When Apple moved from PPC to Intel it was a good thing because not only did Intel finally have some decent chips (remember how much the Pentium 4 and earlier sucked? There's a reason Apple didn't go with Intel in the '90s), but Macs gained the ability to run Windoze software without emulation. Getting one of the new Macs was a very good thing if you didn't need Mac OS 9 compatibility.
Today, there's NOTHING to be gained from this chip switch. The new Macs are going to be slower and less compatible. Virtualization is gone, we'll be back to slow emulation (eventually, no idea how long it will take for SoftPC to come back). No, I do not believe Apple's speed claims, they're almost certainly lying.
People aren't waiting to buy the Apple silicon Macs. People are buying Intel Macs NOW, to avoid the Apple silicon garbage for as long as possible.
I''ll be happy with either, as long as he large screen iMac desktops remain available. It's very difficult to argue against "acreage" when it's needed. Laptops just can't replace acreage.
The switch away from Intel removes the fallback option of Windows for people who were on the fence. The commitment of £1000+ for a device that might not meet all your needs is significant. It's not as bad as it was back in the early Intel era, when Windows was almost required for most jobs, but the need is still there. People here excuse Apple from this consideration by claiming Apple (and somehow they also) knows that it's a "tiny" percentage of customers that run Windows on their Macs, whether this is true or not is debatable, but less debatable is whether Apple cares if they shaft a big proportion of their userbase: they don't. Cook has calculated there is more profit in a transition, so a transition it is, because profit is all he cares about. People hated the butterfly keyboard and the trashcan Mac pro, and Apple pig-headedly continued for years until they finally dropped it.
Our secondary school (UK) had one Mac before Intel. When Apple switched to Intel, most of the school switched to Macs - around 300 Macs in total, because they ran Windows in a VM for all the Windows only admin stuff. The Macs were also pretty cheap, special education iMacs for about £500 each in today's money. Good luck getting an education Mac now for £500. A lot of that Windows only admin stuff still exists today, and as such they will be switching back to PCs. There was a big spike in Mac ownership post-Intel, presumably it can partly be credited to the reassurance that people could still run Windows.
Today, there's NOTHING to be gained from this chip switch. The new Macs are going to be slower and less compatible. Virtualization is gone, we'll be back to slow emulation (eventually, no idea how long it will take for SoftPC to come back). No, I do not believe Apple's speed claims, they're almost certainly lying.
I guess if you're only interested in running Windows on your Mac (as you most certainly seem to be), then yes, you are correct, there's nothing to gain by Apple moving away from Intel. However, Apple isn't much interested in the need for Windows compatibility anymore. They're cutting the cord and moving on. You can hang onto your Intel based Mac as long as you like. Sooner or later there will be emulation software - I wouldn't be surprised if Apple is working with Parallels (and Microsoft), to come up with something feasible soon.
But to think that Apple is lying, is just plain weird. We already have the proof! The DTK has been around for several months and it's been said to be an extremely capable macOS system. That's an SoC from 2 years ago, designed and optimized for high mobility. Plus... the A14!!! Not to mention the supposed "leaked" benchmarks of the A14X, which beats the top end Intel Core i9 in the MacBook Pro 16". And that's an SoC destined for the next iPad Pro.
As I recall the intel transition, we went from a loud locomotive sounding iMac G5 which was well laid out internally and an absolute doddle to open up, service and replace parts (it even had a set of diagnostic lights to help work out what was wrong), to the first intel iMac which was definitely quieter but was an absolute nightmare to open up and try to replace anything, even things like the PRAM battery. I don’t think it was noticeably faster though.
i also remember the transition to PowerPC. That memory is tied up in my transition to the PowerBook G3, the most amazing and incredible laptop on the market at that time. Nobody had anything like it. Before that I had been using an Apple Portable. While it too was ground breaking when launched, It was embarrassing and bad for posture lugging that thing around.
So what does those experiences back then lead to my prognostications of the ASi transition? I suspect in the ASi transition the iMac will become basically a giant iPad on a stick. Completely unable to be upgraded and can only be serviced by Apple, and probably forever in port poverty.
