Anyone else used to be very pro-ARM but now having second thoughts?

[Yebubbleman]

I don't know whose comments you're reading but clearly it isn't mine. I said his latest prediction was wrong. Which it is. He's already backtracked more than half of it. I also said he didn't really know that much more than most people which you seem to agree with while simultaneously stating that he all but never makes a mistake. I'll state again though, his predictions look more accurate and/or informed than they are. I took his prediction into consideration when I bought my 16", but now its looking less and less likely to be accurate.

I said you were defending his honour, not yours.
Again, I much prefer the 16" to stay as is for as long as possible. If you're going to reply again, please make sure you've read and understood what I said otherwise you're just arguing with yourself.
The only sign the 16" won't be first is Kuo's shonky prediction. Clearly you're not reading the rest of the thread either.

I've boldfaced the things that you say that are baseless at this point in time.

I've never said that he never makes a mistake. He DOES make mistakes, but they're few and far between. The most important details are correct. The only one I can recall in recent memory was about the Fifth Generation iPad mini (though I'll bet that's more a result of Apple changing course on that internally after a while).

You can't tell that his current prediction is wrong because the future has not happened yet (unless you're referring to a different prediction).

I don't care about defending the honor of a man I won't ever meet based on what he talks about with Apple. I don't take his word as Gospel, but if I had to go to Las Vegas and bet on Apple's next moves, I'd take what he had to say into consideration more than I wouldn't. That's really all I said. The rest are words you seem to want to put into my mouth without actually reading what I'm saying.

There's nothing to indicate that Apple can surpass the power of the current higher-end Intel 16" MacBook Pros this calendar year. Next calendar year is very likely. But not this one. The processor in the DTK can comfortably overtake any currently shipping 6-core i7 based Mac. Past that, and Intel is still faster for the moment. Apple would need to make a DRASTICALLY faster (and I'm talking much faster than the disparity between the Pentium 4 based Intel DTK and the first 15" MacBook Pro) SoC. Considering the iPad Pro was given an A12Z and not an A13X, I'm guessing we're not there yet. You seem to ignore these FACTS which, by the way, have nothing to do with Kuo. Furthermore, you say nothing to discredit them other than "it makes bad business sense".

Point me in the direction of things that negate what I have to say or yield and stop this stupid argument, please.

 

[UBS28]

People have been making these claims for almost ten years. Still no signs of locking down.
[automerge]1594382863[/automerge]


The issue are the drivers and the API. The architecture does not matter if the software implementation is different. MacOS does not implement DirectX, neither did it ship gaming quality drivers. OpenGL never properly worked for this purpose either. Metal made things much better but still, driver quality was a big issue. The IHVs simply don’t have the incentive to polish their drivers - their sales is not driven by the gaming performance they provide.

With Apple Silicon, Macs will have both gaming grade hardware and gaming grade drivers. They will require the developer to use a new API or at least an engine that uses that API. But this will make low-latency, predictable performance gaming a possibility on macOS - for the first time in the modern times.

You are very naive to think this will not happen at this point. Even blind people can see that Mac‘s have been transitioning towards being an iOS device. (an iPad Pro is an ARM MAC as it is capable of running Mac Big Sur. The hardware inside the ARM Mac Mini is an iPad Pro).

Now bootcamp and Linux is gone (unless you like to play around with the ARM version). Wait for more to happen.

 

[leman]

You are very naive to think this will not happen at this point. Even blind people can see that Mac‘s have been transitioning towards being an iOS device. (an iPad Pro is an ARM MAC as it is capable of running Mac Big Sur. The hardware inside the ARM Mac Mini is an iPad Pro).

I believe it’s the other way around. The iPad has been gaining some flexibility usually reserved to the desktop, things like rudimentary file management, mouse support etc. But it’s still a managed environment. Again, macOS allows me to manage my data and software and external connectivity in any way I want, without restrictions of any kind. This is the core difference to iOS and this has not changed at all in the last decade.

