I know some have said the new iMac won't be ARM but is there any chance Microsoft and Adobe have been given a heads up about this move, or that their iOS ports have been done with an easier transfer to ARM macOS? I wonder if the iMac could be announced at WWDC, on ARM but with shipping not until later in the year?
It really needs an update before that, for the T2.
techconc said: Not likely. Apple needs to set a direction for the platform. Previous transitions were successful because they were actual transitions. Developers got onboard because they had to. Making support for ARM optional is a recipe for failure. Just ask Microsoft about that.
I sort of agree, but they could certainly stage it. They just released the new Mac Pro they took so long on. Do you think they'll just announce ARM at WWDC and then go, sorry all you folks who just bought your $5k+++ Mac Pro, we're now announcing its effective EOL a couple years from now?
techconc said: Intel has become stagnant and uncompetitive. Nobody can really deny that. Apple’s cores are already on par with Intel designs on devices that are extremely constrained for power. Imagine what Apple can do on devices where that constraint is removed. Apple has a competitive advantage on their mobile platforms because of their excellent chip design team. Apple can now gain a competitive advantage on the desktop by moving to their own custom ARM processors.
Well, I think that is a bit more rhetoric than reality. It has become a meme that people really haven't thought through. I remember when the laptops hadn't been updated for so long and people were (here, even AI writers and podcasters) saying how there was no need for updates because.... Intel. Then, a couple weeks later, updates and articles talking about the relatively large performance gains of the new laptops.
No, the updates are no longer leaps and bounds faster each increment, but they don't need to be. There has been quite a bit of gain in terms of cores and more performance for less power. Would ARM make some nice jumps there if done right? Yeah, I suppose for a couple years until they run the rapid advance out there as well.
It's easy to forget the reality, especially when following headlines. I still remember all the 'performance advantage' language around that G5 to Intel transition, when in reality, the G5 was beating anything Intel had at the time. It took some time before the extra cores started to effectively beat the G5. It was a good decision for longevity, and Intel compatibility, but it wasn't, at least initially, about performance. I still hear people tell that story, as history has been rewritten in people's minds.
tht said: My crazy idea is to have 2 to 3 front facing cameras, one on each side, for iPad too. During a video call, the camera video data goes through an image correction routine to make it appear as if I'm looking straight into the camera. Apple was experimenting with this with one camera, probably still are. Well, with one camera's worth of data it sounds like some creation of facial data is being done. 2 or 3 cameras? That's basically just interpolation with some correction for shadows.
I really don't mind a border at all, besides the impact on screen size vs overall dimensions. In fact, I prefer a bit of border. But, the camera thing for mobile devices is a really good point. It sucks to have the camera to the side, when landscape is probably the most used orientation.
CheeseFreeze said: Another reason could be is that the iPad and MacBook Pro are more ‘personal’ devices used on your lap, or holding with your hands, close to your face. A desktop is more public and therefore FaceTime could activate through ‘false positives’. The web camera is quite wide so they’d have to experiment with a FaceID camera that only ‘looks’ for someone clearly engaging with the device.
Yeah, I'd WAY rather have a Touch ID keyboard or trackpad that could be used with any Mac. I'm not crazy about Face ID on the phone, and I'm REALLY not crazy about it on the Mac.
techconc said: Intel has become stagnant and uncompetitive. Nobody can really deny that. Apple’s cores are already on par with Intel designs on devices that are extremely constrained for power. Imagine what Apple can do on devices where that constraint is removed. Apple has a competitive advantage on their mobile platforms because of their excellent chip design team. Apple can now gain a competitive advantage on the desktop by moving to their own custom ARM processors.
Well, I think that is a bit more rhetoric than reality. It has become a meme that people really haven't thought through. I remember when the laptops hadn't been updated for so long and people were (here, even AI writers and podcasters) saying how there was no need for updates because.... Intel. Then, a couple weeks later, updates and articles talking about the relatively large performance gains of the new laptops.