You can see some initial signs of this in Mac OS11. And performance may not be as great as people hope, as Apple only needs to beat intel.
But I also worry that the iOSification of the Mac will lead to reduced versatility and limitations in file management, services like printing and how well it plays in office networks. I can only hope some flows the other way to help iPad be more versatile.
1). There's no reason for Apple to build a giant "tablet", you're not carrying it everywhere and it doesn't have space constrains. With that it means there will always have room for other ports, I don't think they'll do anything to change the current configuration.
Yes they soldered the SSD. They're likely solder their ASi on but so does the old G5. The only thing that remains to be a module it's the Power Mac which they can still do.
The Intel iMac you're talking about only missed out the modem port compared to the old G5. I doubt 99% of the people need that.
2). Will ASi surpass AMD's flagship? I think yes but maybe there's no point. Rumor suggests the ASi on the 24" can achieve the same performance as the 10700K with almost 1/3 of the consumption. They don't have to join the competition with that level of ratio. AMD's 16-core also facing many other issues that no one like to mention: It costs $999 on a consumer platform, take a ton of power and bottlenecked by their memory bus. Most people only bought their 6-core processors because the sweet price-to-performance ratio.
I remember waiting eagerly for the PowerBook G5 announcement. For years. And years. And years.
There were a few leaked photos of the prototype taken in the corner of a dark industrial warehouse lift as I recall.
Looking back there might have been one or two limitations, but at least it had no port poverty.
... I can't tell if that's a joke or you're being serious. If you're selling portable computers with no portability then you're not selling portable computers.
Having owned both the iMac G5 and the intel iMac, I liked the G5 better. The G5 was actually faster in some things but its downside was it was very noisy. While they looked similar the G5 was very easy to service and upgrade whereas the intel one was definitely not. A nightmare in fact. The G5 insides were neatly designed and laid out, the intel was a nightmare wire nest with a confusing logic board layout.
also the pulsing white ambient light in the iMac G5 when it was asleep was nice. Kind of like the MBA and MBP glowing logos of old.
By the next iMac probably being a giant iPad I mean it will be the same design philosophy as the iPad Pro: an squared off aluminium slab with a couple of USB cable ports and Face ID at the top, with a single legged stand. It would be thirty inches and weigh heaps. I did not dream I was implying it was something to carry around. I Also meant there would be no access or expansion possible without using an external case. I would prefer some ports in the stand as otherwise I could see the ports around the edge like an iPad so the cables stick out very unflatteringly. Ok for temporary connections but would suck if you had an external drive to boost capacity,
And yes. Was jesting about the PowerBook G5. Sheesh.
Let's hope if the iMac is thin, it will at least have ports on the side instead of - or in addition to - the back. It's absurd that you have to poke around blind to plug things in, form over function yet again.
One can not extrapolate what happened during the transition of PowerPC to Intel, the similar might happen from Intel to Apple Si. I can say that when MAC with Apple Si comes out, they will be better in all respect from features to price/performance. Apple seems serious about growing MAC product line and wanting more market share. Besides making revenue on selling lots of MACs, additional revenue from service always tag along.
What's media access control address, aka MAC, got to do with it? Or were you referring to MAC Cosmetics?
It's Mac, never MAC. If you're a newbie, you're forgiven. If you've been around the Mac market for years, then you really ought to know better.
FYI the best predictor of future behaviour, is past behaviour.
Today, there's NOTHING to be gained from this chip switch. The new Macs are going to be slower and less compatible. Virtualization is gone, we'll be back to slow emulation (eventually, no idea how long it will take for SoftPC to come back). No, I do not believe Apple's speed claims, they're almost certainly lying.
I guess if you're only interested in running Windows on your Mac (as you most certainly seem to be), then yes, you are correct, there's nothing to gain by Apple moving away from Intel. However, Apple isn't much interested in the need for Windows compatibility anymore. They're cutting the cord and moving on. You can hang onto your Intel based Mac as long as you like. Sooner or later there will be emulation software - I wouldn't be surprised if Apple is working with Parallels (and Microsoft), to come up with something feasible soon.
