Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
We read it. They both have poor single-core performance. That keystore test is also very specialized, so it's not useful look at when considering Mac workload, while Geekbench is. The article you linked doesn't say anything about single-core performance either, but I think the "medium" test is that, based on the AWS tiers documentation.

111160.png


Single Thread *Integer* Performance at 30% lower clock speed against Cascade Lake. I dont even know how anyone define "poor" Single Core performance and not "optimised" for Single Core performance, or Xeon Phi like Performance.
 
Last edited:
All of this seems to hinge on the presupposition that Apple is unhappy with Mac sales volume, and… I'm not sure they are?
I think they’re content with Mac sales volume for how much profit they make. There will eventually come a time, though, when they won’t consider it worth the effort.
But, considering how persistent this rumor has been,
I think part of the insistence of recent rumors is something that had not been a part of prior rumors... a report allegedly from Intel indicating that Apple will be moving to another processor architecture. I’ve always taken that to mean that someone from inside Intel has seen that Apple’s purchase order is far smaller than it had been historically, especially considering Apple’s 20 million per year rate of Mac sales and assumed that:
a. Either Apple will be selling FAR fewer Macs than they had in prior years, or
b. They’ll still sell Macs, but they won’t have Intel inside
There’s stuff in the App Store that won’t even run on Catalina, how are you supposing Apple is going to recompile all those existing apps for a different architecture?
There’s no Electron apps in the Mac App Store that use private API’s. This is a decision Apple made and, whether the devs like it or not, they have to abide by it if they want to keep their apps listed/available to their customers.

And, Apple wouldn’t need to recompile everything. About Bitcode, Apple said developers could “take advantage of new processor capabilities we might be adding in the future, without you re-submitting to the store.” That’s from 2015, so that would mean Apple’s been working on the potential pitfalls for at least 5 years, likely far longer than that.
 
I don’t think that’s correct. As long as they didn’t use any low level coding, they might get away with some minor adjustments and a recompile targeting ARM64 - as will most developers.

This is an architecture switch, not an OS switch with entirely different APIs.

But does it guarantee 100% performance while it can take advantage from ARM?
 
...a USB charger that tells you how much power is being drawn at any given time):

794C5DC4-E734-4975-AE18-87970FA6F3FD.png


Isn't the USB charger limited to output 3A x 5V = 15W?

So that 14.96W number on the chart above is the same as the 3.6 roentgens/hour radiation measument on the Chernobyl HBO's miniseries...😉
 
Last edited:
They have to make a whole new software just for ARM-based devices. But they already made Photoshop for iPad.

But will they is the $64K question.

I bet they don't. And not just because they are Adobe.

Hardware version 1.0 bugs plus Software version 1.0 bugs equals hard pass for people that actually do work.

I am sure it will sell to the Candy Crush crowd, but for those of us that have work to do, probably not.
 
Last edited:
And, Apple wouldn’t need to recompile everything. About Bitcode, Apple said developers could “take advantage of new processor capabilities we might be adding in the future, without you re-submitting to the store.” That’s from 2015, so that would mean Apple’s been working on the potential pitfalls for at least 5 years, likely far longer than that.

Even if Bitcode can be used to transpile from LLVM-IR-x86 to LLVM-IR-ARM, which so far nobody has stated unequivocally it can, that leaves the problem that almost all Mac apps, including those on the Mac App Store, do not use Bitcode at all.

So that's a non-starter.
 
Office for iOS has some feature limitations not because of inherent issues with the ARM architecture (though there might be a feature or two they simply haven't ported yet), but because of user interface paradigm differences between iOS and macOS. There is no reason to believe that a hypothetical Office for macOS on ARM would be particularly limited.

I understand that; office on Microsoft ARM (Surface Pro X) is significantly better than current Apple ARM (IOS / iPadOS).

I have no doubt Office will be fine on macOS ARM also. Today it is next to pointless on iOS etc.
[automerge]1585430810[/automerge]
The cost on ARM-based Macs is a lot less than x86-Macs and it can really help double the annual sales of mac if the overall price is lower.
Some of Apples ARM-based iPhones cost more than x86-Macs
 
Last edited:
  • Like
Reactions: falainber
I ran benchmarks (identical source code compiled by me, so no cheating) where an iPhone XR (two fast + four slow cores) consistently beat an iMac (Quad Core x86) when using all threads. There was a beat of slowdown, about 10%, after ten minutes due to thermals, but the iPhone XR was still significantly faster.