No, the updates are no longer leaps and bounds faster each increment, but they don't need to be. There has been quite a bit of gain in terms of cores and more performance for less power. Would ARM make some nice jumps there if done right? Yeah, I suppose for a couple years until they run the rapid advance out there as well.
It's easy to forget the reality, especially when following headlines. I still remember all the 'performance advantage' language around that G5 to Intel transition, when in reality, the G5 was beating anything Intel had at the time. It took some time before the extra cores started to effectively beat the G5. It was a good decision for longevity, and Intel compatibility, but it wasn't, at least initially, about performance. I still hear people tell that story, as history has been rewritten in people's minds.
If you compare the new 10th Gen 13” MacBook Pro performance to previous generations with 4 cores, there really isn’t much improvement. The latest Geekbench 5 scores show about 1280/4600 with the Core i7-1068NG7. My 2018 13” MacBook Pro gets 1160/4400 with a Core i7-8559U. That’s a 10% improvement in single core and a 4.5% improvement in multi core. Not exactly a huge difference.
The G5 might have been a decently fast CPU but Apple didn’t switch to Intel because the top of the line desktop PPC CPU wasn’t fast enough. They switched away because IBM couldn’t supply a decently fast laptop CPU. The G4 was totally stagnant and the G5 needed water cooling and a huge power supply on the Mac Pro. There was no chance for it to be viable in a battery constrained environment.
It's easy to forget the reality, especially when following headlines. I still remember all the 'performance advantage' language around that G5 to Intel transition, when in reality, the G5 was beating anything Intel had at the time. It took some time before the extra cores started to effectively beat the G5. It was a good decision for longevity, and Intel compatibility, but it wasn't, at least initially, about performance. I still hear people tell that story, as history has been rewritten in people's minds.
That is how I recall it too. The G5 was a beast, but it really needed some [loud] cooling. The 2006 intel iMac was a step back in many ways from the iMac G5, but was a sign of things to come.
blastdoor said: Really, I don't think it's bizarre. They should always be making the best product they can at any given point in time. There's no reason to intentionally hold a product design back.
That's what I actually mean... why has the iMac remained based on such an old design/internals for so long, compared to the entire rest of the Mac lineup?
The intervals are due for an update but aren’t really too old. Comet lake is just a price cut, so the CPU isn’t really out of date (at least, not for Intel). The GPU could use an upgrade, though.
in terms of physical appearance, it has been a very compelling design for a long time. It is hard to improve on a design that good. For example, you cannot find many displays better than what is in the 27 inch iMac.
I’ll guess that whatever the new physical design is, it required multiple technical advances in order to be feasible.
The design feels really dated at this point. The surface studio is much better looking. Apple should have updated this years ago. Huge bezels and a chin. It is overdue for a full refresh. An oversized iPad Pro would be perfect. I wouldn’t mind seeing a lay flat mode and Apple Pencil support (touch not needed) for creative apps on some models, but that is probably expecting too much. Sidecar is a decent solution, but it would be nice to have a single larger screen.
jdb8167 said: If you compare the new 10th Gen 13” MacBook Pro performance to previous generations with 4 cores, there really isn’t much improvement. The latest Geekbench 5 scores show about 1280/4600 with the Core i7-1068NG7. My 2018 13” MacBook Pro gets 1160/4400 with a Core i7-8559U. That’s a 10% improvement in single core and a 4.5% improvement in multi core. Not exactly a huge difference.
The G5 might have been a decently fast CPU but Apple didn’t switch to Intel because the top of the line desktop PPC CPU wasn’t fast enough. They switched away because IBM couldn’t supply a decently fast laptop CPU. The G4 was totally stagnant and the G5 needed water cooling and a huge power supply on the Mac Pro. There was no chance for it to be viable in a battery constrained environment.
Yeah, I suppose it depends on which generations we're talking about. I'm not saying it is rapidly advancing, but it isn't stagnant either. A lot of it has come via more cores and efficiency instead of brute speed increase. I actually prefer more cores though, so I'm a bit biased in that direction.
re: G5 - Yes, for sure. It was about power consumption and laptops initially. I was just correcting the often heard belief that it was because Apple was behind on performance.
entropys said: That is how I recall it too. The G5 was a beast, but it really needed some [loud] cooling. The 2006 intel iMac was a step back in many ways from the iMac G5, but was a sign of things to come.