But to think that Apple is lying, is just plain weird. We already have the proof! The DTK has been around for several months and it's been said to be an extremely capable macOS system. That's an SoC from 2 years ago, designed and optimized for high mobility. Plus... the A14!!! Not to mention the supposed "leaked" benchmarks of the A14X, which beats the top end Intel Core i9 in the MacBook Pro 16". And that's an SoC destined for the next iPad Pro.
I don't think they're lying about the benchmarks, but real world performance is far from the same as synthetic benchmarks. And it depends on a lot more than the CPU itself. If the interconnects are slow, benchmarks can look great but general performance can be poor. Samsung for example were at first praised for beating Apple's silicon, then widely panned when it was found they were optimising for tests and the performance was crappy. Memory interconnects can be a big bottleneck, and Apple usually uses package-on-package for the RAM, something they can do up to about 8GB before it gets prohibitively expensive. Then they need an entirely different CPU package to support external RAM. No doubt this will be an excuse for soldered RAM on the 27" iMac too. Another important peripheral is the GPU, Apple/Imagination's GPUs are good mobile GPUs but are orders of magnitude slower than the latest AMD/Nvidia offerings. And slower than the Intel Integrated GPUs too.
The switch from Intel to Apple Silicon is monumental. Is there any chance that Apple could take this monumental moment to rebrand the Mac lineup? What word could replace “Mac”? Applebook? iApple? Mack? Apple PC? Apple Pro? MacPad? Apple SiliconBook? Silicon iMac? Silicon Mini? SuperMac? Emacs? Mac Express? Ace? Wiz? Hero? Gold? Force?
Hey, at least I tried.
How about something along the lines of WATCH, tv, tv+ …
Cook has calculated there is more profit in a transition, so a transition it is, because profit is all he cares about. People hated the butterfly keyboard and the trashcan Mac pro, and Apple pig-headedly continued for years until they finally dropped it.
It's really easy to play the "greed" card. But Apple is a company that has historically ignored "profit above all else". This is evident in a lot of what they do. Second, there is not more profit in a transition... there is an enormous up front cost in the development and support and engineering resources. It is a very expensive undertaking.
People will always hate. That doesn't make those things invalid, it makes it undesirable by those people. There's huge list of Apple products that people hated. Hell, there's a ton of people who just hate Apple. That's not going to stop Apple from creating and believing in the products they make. If users are having issues, Apple moves to correct them. Sometimes it takes longer, sometimes it doesn't.
Apple obviously believed the butterfly key design was superior - something that probably took a lot of resources and money to design and develop - so instead of just replacing it, they tried to make it less and less prone to becoming faulty. And by the way, a vast majority of people didn't mind that keyboard and never thought twice about it.
Cook has calculated there is more profit in a transition, so a transition it is, because profit is all he cares about. People hated the butterfly keyboard and the trashcan Mac pro, and Apple pig-headedly continued for years until they finally dropped it.
It's really easy to play the "greed" card. But Apple is a company that has historically ignored "profit above all else". This is evident in a lot of what they do. Second, there is not more profit in a transition... there is an enormous up front cost in the development and support and engineering resources. It is a very expensive undertaking.
People will always hate. That doesn't make those things invalid, it makes it undesirable by those people. There's huge list of Apple products that people hated. Hell, there's a ton of people who just hate Apple. That's not going to stop Apple from creating and believing in the products they make. If users are having issues, Apple moves to correct them. Sometimes it takes longer, sometimes it doesn't.
Apple obviously believed the butterfly key design was superior - something that probably took a lot of resources and money to design and develop - so instead of just replacing it, they tried to make it less and less prone to becoming faulty. And by the way, a vast majority of people didn't mind that keyboard and never thought twice about it.
Historically yes, lately not so much. Have you seen the price of the AW travel charger? And the price of the leather sleeve? The options they force on their Macs so that you have to upgrade to the "best" model to get the graphics and CPU options you want? Yes there is some upfront cost, but the ongoing costs fall from $250-500 per CPU to around $50 per CPU, since most of the R&D is done by the iPhone team.
