soon.

People complained continually that Apple had gone USB/lightning/Thunderbolt instead of standard USB-C. Now they implement the standard and people complain that the standard is a clusterfuck...

This whole ranting start to look a bit unfair.

Re: Newest MacBook Pro is the first MacBook not recommended by Consumer Reports

#436

Earlier quoted context omitted.

Actually it isn't precisely Apple's fault. It's the USB-IFs fault. USB 3.1 in general is a mess. The spec allows for Alternate Modes, where the port is hardwired on the Motherboard to be able to be switched to deliver DisplayPort, HDMI, MHL, Thunderbolt, and damn near anything else you can figure out how to mux over the port's pins and cheaply demux on the other end of the cable. The best part is no one will actually…

Apple is lauded for making confusing technology simple. It's not Apple's fault, it's Apple's missed opportunity and it's our high expectations of Apple that are being eroded.

I'm pretty sure if you stick to apple hardware, things just work. So in a sense their technology is simple for those who stay in the ecosystem.

But don't dare buy a low-cost adapter from eBay or be damned for 10 generation.

So yeah it's a closed environment but is it so far worst than the jungle outside?

Re: Newest MacBook Pro is the first MacBook not recommended by Consumer Reports

#437

Earlier quoted context omitted.

I'm curious why Apple's approach of drawing things at 2x or 3x and then downscaling the whole screen hasn't caught on in the Linux world. It seems like scaling the result (a single buffer) should be extremely easy? Laptops and screens targeted at Windows 10 mostly have crappy resolutions for both running at 2x or 1x.

Linux desktop environments do use the 2x/3x scale approach for high-DPI by default. If GNOME detects a high-DPI display, it'll render everything scaled larger by an integer factor. 1440p represents the one case that integer scaling doesn't quite handle right, because you don't typically want it to scale 2x (showing the same amount of information as a 720p screen), you want it to scale roughly 1.33x (showing the same…

Linux desktops do the hard part out of the box: scaling everything by an integer factor. But then there is no built-in way to downscale the result to a usable size, which should be relatively easy?

For example, if you need more space on a 5K iMac than the default 1440p@2x, you can have macOS render everything at 1800p@2x off-screen, then downscale it to fill the screen.

The UI looks like this: https://support.apple.com/en-us/HT202471

I think it can be done with xrandr, but IMHO it should be a built-in option in GTK-based desktops too.

Re: Newest MacBook Pro is the first MacBook not recommended by Consumer Reports

#438

Earlier quoted context omitted.

> if two ports weren't compatible, then they would be physically incompatible they have forgotten about poka-yoke (inadvertent error prevention) which is very remiss. https://en.wikipedia.org/wiki/Poka-yoke

A problem that Apple used to understand, if you look at their keyboard USB cords: http://www.mcelhearn.com/images/articles/20060828181525223_1... I guess the USB-C plug is too small for gimmicks like this, though. Would've been nice to distinguish between USB-C and TB 3.1.

(For context, that is how Apple prevents people from plugging anything but its own keyboard into the extension cord, or daisy-chaining two extension cords.)

Re: Newest MacBook Pro is the first MacBook not recommended by Consumer Reports

#439

Earlier quoted context omitted.

Linux desktop environments do use the 2x/3x scale approach for high-DPI by default. If GNOME detects a high-DPI display, it'll render everything scaled larger by an integer factor. 1440p represents the one case that integer scaling doesn't quite handle right, because you don't typically want it to scale 2x (showing the same amount of information as a 720p screen), you want it to scale roughly 1.33x (showing the same…

Linux desktops do the hard part out of the box: scaling everything by an integer factor. But then there is no built-in way to downscale the result to a usable size, which should be relatively easy? For example, if you need more space on a 5K iMac than the default 1440p@2x, you can have macOS render everything at 1800p@2x off-screen, then downscale it to fill the screen. The UI looks like this: https://support.apple.c…

Ah, I see.

I don't think that would work well with Linux expectations for rendering, such as crisp edges for fonts; if you like Apple-style rendering of fonts and other UI elements, it may work better.

I'd rather see everything rendered for the display resolution I have, and just scaled at a non-integer factor.

Re: Newest MacBook Pro is the first MacBook not recommended by Consumer Reports

#440
post #203

Earlier quoted context omitted.

Oddly the underlying calls in the power api are all still there (so you can just write a quick app to query it) so I have absolutely no idea why in the world they have removed it, other than to completely irritate you. It's infuriating.

They removed it for the announced reason; at least at the moment, it's not accurate. It would be infuriating to display an indicator that is completely unreliable.

But it is accurate - at that point in time, the time left is valid. If you suddenly go mental with the CPU then yes it would be inaccurate.

I would argue that some indication is better than none. On the same machine, I can run Windows and get an estimate on the time left until it is flat.

Post reply on HN