The second part of your argument I can’t follow at all. Yes, iPhone and iPad hardware can run MacOS. So what? It’s about how you configure the software side of the platform, the hardware is flexible enough. Using compatible hardware has clear economical and stability advantages - you can cut costs and development time.

 

[Waragainstsleep]

Point me in the direction of things that negate what I have to say or yield and stop this stupid argument, please.

The 13" MacBook Pro is looking like it will be the first with ARM, not the 16". This isn't a matter of making "business sense" or not.

As I stated, one report says this. You've offered none of the corroborative articles you claim exist citing other sources besides Kuo. Also, business sense always matters in business.

Apple has the capability to put out a machine with their SoCs that handily outperforms any and all Intel-based 13" MacBook Pros TODAY. The same isn't necessarily a given for all Intel-based 16" MacBook Pros TODAY. Apple might be able to best the 6-core i7 models TODAY, but the 8-core i9 models aren't necessarily a given.

There are overwhelming technical indicators (not just speculative analyst guesses) that Apple can easily better the i9 in the current MBP. The part that's not a given is if they can beat the Radeon Pro 5600M.

You do realize that one analyst is spot on like 95% of the time, right?

There've been others corroborating it too; to my knowledge, it's not just him.

Both unsubstantiated.

Apple doesn't yet have something to take on the 9th Gen 8-Core Core i9.

They almost certainly do.

The timing is not most of the trick. Even if he's off by a month or two, he gets the order right most of the time.
You've got Bloomberg, Sonny Dickson and Kuo. All of them are closer to accurate than any source of Apple rumors has been in the history of Mac rumors (and MacRumors). All Kuo does is watch what Apple is doing publicly and speculate from there. He doesn't suck at it even if his timing isn't 100%.

I said, he's right 95% of the time, not 100% of the time.

Benchmarks from machines that never see the light of day are not unheard of.

Something else you could substantiate.

I'm just disagreeing with someone who has already backtracked on a prediction that wasn't as 95% as you claim. You're the one defending his honour like it matters.

I have no honor that I care about defending to someone I'm never going to meet on an online forum.

No one said anything about your honour.

You're saying the man's predictions don't mean squat.

I said no such thing.

That and it sounds like you have your heart set on the 16" MacBook Pro going first when there are a lot of signs to indicate that is not going to happen, Kuo's prediction only being one of which.

The opposite of what I said followed by a repeat of your unsubstantiated claim.

We know that they're at least at the level of the A12Z and we can track their rate of progress.

Yet you seem to be refusing to grant them any progress since 2018.

I've never said that he never makes a mistake. He DOES make mistakes, but they're few and far between. The most important details are correct. The only one I can recall in recent memory was about the Fifth Generation iPad mini (though I'll bet that's more a result of Apple changing course on that internally after a while).

You can't tell that his current prediction is wrong because the future has not happened yet (unless you're referring to a different prediction).

To summarise, his prediction was that the 13" MBP and iMac would be first Intel Macs transitioned to AS. He has since dropped the iMac (almost certainly because a benchmark for a new Intel one has been leaked) which you have ignored, and he has pretty much added in the MacBook Air to his list of first AS Macs. Since everyone knows with 99.9%+ certainty that the Mac Pro will not be the first AS Mac, that leaves 6 candidates. MBA, 2xMBP, Mini and 2x iMac. So far he has predicted half of these. If any one of them is first you will doubtless put a tick in his win column because "timing doesn't matter".

I don't care about defending the honor of a man I won't ever meet based on what he talks about with Apple.

See the above list of quotes of you defending his honour.

There's nothing to indicate that Apple can surpass the power of the current higher-end Intel 16" MacBook Pros this calendar year.

One could argue the MBP being first to Intel was an indicator. Plus the reasonable likelihood Apple can beat the current ones (Made easier by the fact they didn't update the 16" to 10th gen Intel before the 13"). None of it is proof but these are more solid indicators than someone relying on supply chain leaks.

The processor in the DTK can comfortably overtake any currently shipping 6-core i7 based Mac. Past that, and Intel is still faster for the moment.

That you know of. And again, that A12Z is nearly 3 years old.