Having a flawed keyboard that is known to have a very short life and requiring the replacement of the entire top case to fix it - at a cost of $650, is generally something undesirable by most people. I agree, they do usually move to fix it, but then better testing in the first place (and less obsession with thin) and they wouldn't have had the issue blow up quite as it did. This is unrelated to profit though, it was an example of Apple's willingness to shaft its customers. Enough people "mind" the keyboard issues that the 2015 MBP has a higher resale value than the 2016 MBP.
Today, there's NOTHING to be gained from this chip switch. The new Macs are going to be slower and less compatible. Virtualization is gone, we'll be back to slow emulation (eventually, no idea how long it will take for SoftPC to come back). No, I do not believe Apple's speed claims, they're almost certainly lying.
I guess if you're only interested in running Windows on your Mac (as you most certainly seem to be), then yes, you are correct, there's nothing to gain by Apple moving away from Intel. However, Apple isn't much interested in the need for Windows compatibility anymore. They're cutting the cord and moving on. You can hang onto your Intel based Mac as long as you like. Sooner or later there will be emulation software - I wouldn't be surprised if Apple is working with Parallels (and Microsoft), to come up with something feasible soon.
But to think that Apple is lying, is just plain weird. We already have the proof! The DTK has been around for several months and it's been said to be an extremely capable macOS system. That's an SoC from 2 years ago, designed and optimized for high mobility. Plus... the A14!!! Not to mention the supposed "leaked" benchmarks of the A14X, which beats the top end Intel Core i9 in the MacBook Pro 16". And that's an SoC destined for the next iPad Pro.
I don't think they're lying about the benchmarks, but real world performance is far from the same as synthetic benchmarks. And it depends on a lot more than the CPU itself. If the interconnects are slow, benchmarks can look great but general performance can be poor. Samsung for example were at first praised for beating Apple's silicon, then widely panned when it was found they were optimising for tests and the performance was crappy. Memory interconnects can be a big bottleneck, and Apple usually uses package-on-package for the RAM, something they can do up to about 8GB before it gets prohibitively expensive. Then they need an entirely different CPU package to support external RAM. No doubt this will be an excuse for soldered RAM on the 27" iMac too. Another important peripheral is the GPU, Apple/Imagination's GPUs are good mobile GPUs but are orders of magnitude slower than the latest AMD/Nvidia offerings. And slower than the Intel Integrated GPUs too.
Apple hasn't provided any benchmarks regarding their new Apple silicon based Macs. So, no, they can't possibly be lying about that. What they seemed to claim, was that their SoCs will be as powerful as todays desktops, but efficient enough for todays laptops. And as I mentioned, if we go by what they've been capable of with theie AnX SoCs, it's a no brainer.
The Geekbench tests are designed to specifically test the performance of the CPU and GPU cores, not real world performance. So that's is what is being compared. Most people know this does not equate to "real world" performance, but it a base measurement of what that system could be capable of. So when you're comparing macOS on Intel, and macOS and ARM, those benchmarks do actually mean something, because the OS and applications will actually be the same. Give or take a few tweaks here and there for each architecture.
And yes, Apple understands that data interconnects can be an issue, that's one of the biggest reasons why their SoCs share memory between the CPU and GPU. You can bet that every Mac SoC will have at least 8GB onboard, for interaction between the CPU and GPU. The rest of the memory will be offboard, either soldered or on modules, depending on what Apple decides is feasible. Furthermore, you can't compare an SoC (CPU, GPU, ANE, ISP, RAM, etc) that uses 14W with the performance of discreet GPUs that suck well over 100W of power or even a GPU that only uses 45W of power. I honestly don't understand why people think Apple isn't capable of this? They have some of the most talented people designing these chips for them. Their team set out to create a highly efficient, yet powerful mobile SoC... and they have; probably the best mobile chip on the market. There's NO reason, they can't turn their attention towards creating an extremely powerful chip for desktops.
