Apple M4 benchmarks suggest it is the new single-core performance champ
tomshardware.com
tomshardware.com
But the absolute biggest game changer was that it was silent. Completely silent. Hadn’t realised how noisy computers had been until then.
The only hard failure I've hit (admittedly with a small sample size) was The Talos Principle, which has an intel Mac version that won't run on the M2, and running through Whisky immediately errored as well.
Both Unity and Unreal support metal well (and for a long time, better than Vulkan), especially due to iOS. Even many proprietary engines do.
Even before Vulkan and Metal existed, OpenGL was consistent (before Apple stopped supporting it) and there weren’t many more native games then either.
Someone will no doubt bring up reusing Proton, but they’ll gloss over the differences in platform and also that Game Porting Toolkit exists and does the same thing.
The reality is that, until recently, the number of Mac systems that were gaming ready was a very small number that wasn’t worth focusing on.
I also had a Ryzen 3700X, high-end CPU and case fans to make it somewhat quiet under load. Then I got the MacBook Air M1, which was as fast in multi-core use, but felt much faster in daily use to much better single thread performance (turns out that large builds spend quite some time on portions where everything blocks on one unit and the amount of parallelism is 1). And it was completely quiet (no fan).
The Air M1 felt like magic at the time, I could have the power of the 3700X on the train and insanely long battery life.
[1] That said, I got an M3 Pro for work and my private laptop has an M1 Pro and the difference in single-thread performance is certainly noticeable.
I wouldn't call my ThinkPad P14s AMD gen whatever loud either. The fan will spin up when multiple cores are under heavy load but compared to the 2018 MBP it replaced it's essentially silent.
The iPads have had the hardware in the M-series chips and the software in the form of Apple's hypervisor framework in iPadOS for a couple of generations now, but Apple hasn't enabled it to be used officially.
I really wish they would just allow this on iPadOS. It still maintains the sandbox model Apple wants for iOS, it would just give a (contained) outlet for doing things that are difficult in native iPadOS.
In any case, Apple still wants to "review" apps, and we want (arbitrary) user code execution on device. That's something Apple strictly forbids on iOS/iPadOS AFAIK (which is why we can't even have Firefox addons). Unless we can have at least true side loading, a DMA extension to iPad won't help.
The DMA isn't really the right tool to liberate devices, since it's about market competition not consumer rights. I think it would be better to widely address this along right-to-repair, electronic waste reduction and consumer rights regarding actual ownership. Unconditionally locked hardware is ridiculous.
I wish they would simply unlock the bootloader, so we can have Asahi Linux for iPad. They don't have to do anything else. Although Asahi is on trajectory to exceed MacOS performance and dev usability, I don't think they would lose their existing appstore cattle to Linux, but rather gain new hardware only customers.
Though, I don't see true sideloading (like on Android) specified. If apps still need Apple's approval, we will get "freedom" who to pay, not what to run. I still don't see device liberation within the scope of the DMA.
However, if we're lucky, Apple may decide the app approval process may not be worth it, if they are not allowed to extort developers anymore, so they may allow unsupervised sideloading as a consequence.
In any case, requirements and reality may take much longer to align than 6 more months, considering Apple's cringeworthy tantrums so far...
But if Apple abuses its position as platform gatekeeper with the app approval process and rules, the EU will probably slap them down. The DMA doesn't care about device liberation, but it does care about fairness, so Apple will probably only be allowed to continue this if they act in very good faith towards 3rd parties, which doesn't seem like an Apple thing to do.
They removed the hypervisor framework in addition to the kernel support for virtualization a few months ago unfortunately.
Plug it into a usb-c dock connected to a screen/keyboard/mouse and it's a Mac, put it on the little stand with the magic keyboard and it's a MacBook, hold it in your hand and it's an iPad
My dream device
Apple is very clear about their product differentiation and would never make macOS and ipadOS exact copies. Case in point: when they brought cursor support over, they painstakingly engineered it to work differently and snap to objects on the screen. I think if they did OP's suggestion and had a macOS screen show up when an iPad is connected to peripherals it would actually work out well.
Switching between entirely different and separate "UI personalities" could work.
I don't think the pointer vs touch UI is the main thing that people have in mind when wishing for macOS on the iPad though, instead a "proper" UNIX-style filesystem and shell, and the ability to install any software outside the app store via a package manager and without Apple's nanny reflex getting in the way.
It goes back further than that too. Microsoft bought the Sidekick and squandered their lead by bringing it onto the Microsoft platform.
