spronkey

Couldn't agree more, modern macOS is just kind of... annoying. I also don't feel like it really does much if anything better than e.g. Snow Leopard, while simultaneously performing a lot of mostly invisible magic behind the scenes that I wish had some feedback and gave some control over - like all the photo analysis stuff.

Do you know what plan you are on for Protonmail? I can't see anything that would allow that many mailboxes for anything less than 288 EUR a year?

Let's not forget that Google's spam filtering racket has made it next to impossible to host your own email in any kind of practical manner, or even use many hosted email services effectively over the years.

Yes. Through a combination of increased network-dependence of apps, increased pace in general of updates, system-level updates that are required for security, app updates that are required for system updates and/or security, or simply because the app just ceases to function at all without the updates, and SaaS, we have in many ways lost the ability to say "no" to updates by remaining on older software.

I generally agree. MacOS appears to be stagnating in its capabilities, even if there is a constant churn of UI and features. The Human Interface Guidelines-based approach of old seems to have been blurred thanks to the iOS mashups we are now seeing, and native apps that used to feel substantially better aren't really better enough anymore.

However, for me the big issue with Electron or other web-tech apps is the performance - they all feel incredibly slow and laggy, engorge themselves on my machine's resources, and generally slow down substantially the longer they have been open.

Considering my main gripe about macOS when compared to say, Windows, has been the general performance of the UI and apps, this isn't a particularly welcome phenomenon.

Rotten Tomatoes is the epitome of why trusting expert ratings on their own is a bad idea. The number of times that the tomatometer is wildly off the general perception is insane, enough to make the tomatometer largely useless for myself at lease.

It certainly does seem, though, that a company with Google's prowess in search, NLP, ML, algorithms, categorisation and the like should be able to do a lot better, even if the problem is incredibly difficult.

My only explanation is that it doesn't suit the business needs of Google to do so.

I find DDG is generally good or at least good enough when the data you need is relative popular and relative well distributed around the internet. DDG falls very, very flat when it comes to needing to scour the depths of the internet for something very specific.

Case in point is car parts - find an obscure part from a Japanese Honda, for example, and attempt to find it in DDG. I'll wait. :P

First comment in this thread that's hit the nail on the head with regards to why software requires maintenance. After all, if I write an algorithm, how on earth does it just stop working after a while?

Fixing flaws is a separate and legitimate concern, but whether feature or security, this is solved by the idea of a warranty where the purchase price includes the product, and some time-limited obligation to ensure that it is fit for purpose. Once that's up, it's up to the developer to determine whether it's worth their while continuing to support it - but at least the end user can run it on the platform they originally purchased it on. Theoretically.

Then comes the hiding-in-plain-sight pernicious moneymaker mentioned: platform planned obsolescence. It comes in many forms, forced upgrades as mentioned, but also things like "turn on auto-updates because that's just what you do these days", "oh this is a 32-bit app so you can't run it anymore - pester the developer". Then there's all the developer-side stuff like "32-bit apps don't work anymore, update", "Oh neither does the older SDK, update that too", "Oh didn't we mention there's this new screen shape and you have to add support for that too", and "oh if you don't do all this stuff your app will disappear from the store in X days". What about the less obvious stuff like "this new SDK/toolkit/platform library is the way to build new apps! There's not really a clear upgrade path, but it's the one true way. Until the next one."

It's next to impossible for developers to solve this problem because it is an externality imposed on them by platform vendors. It is the main reason why software, in the short term, needs to be maintained. Of course, long term you expect that big developments and innovations would result in some level of obsolescence, but the devices and software industries have an incredibly fast rate of churn even for the pace of innovation. It's very interesting using systems from 5, 10, 20 years ago just to see how far we've come (or in many cases, how far we haven't).

Apple in particular has driven this by consistently eschewing backwards compatibility, constantly churning their SDKs, and continuously breaking older platforms. It's no surprise, they have two strong financial incentives to do so: if new software requires new Apple hardware, that's new revenue for them, providing they can convince us to buy it. If new hardware requires new software, that's new revenue for them thanks to their app stores. If they can get developers to move to subscription models well, that increases the recurring revenue via IAPs. Force them to use your payment and accounting services too, clipping the ticket along the way? More revenue!