"The sheer volume of iPhones being released means that this device alone has a stranglehold on international shipping in October and November"
I believe that to be a huge exaggeration, therefore very untrue. It is certainly not a reason for Apple not to ship Apple Silicon Macs before December.
Let's hope if the iMac is thin, it will at least have ports on the side instead of - or in addition to - the back. It's absurd that you have to poke around blind to plug things in, form over function yet again.
I agree it's a faff but I think it would look much uglier having all the wires coming off the edge of the screen rather than hanging down the back.
Let's hope if the iMac is thin, it will at least have ports on the side instead of - or in addition to - the back. It's absurd that you have to poke around blind to plug things in, form over function yet again.
I agree it's a faff but I think it would look much uglier having all the wires coming off the edge of the screen rather than hanging down the back.
I agree it would, but for things like cameras or memory sticks that are plugged in temporarily it’d be great in addition to the rear ones. Then it’s up to users how whether they use them or not, not Apple.
When Apple moved from PPC to Intel it was a good thing because not only did Intel finally have some decent chips (remember how much the Pentium 4 and earlier sucked? There's a reason Apple didn't go with Intel in the '90s), but Macs gained the ability to run Windoze software without emulation. Getting one of the new Macs was a very good thing if you didn't need Mac OS 9 compatibility.
Today, there's NOTHING to be gained from this chip switch. The new Macs are going to be slower and less compatible. Virtualization is gone, we'll be back to slow emulation (eventually, no idea how long it will take for SoftPC to come back). No, I do not believe Apple's speed claims, they're almost certainly lying.
People aren't waiting to buy the Apple silicon Macs. People are buying Intel Macs NOW, to avoid the Apple silicon garbage for as long as possible.
This statement was brought to you from a car park next to a porn shop and a crematorium.
I'm am fully betting on Apple Silicon MacBooks to crush sales to students and people needing to work from home. However, Apple will need to set some reasonable prices to attract the cheapskates who are willing to settle for Chromebooks. Apple absolutely must gain laptop market share against Windows laptops running Intel processors. A powerful, super-slim MacBook design with all-day battery life should be extremely attractive to consumers. Apple better not blow this opportunity. It would be too much to hope for if Apple can upset the entire X86 processor laptop market with its Apple ARM processors. Those power-sucking Intel chips have just about hit a wall but the AMD chips have a much better chance.
Complete bull excrement in my opinion. Apple does NOT need to worry about cheapskates. Apple has never worried about the bottom feeders because they will never be able to satisfy that market and shouldn’t even try. Going after the cheapskate market would sully the brand, not enhance it. Apple does not have to gain marketshare in the laptop market either. It’s doing just fine with its premium brand cachet. The Apple Silicon migration will take care of the spec monkeys on its own.
And as for AMD? That’s laughable and I don’t understand the nerd herd’s constant gushing and salivating over AMD when it’s just a bit player in the X86 playfield. Where is AMD’s 5nm chip?
As excited as I am with the promise of Apple Silicon, I can’t help to think if that’s going to be a good move, for me at least.
Even though Macs never really got a sizeable piece of HPC market, I still like to compile some research software (mostly written in C or FORTRAN) for quick calculations (or as an proof of concept) on my MPB. And nothing really comes close to Intel Compilers (even though Intel MKL performance library is only slightly better than Apple Accelerate framework).
Worst than that, CUDA is still king of the hill on GPGPU, and I’ve seen little traction on Metal in changing that.
How did you come to this conclusion? Did you test any of this on a A series processor? I fully expect Apple to exclusively use their own GPUs in the future.
Today, there's NOTHING to be gained from this chip switch. The new Macs are going to be slower and less compatible. Virtualization is gone, we'll be back to slow emulation (eventually, no idea how long it will take for SoftPC to come back). No, I do not believe Apple's speed claims, they're almost certainly lying.