I'm reminded of this quote: "A Foolish Consistency is the Hobgoblin of Little Minds."
Laptops and tablets are different devices. We should stop trying to merge them. Every time we do, we end up with something that is a worse version of both things. Tablets need a battery. If the display is detachable, then that creates a weight problem for the laptop. The tablet keyboard is worse than any laptop keyboard.
I said over a decade ago I thought Apple was smart to just have a tablet OS as well as OSX. Don't spend 3 years trying to merge those things. It's a waste of time and gains you nothing. Even the tablet and phone OSs are somewhat distinct (but a lot less so than with OSX).
I always thought Eric Schmidt (at the time) did the right thing with ChromeOS and Android too. There were always questions from people seeking faux "consistency" like "why have 2 OSs? Shouldn't we merge them?" Again, phones and laptops are different things. Let each OS evolve and see if one emerges as a "winner". Otherwise, leave them alone.
And then there were the checkboxes which allowed you to extend the OS beyond Chrome OS's initial limitations. Enable Android. Enable Linux. Enable developer mode. But the user (or administrator) could ignore those checkboxes and keep the machine in a fairly locked down state.
Keep imagining a similar abilities on iPads. A seamless transition from desktop mode to touch and back again. With options allowing you to make it into more of a general purpose computer if you want. But the options can be ignored in favor of the walled garden, if that's what the user or school or corporate owners want.
Even if they made every single one of them buy a new ipad every single year that would still be single digit change of revenue for apple
https://www.bloomberg.com/news/newsletters/2024-05-05/apple-...
Wouldn’t there actually be some popular Linux tablets out there? (I am aware that there are a couple of niche options)
Touch screen interfaces are just not good for programming. Apple already brought all of the useful parts of the iPad to the Mac when they switched to the m series SOCs. Fanless design, long battery life, instant sleep/wake.
An iPad running macOS has some niche appeal for people who want to travel light but I really don’t see it being a game changer at all.
Not necessarily. The surface line has several hardware fumbles (especially regarding power budgets/efficiency). The A & M series chips could easily whip up most of their competition in the low power (>10 watt) segment, if Apple wanted. AMD and Intel push for high performance at 15-28watts on portables, which is too high for thin tablets.
For maximum contrast: x86 tablets have fans. M1/m2 laptops can be fanless.
(This isn’t to say it’s impossible but rather no company with deep enough pockets cares enough.)
Also the Surface Pro 7 was fanless with x86 processors, plus the small one (surface go, I think) is fanless and x86 but does have performance issues.
It was much lighter than what I have now, a 2-in-1 with a fold-back keyboard. This opened up possibilities like using a car suction cup mount and a lap desk to get a slightly taller computer while on the couch. Or a lightweight armature.
Plus, vertical orientation on a “laptop” felt really novel and nice. My 2-in-1 can be vertical, but it is clearly an afterthought. The iPad ~4:3 aspect ratio is much nicer for vertical use, and there’s something about the pixel alignment or maybe the screen viewing angles… my laptop screen doesn’t work quite as well sideways.
I switched because I missed i3wm mostly, and generally all of the local Linux software. But no complaints about the hardware.
M-series is so far ahead of the game than an air-cooled M4 has better single-core performance than the best desktop PC chip money can buy. Incredible.
I get that Apple has an advantage over companies like Dell and Lenovo in that they also build the OS but then there are Linux computer companies that have the source for everything and their machines are usually even worse from a performance / watt perspective.
Here Jim Keller explain why CPU architecture doesn't matter a lot in performance now: https://www.anandtech.com/show/16762/an-anandtech-interview-...
But then I'm back to my original question: where are the Linux or Windows machines with great battery life and good performance. Is there anything like what Apple sells at Walmart: a 13.3" MacBook Air with the M1 chip, retina display, and 8 GB of RAM for $700. This is a very old machine at this point and still very hard to beat.
Sometimes it feels like Apple is the only company even trying. It sucks.
Regardless, it certainly feels like an own goal that AMD or Intel hasn't yet released a CPU with at least 256 bit wide on package RAM. They'd be able to capture a higher percentage of the BOM cost of the laptop if they did so on top of the speed benefits.
I don't really track this directly, but periodically I look for passively cooled (fanless) laptops. I haven't looked into Ryzen, but it would be a very nice surprise if there are fanless Ryzen laptops out there.
If you tell me TDP is a lie, I’ll believe you.