Apple would need to make a DRASTICALLY faster (and I'm talking much faster than the disparity between the Pentium 4 based Intel DTK and the first 15" MacBook Pro) SoC. Considering the iPad Pro was given an A12Z and not an A13X, I'm guessing we're not there yet. You seem to ignore these FACTS which, by the way, have nothing to do with Kuo. Furthermore, you say nothing to discredit them other than "it makes bad business sense".

I'm guessing they decided they didn't need to make an A13X/Z as nothing else on the market is close to touching the iPad Pro.

Point me in the direction of things that negate what I have to say or yield and stop this stupid argument, please.

Done.

 

[dogslobber]

The problem with these ARM chips is Intel. They’re planning to release hybrid x86 chips which are similar to what Apple is claiming will be their strength. I can see Apple going back to Intel in the future for some machines.

 

[KPOM]

The only concern I have with the first Apple Silicon Macs is that they likely are built on ARMv8 and ARMv9 is coming next year. This could be like the first Intel Macs being built on x86-32 with the x86-64 chips arriving less than a year later. The first Intel Macs got only 1 more OS update than the last PowerPC Macs, though the second generation Intel was supported for longer.

 

[Yebubbleman]

As I stated, one report says this. You've offered none of the corroborative articles you claim exist citing other sources besides Kuo. Also, business sense always matters in business.

Business sense doesn't trump what is technically feasible and available. Also, you do a piss poor job of defending the notion that releasing a 13" MacBook Pro on ARM before the 16" MacBook Pro on ARM is poor business sense (especially when the former is a more popular model). You haven't explained or defended anything there. You've just done the equivalent of what a five year old does when he tells another five year old that he or she is wrong.

There are overwhelming technical indicators (not just speculative analyst guesses) that Apple can easily better the i9 in the current MBP.

NAME ONE!

Yet you seem to be refusing to grant them any progress since 2018.

In October of 2018, the 2018 iPad Pro is released with the A12X processor.

Flash forward to just a few months ago in 2020 and the 2020 iPad Pros is released with the A12Z processor. The A12Z processor saw the light of day THIS YEAR, not in 2018. Furthermore, if Apple had a better processor to either include in this year's iPads or in the DTK, what reason would they have, "business sense" or otherwise, or otherwise?

To summarise, his prediction was that the 13" MBP and iMac would be first Intel Macs transitioned to AS. He has since dropped the iMac (almost certainly because a benchmark for a new Intel one has been leaked) which you have ignored, and he has pretty much added in the MacBook Air to his list of first AS Macs. Since everyone knows with 99.9%+ certainty that the Mac Pro will not be the first AS Mac, that leaves 6 candidates. MBA, 2xMBP, Mini and 2x iMac. So far he has predicted half of these. If any one of them is first you will doubtless put a tick in his win column because "timing doesn't matter".

He never backpedaled on the iMac. He just started talking about MacBooks. A lack of chatter isn't a backpedaling. Also, you REALLY don't know your Apple history. The Intel iMac was the first Mac to make the jump to Intel, not the MacBook Pro. If timing is everything, you're showing that there's a lot you don't know...

Also, dude, you're on MacRumors. Find me someone who is 100% right on their Apple predictions or 100% of the time. Also, for the last freakin' time, I never said Ming Chi Kuo was 100% right. I said he's right enough of the time to be worth giving attention to. This site and most news outlets agree. Get over it.

One could argue the MBP being first to Intel was an indicator. Plus the reasonable likelihood Apple can beat the current ones (Made easier by the fact they didn't update the 16" to 10th gen Intel before the 13"). None of it is proof but these are more solid indicators than someone relying on supply chain leaks.

The difference in performance between the 9th Gen Core i9 H processors and their 10th Gen equivalents is minimal. Apple isn't always prompt to upgrading 15"/16" MacBook Pros. However, the 8th Gen based 13" needed the upgrade much more. Take what you will from that.

Also, you really don't know your Apple history. The 17" and 20" iMacs were first to transition to Intel from PowerPC, not the 15" PowerBook G4/15" MacBook Pro.