We're the suckers though - we've been conditioned so hard to expect their stuff to be somewhat fragile, expensive to repair, and severely limited in lifespan that we almost like it. You dropped your phone? Oh the glass on the back has all broken. That's integral to the chassis, so it's a really expensive repair - over half the cost of a new one. Might as well treat yourself. No don't listen to that Rossmann guy, he's obviously <unsafe/counterfeit/untrustworthy>, eyes over here buddy, look at my pretty blue shirt. Oh hey, if you go to him Face ID won't work, because a nation-state actor might have installed an intercept on his dodgy counterfeit screen. You want to keep secure right? Anyway what were we saying again? Oh yeah, the new one. Oh you also ran out of storage? It's only been 5 years, the base model still has the same amount, so why don't you pay a bit more and get the upgrade. After all it only makes sense that you'd accumulate stuff over time right? What did you say, SD card? RAM stick? nVME SSD? No no no, they're way too slow/insecure/low-tech/dangerous to work with iPhone/Mac. No no your friend who put one in his computer has a PC - these are Macs, entirely different type of thing. Of course software needs to be maintained, of course it does. How could it possibly work on your new machines, how on earth! I digress...

I've been on record here for as long as I can remember saying something similar in defense of Apple's laptops, and I've done something similar to you in buying (usually cheap ex-lease) Windows machines every year or two to try them out, but I wouldn't go so far as to say Macs are superior in "every way".

Typically with an Apple device, except for the butterfly keyboards, you could be reasonably sure that all the interaction points were at least mostly decent - display, inputs, speakers microphone, charger, system noise (fan and electrical). You could also be sure that the inbuilt feature set such as WiFi, Bluetooth, Thunderbolt, USB (USB3 was a clusterf... on PCs in the early days), even charging would work reasonably reliably.

The other nice thing about Apple's laptops for a long time has been longevity of batteries - Apple were the first manufacturer to move to 500 cycle batteries with the 2008 Unibodies, and then 1000 cycle batteries later on. Additionally, they have pretty good charging algorithms to preserve battery life over the long term - not something that is the case with many other brands.

However, I think Apple have regressed on some of these fronts since 2015 (maybe they'll redeem themselves with the M1). The touchbar and T2 have been notorious for strange issues. The USB-C only setup has been both obnoxious for anyone who ever needs to use a Type A device, and notorious for strange issues such as heat generation when charging from the "wrong" side - a clear step backwards from MagSafe. The quality of their machines has suffered with the butterfly keyboard, flexgate, less robust internals (see Louis Rossmann for more details!), and anaemic cooling (one of the most impressive things about the original retina MBP was the improvement in the cooling system for sustained performance vs the thicker unibody). MacOS itself also declined in quality somewhat obviously since somewhere around Snow Leopard, with mere flashes of brilliance inbetween since then.

At the same time, Windows 10 has improved dramatically and is now solidly Microsoft's best OS since Windows 7, beating even that OS out in many ways. And, while I don't think the PC laptops have caught up, many of them are now within firing range of Apple devices - certainly close enough that in many cases, for the few that actually get all the touchpoints right, they offer an experience that isn't too far away from the best Mac experience I've had.

I'm seeing more and more this sentiment about "not being the target group", and using it to justify crappy business decisions.

Around the PPC change, Apple took the grand strategy of backing away from this idea of controlling what their users were able to do with their machines. Now they're moving back to their "we say what our machines are for and if your use case doesn't match then screw you" attitude.

Many of Apple's business decisions of late have been nothing but hostile to the idea of allowing users to do what they want with their devices. As a very long time Mac OS user, this whole idea of a "walled garden" was never central to the Mac, and while Apple have always been a sufferer of NIH syndrome, it seemed in the mid-2000s that they had finally got over this and started playing nice with the rest of the industry. So much for that.

I don't like your attitude here, because ultimately its harder and more expensive to mitigate against these types of failure than it has been in the past.

First, on Apple devices there are more software-induced failure modes than there had been in the past, and many of these failure modes are far more difficult or obscure to recover from. An OS update bricking your system in a way that Time Machine (or a disk image or similar) couldn't recover from was incredibly rare until rather recently.

Second, Apple devices have become so integrated and locked down it's now impossible to use some of the historical strategies to get yourself back up and running faster in the event of an unexpected failure. For example, you can no longer take the drive out of your current device and stick it in a replacement device to continue working. You've gotta make sure you have a perfectly working backup system to do that now, and even that takes waaay more time then the old-school drive swap. We're still not at the point where it's easy to shift your entire life from one machine to the next unless everything you do is cloud-based (which isn't the case for most).

Third, Apple's devices in particular have become harder to repair. They have more proprietary features than they did in the past, and a less friendly attitude towards repair than in the past. Apple don't release detailed documentation or repair manuals on how to e.g. revive the T2 chip, and AFAICT some of the tools needed to diagnose and repair Apple devices are internal only. It's now also impossible for all but an SMD repair shop to replace certain components if they fail because they are now soldered to the logic board. This, combined with the turnaround time for repairs on Apple products being ridiculous, has put even more pressure on people to have more robust backup plans than before, and penalises those without these plans moreso than in the past.