At the time, the 3D software I used had a user-created performance tracking system. A G5 was at the top for quite some time (even after the transition), and all sorts of wild custom-built machines were trying to dethrone it. It had a pretty good margin.
I sort of agree, but they could certainly stage it. They just released the new Mac Pro they took so long on. Do you think they'll just announce ARM at WWDC and then go, sorry all you folks who just bought your $5k+++ Mac Pro, we're now announcing its effective EOL a couple years from now?
Yes, that's exactly what I expect. Apple has done the CPU transition twice before. Let's not pretend this is the first time. Of course, they will acknowledge how wonder the Intel based Macs are today. Then in about 2 years, Apple will brag about how smooth the transition went. So smooth that it's going to stop including Rosetta based emulation, etc. Then, the requirement for fat binaries will go away, etc...
It's easy to forget the reality, especially when following headlines. I still remember all the 'performance advantage' language around that G5 to Intel transition, when in reality, the G5 was beating anything Intel had at the time. It took some time before the extra cores started to effectively beat the G5. It was a good decision for longevity, and Intel compatibility, but it wasn't, at least initially, about performance. I still hear people tell that story, as history has been rewritten in people's minds.
Yes, the G5 was competitive with the Pentium 4 series, but no so much with the new Core Duo. Motorola had effectively dropped out with left just IBM. IBM failed to deliver 3GHz chips as promised. Also, they ran hot. There was no way you were getting a G5 into a laptop. Given how the direction for everything was going mobile, that was a big deal. That's not rewriting history, that was the motivation to switch. The transition presentation wasn't about top end speed, it was about performance per watt and how Intel's new Core Duo chips were much better aligned for future products.
Fast forward to today and the message is the same. ARM based designs are much more efficient in a performance per watt basis and Intel has become stagnant. Intel used to lead the industry by being first to market with the latest chip fabrication process. That used to make up for their shortcoming in chip design. Today, they neither lead in chip design nor in chip fabrication process. Intel is so proud about finally getting their 10nm process to market. Meanwhile the rest of the industry has been at 7nm and now moving to the 5nm process. Each new generation of chip comes with the most meager performance increases these days. Meanwhile, the cores in Apple's designs for phone class devices have matched the performance of Intel's desktop class chips. I'm talking single core performance, not multi-core. It's pretty ridiculous that a 2 year old iPad Pro is more powerful than a brand new MacBook Air. Obviously, Apple is aware of this disparity which is why the current iPad Pro didn't get a CPU bump to an A13x class chip. It's an awkward situation for Apple and they need to address it.
Let's see... A super thin, battery-powered iMac with only a single port.
No thanks. I'll take USB-A, the SD card slot, TB3 and ports galore, even if that means keeping the bezels and chin.
Practicality... It does every Mac user good.
No doubt... most stuff is thin/small enough already, especially iMacs. And, most battery powered things now last all day, which doesn't add much more benefit if they lated two days.
re: ports - especially on a desktop. It is silly to make a machine where you immediately have to add a dock unit (extra $, extra clutter) just some some industrial designer could get off on some kind of port-minimalism.
techconc said: ... That's not rewriting history, that was the motivation to switch. The transition presentation wasn't about top end speed, it was about performance per watt and how Intel's new Core Duo chips were much better aligned for future products. ...
Yes, I don't disagree. The rewriting history part, though, is that I often hear it put in terms of Apple falling further and further behind in performance, so they had to switch to Intel to catch up. Performance per watt, yes. But, in terms of raw performance, Apple was killing EVERYTHING else on the market at that point, even custom builds.
Comments
I sort of agree, but they could certainly stage it. They just released the new Mac Pro they took so long on. Do you think they'll just announce ARM at WWDC and then go, sorry all you folks who just bought your $5k+++ Mac Pro, we're now announcing its effective EOL a couple years from now?