That you know of. And again, that A12Z is nearly 3 years old.

The A12X is not even two years old. The A12Z came out THIS YEAR with the 2020 iPad Pro and it's what was put in the DTK. If timing is everything, you're showing that there's a lot you don't know...

 

[macsplusmacs]

The problem with these ARM chips is Intel. They’re planning to release hybrid x86 chips which are similar to what Apple is claiming will be their strength. I can see Apple going back to Intel in the future for some machines.

Similar.

In the same way, I am similar to Mr. Jordan when I pick up a basketball. Plus. why would ANYONE believe the future plans and a meeting of deadlines from Intel, guess maybe someone that was in a coma the past six years maybe?


Going back to intel?

Nope. Never. going. to. happen.

 

[dogslobber]

Similar.

In the same way, I am similar to Mr. Jordan when I pick up a basketball. Plus. why would ANYONE believe the future plans and a meeting of deadlines from Intel, guess maybe someone that was in a coma the past six years maybe?

I've no idea what you are talking about.

Going back to intel?

Nope. Never. going. to. happen.

That is your unqualified opinion. Doesn't have any relevance to really, obviously.

 

[macsplusmacs]

Seeing the progress of the A Series and the Mollases like "progress" of the Intel chips makes me more then qualified enough.

"I've no idea what you are talking about."

Research intel chiplets. maybe you will.


And bookmark my comment. In 10 years I will be right and you will be wrong about them going back to intel.

 

[Waragainstsleep]

Business sense doesn't trump what is technically feasible and available.

But you have no idea what is technically feasible or available. Only what was feasible many many months ago, because you keep ignoring strong indicators that don't support your point.

Also, you do a piss poor job of defending the notion that releasing a 13" MacBook Pro on ARM before the 16" MacBook Pro on ARM is poor business sense (especially when the former is a more popular model).

Theres no job to do as far as any sane person is concerned. A 13" that is better than the 16" will eat into 16" sales. The 16" is substantially more expensive so that is bad for business.

NAME ONE!

So you can ignore them again? I'm getting sick of repeating myself but luckily there is new one to add.
Apple has a chip over two years ago that can beat anything Intel has below 8 cores when it comes to single threaded performance. It did this with less RAM, and passive cooling. Now they have one that can beat that one with two fewer cores and in a water tight case the size of 3-4 credit cards.

In October of 2018, the 2018 iPad Pro is released with the A12X processor.

Flash forward to just a few months ago in 2020 and the 2020 iPad Pros is released with the A12Z processor. The A12Z processor saw the light of day THIS YEAR, not in 2018. Furthermore, if Apple had a better processor to either include in this year's iPads or in the DTK, what reason would they have, "business sense" or otherwise, or otherwise?

The A12Z is exactly the same as the A12X but with one extra GPU core enabled. The core was always there, they just improved the yield enough to make use of it now. The design is a 2018 chip.
They have an A14 6 core chip that beats the A12Z now. They didn't put that in the DTK because the DTK doesn't need anything fancier than the A12Z which they are making in numbers with excellent yield and everyone already knows about. They need the A14 for iPhone 12. The Pentium 4 in the Intel DTK was public knowledge even before it came out as it would have been on Intel's roadmap in advance. It was bought off the shelf. Apple just used their own shelf and they didn't use anything newer because they weren't about to design a chip for a low unit dev box and they don't ever give away their new tech before they need to if there is any benefit to surprising people. Which there almost always is with Apple.

He never backpedaled on the iMac. He just started talking about MacBooks. A lack of chatter isn't a backpedaling.

He left it out of his updated prediction after benchmarks of a new Intel one leaked. Its called hedging your bets. If the iMac comes out, he was right. If it doesn't, that why he left it out so he was right. Its how horoscopes and mediums work too.

Also, you REALLY don't know your Apple history. The Intel iMac was the first Mac to make the jump to Intel, not the MacBook Pro. If timing is everything, you're showing that there's a lot you don't know...