Forth, Apple's devices have become more expensive in the first place - or at least, there are now fewer options for folks to get the machines they need from third parties via upgrades. I typically work in smaller businesses and with freelancers and contractors, and many of these folk used to buy lower end models and upgrade the components over time as their needs improved and components got cheaper.

With soldered storage these days it can also be more difficult to find a suitable machine to buy e.g. used while you wait for your main machine to be repaired. Need a 1TB device? In the past, you could pick up any used machine on the market of approximately the specs you need, and simply swap in aftermarket storage. Now? Well, you're limited to the tiny portion of Macs originally sold with 1TB drives. There may be some way to use an external drive for this, but damned if I can find instructions on how to do so on Apple's website. What I can see however is a disclaimer that Macs with a T2 chip might be more difficult to use external startup disks with. I'm sure I could figure it out eventually, but I'm pretty sure your average freelancer would struggle.

So while ultimately yes, you as a professional can do things to mitigate the circumstances of any single point of failure and should do so where it makes sense, from where I sit this is an increasingly difficult and time consuming process. And all of that extra difficulty has been forced upon its customers by Apple.

Yeah, Anandtech's dive revealed some interesting benchmark results - such as the Zen 2 4200U part being quite close in both performance and power consumption (a little higher) to the M1 in heavily multithreaded tasks, but getting stomped in both aspects in single-threaded ones.

The M1 isn't a tiny power-sipping mobile part - it sucks power down just like AMD's 15W TDP CPUs do. The efficiency gains Apple are getting here are likely to be the result of several factors, only one of which are the CPU cores themselves.

Well, it's more like people celebrating an electric scooter that has twice the range of the electric bike.

Not saying it isn't a novel form of transport or equally useful in most cases, but... we're comparing a very stripped-down SoC to systems that have vastly more complexity for several different reasons - not least the ability to support more modular CPU/memory/GPU configurations.

Rework the existing x86 cores that we have into similar configurations and we'll likely see pretty substantial efficiency gains there too.

The issue with this is that Apple has the same problem that they have had for years - they don't really offer mainstream products.

Even their most inexpensive products are on the higher end of the pricing spectrum, and they could be 5x the performance but it still wouldn't matter.

There's a reason why Chromebooks are so popular and it sure isn't anything to do with their performance.

Multicore scaling on M1 doesn't appear to be as efficient as that on Ryzen - for example, the 4200U is able to beat the M1 in multicore tasks at quite a similar power draw, but gets soundly beaten in single core. Ryzen's big advantage over previous multicore implementations was infinity fabric - so clearly there's more to scaling than just the actual compute cores.

Don't forget that scaling up is also not just about frequencies, there are also packaging considerations - the CPU dies have to actually be able to dissipate the heat generated, and the package itself has to be able to do so as well. I'd expect that this is something AMD and especially Intel have a leg up over Apple with - although, considering they've already tread that ground it makes Apple's job a bit easier too.

Sequential performance is only one facet of overall SSD performance. Some older Apple SSDs weren't the greatest at random reads and writes - it's possible these new ones are quite a bit faster in that department, FWIW.

It wasn't actually that bad... Everyone likes to remember it as this massively power hungry beast with terrible performance. It was power hungry compared to what came later, but it did also provide some pretty good performance at the time - especially if you were the kind of tweaky nutter who liked to overclock.

Northwood was a relative bargain when you cranked the bus clocks up well beyond what it said on the box.

One of the questions I have is just how much of their relatively recently-gained process optimisation knowledge of e.g. 14+++ will be transferrable to future node shrinks, and, will this give them an IP edge over TSMC if they can get back on track with node shrinking?

They are riveted to the aluminium top case, that also contains the trackpad and a battery glued in on top of it. So... you're forced to buy all that for a keyboard - a part that is likely to fail in the lifespan of the machine.

Absolutely this - even some aggressively optimised stuff like game engines don't have insurmountable amounts of micro optimisations in them.

Also - why would Apple massively up and change uArchs? Even if they did decide to turn Apple Silicon into not-ARM, I'd wager it would look a lot more like ARM than for example, x86 does.

Yeah, the DTK hasn't exactly shown that Apple Silicon can outdo Intel, let alone AMD.

I remain rather skeptical. I think this is a business win for Apple moreso than a technical win.

I'm skeptical. Geekbench seems like just about the perfect use case for Rosetta 2 translation - I wouldn't be surprised if performance is close to native. If that's the case, the A12Z is only approximately competitive with Intel of a similar vintage despite having a process node advantage.

Have you looked at the network traffic coming out of macOS lately?