Well, I think that is a bit more rhetoric than reality. It has become a meme that people really haven't thought through. I remember when the laptops hadn't been updated for so long and people were (here, even AI writers and podcasters) saying how there was no need for updates because.... Intel. Then, a couple weeks later, updates and articles talking about the relatively large performance gains of the new laptops.
No, the updates are no longer leaps and bounds faster each increment, but they don't need to be. There has been quite a bit of gain in terms of cores and more performance for less power. Would ARM make some nice jumps there if done right? Yeah, I suppose for a couple years until they run the rapid advance out there as well.
It's easy to forget the reality, especially when following headlines. I still remember all the 'performance advantage' language around that G5 to Intel transition, when in reality, the G5 was beating anything Intel had at the time. It took some time before the extra cores started to effectively beat the G5. It was a good decision for longevity, and Intel compatibility, but it wasn't, at least initially, about performance. I still hear people tell that story, as history has been rewritten in people's minds.
I really don't mind a border at all, besides the impact on screen size vs overall dimensions. In fact, I prefer a bit of border. But, the camera thing for mobile devices is a really good point. It sucks to have the camera to the side, when landscape is probably the most used orientation.
Yeah, I'd WAY rather have a Touch ID keyboard or trackpad that could be used with any Mac. I'm not crazy about Face ID on the phone, and I'm REALLY not crazy about it on the Mac.
The G5 might have been a decently fast CPU but Apple didn’t switch to Intel because the top of the line desktop PPC CPU wasn’t fast enough. They switched away because IBM couldn’t supply a decently fast laptop CPU. The G4 was totally stagnant and the G5 needed water cooling and a huge power supply on the Mac Pro. There was no chance for it to be viable in a battery constrained environment.
re: G5 - Yes, for sure. It was about power consumption and laptops initially. I was just correcting the often heard belief that it was because Apple was behind on performance.
At the time, the 3D software I used had a user-created performance tracking system. A G5 was at the top for quite some time (even after the transition), and all sorts of wild custom-built machines were trying to dethrone it. It had a pretty good margin.
No thanks. I'll take USB-A, the SD card slot, TB3 and ports galore, even if that means keeping the bezels and chin.
Practicality... It does every Mac user good.
Yes, the G5 was competitive with the Pentium 4 series, but no so much with the new Core Duo. Motorola had effectively dropped out with left just IBM. IBM failed to deliver 3GHz chips as promised. Also, they ran hot. There was no way you were getting a G5 into a laptop. Given how the direction for everything was going mobile, that was a big deal. That's not rewriting history, that was the motivation to switch. The transition presentation wasn't about top end speed, it was about performance per watt and how Intel's new Core Duo chips were much better aligned for future products.
Fast forward to today and the message is the same. ARM based designs are much more efficient in a performance per watt basis and Intel has become stagnant. Intel used to lead the industry by being first to market with the latest chip fabrication process. That used to make up for their shortcoming in chip design. Today, they neither lead in chip design nor in chip fabrication process. Intel is so proud about finally getting their 10nm process to market. Meanwhile the rest of the industry has been at 7nm and now moving to the 5nm process. Each new generation of chip comes with the most meager performance increases these days. Meanwhile, the cores in Apple's designs for phone class devices have matched the performance of Intel's desktop class chips. I'm talking single core performance, not multi-core. It's pretty ridiculous that a 2 year old iPad Pro is more powerful than a brand new MacBook Air. Obviously, Apple is aware of this disparity which is why the current iPad Pro didn't get a CPU bump to an A13x class chip. It's an awkward situation for Apple and they need to address it.
re: ports - especially on a desktop. It is silly to make a machine where you immediately have to add a dock unit (extra $, extra clutter) just some some industrial designer could get off on some kind of port-minimalism.
Yes, I don't disagree. The rewriting history part, though, is that I often hear it put in terms of Apple falling further and further behind in performance, so they had to switch to Intel to catch up. Performance per watt, yes. But, in terms of raw performance, Apple was killing EVERYTHING else on the market at that point, even custom builds.