And the processor in an iPhone XR is designed to run in a tiny device with no cooling. An ARM processor inside an MacBook Pro case could easily have four fast cores (and a few slow ones) running at higher clock rate.
Yeah, what I said was old news. Apple made a big leap recently, and I didn't understand Geekbench's site, but turns out they show iPhone CPUs pulling ahead of mid-range desktop ones in some ways.
 
Isn't the USB charger limited to output 3A x 5V = 15W?

So that 14.96W number on the chart above is the same as the 3.6 roentgens/hour radiation measument on the Chernobyl HBO's miniseries...😉
I thought about that possibility too, but I checked and the standard charger that comes with the 2018 iPad Pro is 18W.
I agree that those numbers look too good to be true, but it is what it is. The A12X is a little marvel of a processor.
 
  • Like
Reactions: Roode
...a USB charger that tells you how much power is being drawn at any given time):

I thought about that possibility too, but I checked and the standard charger that comes with the 2018 iPad Pro is 18W.

Can't be the iPad's standard charger, since it doesn't tell the power being drawn.

He probably used a 15W charger like this one I have here (it tells the current being drawn):

424AD9C2-4CA3-40B5-8EB1-AEDAC42265D8.jpeg
 
Last edited:
Can't be the iPad's standard charger, since it doesn't tell the power being drawn.

He probably used a 15W charger like this one I have here (it tells the current being drawn):

View attachment 902035
You kind of misquoted me. I said "Let's review together the data they gathered measuring the actual power consumption (you can do this if you have a USB charger that tells you how much power is being drawn at any given time)".

Meaning you could do it like that, but they actually used something else. The review mentions using a MetraHit Energy, which lets you measure up to 16A of current by plugging the standard iPad Pro charger into the PMA16 adapter.
 
I look forward to this. Maybe this will finally push me off of my 2009 MacBook as a daily driver (although probably not ;)).

They should bring back the iBook/PowerBook/Power Mac lines because they're switching back to a RISC architecutre (probably unlikely, but you never know). It would be a nice nod to the past for sure.
 
You kind of misquoted me. I said "Let's review together the data they gathered measuring the actual power consumption (you can do this if you have a USB charger that tells you how much power is being drawn at any given time)".

Meaning you could do it like that, but they actually used something else. The review mentions using a MetraHit Energy, which lets you measure up to 16A of current by plugging the standard iPad Pro charger into the PMA16 adapter.
👍
 
Given that there isn't a single ARM-based Mac for sale,

Add another USB port or 2 to the newest iPad Pro, plus the new trackpad with keyboard case, and the only big differences between it being an entry-level MacBook and an iPad are the upper layers of the OS and utilities that Apple allows to be installed. However, the larger iPad Pro + keyboard currently costs more than the entry level MacBook Air. As an advantage the iPad is faster, has a touchscreen, and runs fan-less.
 
almost all Mac apps, including those on the Mac App Store, do not use Bitcode at all.

So that's a non-starter.
Well, that’s just one “App’s must be delivered to the App Store with bitcode turned on” requirement away. :) If you can’t get an update into the store without that turned on, you’ll turn it on or be removed.

Steve Troughton-Smith has a blog post from May of last year where he extracts bitcode from an ARM app and compiles it for x86, so someone has shown that it’s possible.
 
Wrong. The idea of using ARM-Based on some of the macs is to provide better performance without the need for active cooling to prevent overheating. Apple does not necessarily need to stop selling x86-Macs when that happens.
Starting posts with “Wrong.“ makes you sound arrogant and unnecessarily argumentative. I know tone doesn’t always come across as intended online and across cultures so I’m just letting you know.

The idea of using ARM-Based on some of the macs is to provide better performance without the need for active cooling to prevent overheating.
You know this for a fact? I wasn’t aware it had been revealed that their ARM processors would be passively cooled.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.