Bullshit. Stop spreading false rumours. There's a big difference between counterfeit goods and refurbished goods.

You lack imagination here :) There are several solutions that could have been pursued, including introducing a completely new type and effectively duplicating the existing string library for it.

It's also worth considering that it extends the usable life of the machines. In some companies, older high end machines that are still in good condition are recycled and passed down to less demanding users - e.g. developer machines may get passed down to tech support.

For personal users, it means that passing it down to a family member may no longer be an option.

In addition, the lack of upgradability has tanked the resale value of lower end Macs - so lease companies aren't recovering as much value, and consumers machines are depreciating faster than Macs of old.

It's not just buyers remorse. If you use the machine for work and buy something under-spec'd - especially in RAM, you lose money by being less efficient with a slower machine.

Maybe in Silicon Valley with VC money.

Even then, it's getting more and more difficult to justify the increasing gap between Apple pricing and say - Dell or Lenovo. Almost $1k in it now for equivalently spec'd non-base MBP vs XPS15/X1 Extreme, and the gap just gets higher as you need higher requirements.

Let's have a look at what Apple have done since 2012.

1) Inflate the base prices of the machines and attempt to justify it via non-optional "features" such as the touch bar, wide gamut displays, extra thunderbolt ports, obsessively thin designs, T2 chip.

2) Solder everything, requiring customers to buy the specs they think they'll need in ~3 years' time upfront, when prices are at their highest. Look how expensive 1TB of flash was 3 years ago vs today, for example. Heck, in 2008-2012, several Apple machines could be upgraded to beyond their original BTO capabilities thanks to technology advancements and firmware updates by Apple at the time.

3) Where they didn't solder storage in the 2012-2015 machines, they used several different proprietary form factors for blade card SSDs when standardised form factors have existed the whole time (mSATA, M.2 SATA/PCIe AHCI/NVMe).

4) Removed the ability for customers to restore machines to working state either in the field or in a timely manner, and pushing customers toward Apple service and AppleCare.

5) Literally glue in the one consumable item in the machine (battery) that is almost certainly going to fail before the usable lifespan of the machine, pushing the price of a battery service up dramatically, reducing the economical lifespan of the machines.

6) Reduce serviceability of other components likely to fail or get damaged over time such as the keyboard and trackpad by riveting, glueing, sandwiching etc to ensure older machines are uneconomical to repair as soon as they can be, pushing customers toward buying a new machine.

This is a company that is doing everything to take away your choice as a customer, trying to turn expensive computers into disposable appliances. Don't try to justify this crap - just say no.

All the above, combined with the design flaws almost every 2016+ MacBook has (butterfly keyboard, flexgate, staingate, display connector issues, T2 chip integration issues), the seriously declining quality of Apple's OSs, the removal of useful features (MagSafe!, sleep light, external battery status meter, IR remote, non-type C ports, SD reader), have me now in the position where I not only don't want to buy any of there new MacBooks, I'm actively encouraging others not to as well.

Me, a once huge Apple fan whose personal portable machines have been Apple almost exclusively since the 90s. Whose OS of choice has been OS X/macOS since Jaguar. Who used to go out of his way to explain why Apple machines were worth it.

Nope. No more.

T (and X) series solders some stuff now too. Dell XPS15 and X1 Extreme are mostly maintainable.

Completely agree. I've been saying this since 2012 when the Retina machines came out, and skeptical since 2009 when they moved to non-user-replaceable batteries in the MacBook Pros.

I don't think I've actually ever had a non-Apple main notebook, and that goes back to the 1990s!

That said, I only tolerate the 2013-2015 rMBPs, use one for my main personal and work laptops, but the soldered RAM pisses me off (a lot, because my personal machine only has 8GB and my god is it hard to find a 16GB model for a reasonable price in the used market), and the proprietary storage irks me. Thankfully 10.13 supports NVMe with an adaptor, which to me basically confirms that there was zero reason for Apple to use the proprietary stupid thing in the first place.

As for any machine they've built after 2016, well, I don't want them. I don't want a butterfly keyboard with no travel that breaks with a skin flake. I don't want screens that stop working because they use a flex cable connector that's too short. I don't want a touchbar if it means no function row. I don't want to give up MagSafe. I don't want to give up my SD card slot. I don't want to give up USB Type A. I don't want a massive trackpad, and I don't want the fscking T2 chip.

In fact, the only things on the >2016 machines I do want are the faster CPUs and GPUs, the better quality displays, and Touch ID!

... As for Lenovo though, they're slowly into Apple 2.0. Have you seen the T/X x90 and X1 series? Soldered RAM. At least they still make the X1 Extreme and P series.