However, that doesn’t change the fact that there still aren’t any great fanless laptops from anybody other than Apple. I was attributing it to TDP, but maybe it’s something else.
Uh, but M2 on TSMC's N5 node is faster and more efficient than AMD's products fabbed with TSMC's N4 node. Edit: more efficient in general, but I meant "faster" in comparison with AMD's latest laptop chips.
> Here Jim Keller explain why CPU architecture doesn't matter a lot in performance now: https://www.anandtech.com/show/16762/an-anandtech-interview-...
Let's see what he has to say:
> But instruction sets only matter a little bit - you can lose 10%, or 20%, [of performance] because you're missing instructions.
20% is quite a lot in the current competitive landscape. Jim Keller is stuck thinking like it's 10+ years ago.
The 7940HS is way faster than the M2. Efficiency is harder to compare since TDP doesn't mean anything theses days.
because even in cinebench the m3 max is already ahead in cinebench 2024 let alone R23...
https://www.notebookcheck.net/R9-7940HS-vs-Apple-M3-Max-16-C...
The M3 max is on TSMC N3B, which have a +30% of transistor density compared to the TSMC nodes AMD is using (N4).
Does Apple really sell more iPad Pros than MacBooks to make it worthwhile? I was under the impression that the iPad Pro is a somewhat niche product and the Air is far more popular.
I think this is one of those things that only enthusiasts care about, and if you care that much about performance you’re probably one of the people who buys a new one every year or two anyways so they aren’t really missing out on your revenue.
Most people just need or want a new laptop and order whatever one is currently available.
People who buy MacBooks aren't suddenly going to stop or switch to Windows because of some Osborne effect. The entry-level M3 MBA is still extremely hard to beat at the same price point.
Like how long? M4 MacBook Pros could in theory launch in a month (i.e. WWDC). That's worthwhile.
Most believed that M4 was not coming either until a last minute rumor. So who knows.
Related rumors: - M3 process is expensive so it's getting dropped ASAP - M4 needed for AI so the whole line is getting refreshed - There's competition from Nuvia etc and M4 is needed to get ahead
I think it's more likely that current owners of m3 macbooks will be enticed to buy m4 ipads and thus apple makes more money.
And owners of older macbooks would buy both the m3 macbook and the m4 ipad
It sounds backwards, but that's how the apple ecosystem works.
- first access to the best process
- a great team
- high bandwidth / latency DRAM
Intel & AMD aren't that far behind on #1 and #2, but I'm surprised at how long it's taking them to catch up on #3. If they created and sold a chip with wide low latency DRAM on package they would simultaneously significantly increase their performance and also they would capture a higher proportion of the BOM cost of a laptop.
Pedantic: AMD's MI300X has on package RAM, but it's not exactly a laptop chip...
1: Jim Keller https://www.anandtech.com/show/16762/an-anandtech-interview-...
Even without the M4 it would be awkward when those chips appear in the real World in real situations.
IMO some people inside Qualcomm have been stretching the truth way too much even internally.
I know us nerds aren’t the market, but an mATX motherboard with elite x for like $700-800 would be a seller.
Before you ask, I'm one user who has a Razer Blade 16 (4090) and Galaxy Book 4 Ultra that I use for CAD and BIM work. I'll pass on having a dedicated video editor.
The main issue with M series is that a much of performance comes from the memory and that is very expensive to scale.
That's is why Apple is still selling 8GB laptops.
Punching within its own weight class, laptop chips rather than expecting it to outright beat desktop chips... it beat the 5980HS and 11980HK in every single test.
It just didn't sink into the public consciousness too well at that point because everyone assumed it must have been some erroneous test/outlier.
https://www.anandtech.com/show/17024/apple-m1-max-performanc...
https://www.anandtech.com/show/17024/apple-m1-max-performanc...
While M1 Max can be used in laptops, the benchmarks I've seen have been for the mini/imac where it doesn't have power and thermal limitations of a laptop.
Besides, the cheapest M1 Max laptop is $1000 more expensive than the most expensive 5980HS laptop :(
Would love to see some blade or sled hardware based on Apple Silicon, think of the density of cpu/gpu/ram/npu :)
I wonder how many people are going to get Apple-shaped burns and will get told that they're holding it wrong.
Since they got rid of the stickers, they should include Apple-shaped burn pads.
On very intense workloads, the fanless products do eventually throttle slightly. It's after several minutes, not a fraction of a second.
If you have reason to believe this is only true for M1-3, but not M4, let's hear them.