I guess if you're only interested in running Windows on your Mac (as you most certainly seem to be), then yes, you are correct, there's nothing to gain by Apple moving away from Intel. However, Apple isn't much interested in the need for Windows compatibility anymore. They're cutting the cord and moving on. You can hang onto your Intel based Mac as long as you like. Sooner or later there will be emulation software - I wouldn't be surprised if Apple is working with Parallels (and Microsoft), to come up with something feasible soon.
But to think that Apple is lying, is just plain weird. We already have the proof! The DTK has been around for several months and it's been said to be an extremely capable macOS system. That's an SoC from 2 years ago, designed and optimized for high mobility. Plus... the A14!!! Not to mention the supposed "leaked" benchmarks of the A14X, which beats the top end Intel Core i9 in the MacBook Pro 16". And that's an SoC destined for the next iPad Pro.
I don't think they're lying about the benchmarks, but real world performance is far from the same as synthetic benchmarks. And it depends on a lot more than the CPU itself. If the interconnects are slow, benchmarks can look great but general performance can be poor. Samsung for example were at first praised for beating Apple's silicon, then widely panned when it was found they were optimising for tests and the performance was crappy. Memory interconnects can be a big bottleneck, and Apple usually uses package-on-package for the RAM, something they can do up to about 8GB before it gets prohibitively expensive. Then they need an entirely different CPU package to support external RAM. No doubt this will be an excuse for soldered RAM on the 27" iMac too. Another important peripheral is the GPU, Apple/Imagination's GPUs are good mobile GPUs but are orders of magnitude slower than the latest AMD/Nvidia offerings. And slower than the Intel Integrated GPUs too.
Apple hasn't provided any benchmarks regarding their new Apple silicon based Macs. So, no, they can't possibly be lying about that. What they seemed to claim, was that their SoCs will be as powerful as todays desktops, but efficient enough for todays laptops. And as I mentioned, if we go by what they've been capable of with theie AnX SoCs, it's a no brainer.
The Geekbench tests are designed to specifically test the performance of the CPU and GPU cores, not real world performance. So that's is what is being compared. Most people know this does not equate to "real world" performance, but it a base measurement of what that system could be capable of. So when you're comparing macOS on Intel, and macOS and ARM, those benchmarks do actually mean something, because the OS and applications will actually be the same. Give or take a few tweaks here and there for each architecture.
And yes, Apple understands that data interconnects can be an issue, that's one of the biggest reasons why their SoCs share memory between the CPU and GPU. You can bet that every Mac SoC will have at least 8GB onboard, for interaction between the CPU and GPU. The rest of the memory will be offboard, either soldered or on modules, depending on what Apple decides is feasible. Furthermore, you can't compare an SoC (CPU, GPU, ANE, ISP, RAM, etc) that uses 14W with the performance of discreet GPUs that suck well over 100W of power or even a GPU that only uses 45W of power. I honestly don't understand why people think Apple isn't capable of this? They have some of the most talented people designing these chips for them. Their team set out to create a highly efficient, yet powerful mobile SoC... and they have; probably the best mobile chip on the market. There's NO reason, they can't turn their attention towards creating an extremely powerful chip for desktops.
Surely they could be aiming the design at largely free of volitile memory?
Tap Micron XPoint production for 64Gb extra fast non-volitile RAM then use the old hydrid drive code to manage between slow and fast storage.
Comments
Our secondary school (UK) had one Mac before Intel. When Apple switched to Intel, most of the school switched to Macs - around 300 Macs in total, because they ran Windows in a VM for all the Windows only admin stuff. The Macs were also pretty cheap, special education iMacs for about £500 each in today's money. Good luck getting an education Mac now for £500. A lot of that Windows only admin stuff still exists today, and as such they will be switching back to PCs. There was a big spike in Mac ownership post-Intel, presumably it can partly be credited to the reassurance that people could still run Windows.
I guess if you're only interested in running Windows on your Mac (as you most certainly seem to be), then yes, you are correct, there's nothing to gain by Apple moving away from Intel. However, Apple isn't much interested in the need for Windows compatibility anymore. They're cutting the cord and moving on. You can hang onto your Intel based Mac as long as you like. Sooner or later there will be emulation software - I wouldn't be surprised if Apple is working with Parallels (and Microsoft), to come up with something feasible soon.