Both announced January 2006, the MBP just shipped a little later in February.
Things that aren't available to buy immediately, get delayed for all kinds of reasons.


The Intel iMac was the first Mac to make the jump to Intel, not the MacBook Pro. If timing is everything, you're showing that there's a lot you don't know...Also, dude, you're on MacRumors. Find me someone who is 100% right on their Apple predictions or 100% of the time. Also, for the last freakin' time, I never said Ming Chi Kuo was 100% right. I said he's right enough of the time to be worth giving attention to. This site and most news outlets agree. Get over it.[/QUOTE]

And yet you haven't found me anyone else claiming his hit rate is 95%. Which is what you claimed. While treating him as if it was 100%.
If he's not right all the time, why is it such a wound upon your pride that I disagree with him?

Also, you really don't know your Apple history. The 17" and 20" iMacs were first to transition to Intel from PowerPC, not the 15" PowerBook G4/15" MacBook Pro.

That was certainly the idea, but the Intel iMac was delayed. It was supposed to be the first one to get the iSight built in but it wasn't ready so they shipped a G5 version that only lasted 4 months. I feel like I said this once already.

The A12X is not even two years old. The A12Z came out THIS YEAR with the 2020 iPad Pro and it's what was put in the DTK. If timing is everything, you're showing that there's a lot you don't know...

See above for how you are utterly clueless about this. Chips take a while to develop you know. They don't just pop out of the ground on release day. This years A series chips are already in production for iPhones, iPads and these early AS Macs. Hence we can allow a little licence when it comes to the age of a given technology. Especially since we are talking about devices that will be shipping two or more years after the A12X did.

 

[Waragainstsleep]

Seeing the progress of the A Series and the Mollases like "progress" of the Intel chips makes me more then qualified enough.

"I've no idea what you are talking about."

Research intel chiplets. maybe you will.


And bookmark my comment. In 10 years I will be right and you will be wrong about them going back to intel.

10 years is a long time, especially in tech but right now its hard to see Apple going back. Intel will have to cover a lot of ground in order to catch up to where Apple is now and to compel Apple to switch back they will need to get far enough ahead and stay there a while to convince them to start stumping up the extra cash again.

In the meantime, Intel now has two competitors ahead of them and it was already lagging with only one. This new kid on the block though is no plucky underdog with less resources and a weaker brand, its a behemoth that could probably swallow Intel if it wanted to. They can certainly poach talent and likely have been if not directly from Intel and AMD then ahead of them.

Right now Intel's position looks precarious. If Apple is as far ahead as some of us believe they could be in real trouble once Apple shows their hand.

 

[jerwin]

I prefer a lot more an Apple GPU than any other GPU vendor in a Mac.

True. There's a lot to be said about the fact that GPUs peaked in 2012, and 768p looks damn good.

 

[pldelisle]

True. There's a lot to be said about the fact that GPUs peaked in 2012, and 768p looks damn good.

???
Not sure if I understand.

GPUs didn’t peak in 2012. Nvidia makes great improvements every generation. How long will they be able to keep it? No one knows.

 

[jerwin]

???
Not sure if I understand.

GPUs didn’t peak in 2012. Nvidia makes great improvements every generation. How long will they be able to keep it? No one knows.

Indeed. Here I am with a 5k imac from 2014, and I can't run Surviving Mars at 1440p at any detail settings above Medium. My gpu "scores" 24716 on geekbench metal. The gpu that brings this forum to such paroxysms of delight scores 9799.

Obviously I'm long past due for an upgrade. But considering what sort of things show off my current system's limitations. I'd like an upgrade-- not an assurance that my imac will be even thinner than ever before.

 

[matrix07]

I've no idea what you are talking about.

That is your unqualified opinion. Doesn't have any relevance to really, obviously.

I’m perfectly understand what he was talking about. He asked why do you still trust Intel to keep the promise?

 

[thunng8]

Indeed. Here I am with a 5k imac from 2014, and I can't run Surviving Mars at 1440p at any detail settings above Medium. My gpu "scores" 24716 on geekbench metal. The gpu that brings this forum to such paroxysms of delight scores 9799.

Obviously I'm long past due for an upgrade. But considering what sort of things show off my current system's limitations. I'd like an upgrade-- not an assurance that my imac will be even thinner than ever before.

Just a quick note, that Geekbench Metal is how fast the GPU does computations - it does not necessarily equate to graphics performance.

There generally is a correlation, but it is not guaranteed. For example GFXBench would be a better benchmark to test graphics performance rather than Geekbench metal if all that you care about it graphics performance.

A case in point is for the 2018 15" Macbook Pro (AMD Radeon Pro 560X) vs iPad Pro (A12Z)

Gfxbench Metal Aztec High offscreen:

AMD Radeon Pro 560X

GFXBench - Unified cross-platform 3D graphics benchmark database

The first unified cross-platform 3D graphics benchmark database for comparing Android, iOS, Windows 8, Windows Phone 8 and Windows RT capable devices based on graphics processing power.

gfxbench.com

43.4fps

iPad Pro 11inch (A12Z)

GFXBench - Unified cross-platform 3D graphics benchmark database

The first unified cross-platform 3D graphics benchmark database for comparing Android, iOS, Windows 8, Windows Phone 8 and Windows RT capable devices based on graphics processing power.

gfxbench.com

53.9fps

Here, the ipad Pro has higher performance in GFXBench while the 560X has a significantly higher score in Geekbench Metal (~17,500 vs ~10,000)

 

[dogslobber]

I’m perfectly understand what he was talking about. He asked why do you still trust Intel to keep the promise?

I don’t know of Intel’s publicly disclosed schedule to distrust them. What I know is that Intel absolutely will deliver the Lakefield hybrid chips and they are a substantial leap. That’s all I’ll say.

 

[jerwin]

There generally is a correlation, but it is not guaranteed. For example GFXBench would be a better benchmark to test graphics performance rather than Geekbench metal if all that you care about it graphics performance.

Indeed, but GFXbench consistently crashes on launch. Ah well. Some of these benchmarks are very old, akin to bragging that your computer gets 500 fps in Quake III Arena.

 

[matrix07]

I don’t know of Intel’s publicly disclosed schedule to distrust them. What I know is that Intel absolutely will deliver the Lakefield hybrid chips and they are a substantial leap. That’s all I’ll say.

That won’t be enough for Apple ”to go back to Intel” after the transition.
Apple can no more wait for 3rd party to advance their products. Not when their chip department is so good.

 

[dogslobber]

That won’t be enough for Apple ”to go back to Intel” after the transition.

We‘ll see. I think Apple will be hedging their bets until the very last minute of product release. Apple continues to be taking early access to Intel‘s prototype chips as of today.

 

[thunng8]

Indeed, but GFXbench consistently crashes on launch. Ah well. Some of these benchmarks are very old, akin to bragging that your computer gets 500 fps in Quake III Arena.

What are you talking about? GFXBench 5.0 Metal was just recently released. Even the highest end NVIDIA Titan RTX "only" gets 360fps in Aztek High.

 

[jerwin]

Even the highest end NVIDIA Titan RTX "only" gets 360fps in Aztek High.

More challenging benchmarks appear to be necessary.

 

[leman]

The problem with these ARM chips is Intel. They’re planning to release hybrid x86 chips which are similar to what Apple is claiming will be their strength. I can see Apple going back to Intel in the future for some machines.

I suppose you are referring to the asymmetric performance+efficiency core configurations. That will help Intel with battery life for sure.

Intel will have to work very hard to match Apples performance though. Currently Apple has a 50% IPC advantage. That’s not something you can overcome over night. Intel needs a radically new architecture if they want to match that.

 

[Waragainstsleep]

I don’t know of Intel’s publicly disclosed schedule to distrust them. What I know is that Intel absolutely will deliver the Lakefield hybrid chips and they are a substantial leap. That’s all I’ll say.

They appear to be a single high speed core and four efficiency cores. You know Apple already has 4+4 right? Plus whatever they haven't told us about yet.