But to think that Apple is lying, is just plain weird. We already have the proof! The DTK has been around for several months and it's been said to be an extremely capable macOS system. That's an SoC from 2 years ago, designed and optimized for high mobility. Plus... the A14!!! Not to mention the supposed "leaked" benchmarks of the A14X, which beats the top end Intel Core i9 in the MacBook Pro 16". And that's an SoC destined for the next iPad Pro.
What's media access control address, aka MAC, got to do with it? Or were you referring to MAC Cosmetics?
It's Mac, never MAC. If you're a newbie, you're forgiven. If you've been around the Mac market for years, then you really ought to know better.
FYI the best predictor of future behaviour, is past behaviour.
How about something along the lines of WATCH, tv, tv+ …
Book
Book Pro
Mac mini
Mac
Mac Pro
It's really easy to play the "greed" card. But Apple is a company that has historically ignored "profit above all else". This is evident in a lot of what they do. Second, there is not more profit in a transition... there is an enormous up front cost in the development and support and engineering resources. It is a very expensive undertaking.
People will always hate. That doesn't make those things invalid, it makes it undesirable by those people. There's huge list of Apple products that people hated. Hell, there's a ton of people who just hate Apple. That's not going to stop Apple from creating and believing in the products they make. If users are having issues, Apple moves to correct them. Sometimes it takes longer, sometimes it doesn't.
Apple obviously believed the butterfly key design was superior - something that probably took a lot of resources and money to design and develop - so instead of just replacing it, they tried to make it less and less prone to becoming faulty. And by the way, a vast majority of people didn't mind that keyboard and never thought twice about it.
Having a flawed keyboard that is known to have a very short life and requiring the replacement of the entire top case to fix it - at a cost of $650, is generally something undesirable by most people. I agree, they do usually move to fix it, but then better testing in the first place (and less obsession with thin) and they wouldn't have had the issue blow up quite as it did. This is unrelated to profit though, it was an example of Apple's willingness to shaft its customers. Enough people "mind" the keyboard issues that the 2015 MBP has a higher resale value than the 2016 MBP.
Apple hasn't provided any benchmarks regarding their new Apple silicon based Macs. So, no, they can't possibly be lying about that. What they seemed to claim, was that their SoCs will be as powerful as todays desktops, but efficient enough for todays laptops. And as I mentioned, if we go by what they've been capable of with theie AnX SoCs, it's a no brainer.
The Geekbench tests are designed to specifically test the performance of the CPU and GPU cores, not real world performance. So that's is what is being compared. Most people know this does not equate to "real world" performance, but it a base measurement of what that system could be capable of. So when you're comparing macOS on Intel, and macOS and ARM, those benchmarks do actually mean something, because the OS and applications will actually be the same. Give or take a few tweaks here and there for each architecture.
And yes, Apple understands that data interconnects can be an issue, that's one of the biggest reasons why their SoCs share memory between the CPU and GPU. You can bet that every Mac SoC will have at least 8GB onboard, for interaction between the CPU and GPU. The rest of the memory will be offboard, either soldered or on modules, depending on what Apple decides is feasible. Furthermore, you can't compare an SoC (CPU, GPU, ANE, ISP, RAM, etc) that uses 14W with the performance of discreet GPUs that suck well over 100W of power or even a GPU that only uses 45W of power. I honestly don't understand why people think Apple isn't capable of this? They have some of the most talented people designing these chips for them. Their team set out to create a highly efficient, yet powerful mobile SoC... and they have; probably the best mobile chip on the market. There's NO reason, they can't turn their attention towards creating an extremely powerful chip for desktops.
I believe that to be a huge exaggeration, therefore very untrue.
It is certainly not a reason for Apple not to ship Apple Silicon Macs before December.
And as for AMD? That’s laughable and I don’t understand the nerd herd’s constant gushing and salivating over AMD when it’s just a bit player in the X86 playfield. Where is AMD’s 5nm chip?