The Surface Laptop scored 1,745 on the Procyon AI Score, while the MacBook Air managed 889. Qualcomm’s Snapdragon X Elite has 45 TOPS of AI acceleration performance, much more than the 18 TOPS found on the M3.
TOPS (Trillion Operations Per Second) is a meaningless score without including the precision. Are they INT4, INT8, INT16 or FP16? Microsoft's qualifications for a Copilot+ laptop require 40 TOPS at INT8: https://www.theregister.com/2024/05/21/qualcomm_windows_micr.... The new Snapdragon X Elite can do 45 TOPS at INT8.
I believe there's some work done by Qualcomm developers on Linux, and they even have some marketing site about mainlining Linux support here[1].
Linux 6.10 has some drivers for hardware on the Snapdragon X Elite platform[2].
That said, Linux support does not come down to just SoC support. Each device using the Snapdragon platform will need support from manufacturers or volunteers to get Linux running on them, as they'll need DeviceTrees written for them, drivers for peripherals and whatnot.
It would be nice to be able to purchase a Linux laptop with proper power management. I haven’t used a Linux laptop in a long time, but when I did, I found it didn’t manage different power modes well at all. I feel that’s the main drawback to a Linux laptop, so if arm chips fixed that it would be awesome.
This is one of the main reasons I use Pop! on my Thinkpad. Their power management works great out of the box, I get three times the battery life compared to Win 10. Suspend often doesn't work properly, but to me, that's a minor issue, likely hw related.
My Latitude 7370 is just about perfect. With gnome power profiles, I can easily switch between power saver, balanced, and performance, suspend/resume works fine. All peripherals work out of the box. It has a crisp 3200x1800 touchscreen that works great under Gnome. It is pretty much my perfect Linux laptop. I only wish the battery life was better, and that it had a more powerful CPU. It is fanless, so it throttles pretty quickly under load.
I ran some tests, and completely idle at 30% brightness I can get 7h under Linux and 8h under Windows, so Linux still has some room for improvement. But as someone who uses Windows at work, there is no way I could go back to that personally. Not unless they released an open source version completely cleaned of crapware and bloat. The Enterprise / IOT versions are the closest I’ve come to that.
I have both a Latitude 7370 for personal use and a Precision 3520 for work. Lugging that Precision around is a pain in the butt. It is big and bulky, noisy and hot. That said, as my eyes continue to age, the 13” screen of my Latitude does seem smaller than it once was, though the crispness of the 3200x1800 is much better than the grainy 1080p in the Precision. And I do miss the number pad sometimes. My perfect form factor these days would probably be a 14-15” in a similar profile as my Latitude (it is essentially a poor man’s XPS 13).
IMO for many people a laptop means highly portable, wire free 'desktop' computing experience that you can't get on a tablet. Therefore the comparison for mobility is with a tablet and not a powerful Windows laptop.
Plus many people working in tech have specific systems for specific tasks; I use a spec'ed up Windows laptop for heavy lifting and games, and an M1 Air as my main computer.
But at the same time, many people also asks for performance improvements, and sometimes extra few USB A ports too. Plenty people makes hyperbolic claims such as which Apple Mx CPUs has x times more computation than which NVIDIA GPUs, superficially focusing on raw CPU performances.
I know deep down that thinnest laptops with brightest displays sells the most in reality, marketing wise, but it just sounds a bit hypocritical to me.
For ultraportables I think people just want performance to be “good enough” while also not getting hot or spinning up a fan. The M1 Air struck this balance very well – it was no speed demon but wasn’t pokey either, and battery life didn’t suffer for it.
This was in stark contrast to contemporaries where any semblance of decent battery life required low power mode, which made them dog slow. Even today this is something that x86 laptops with current chips struggle with… there are a handful of x86 ultraportables that can get M1 MBA like battery life, but they have larger batteries than the MBA had and still need low power mode to pull it off reliably.
That’s where a lot of the hype around “M-series SoC is capable of X performance” is rooted. It’s not the performance in and of itself, but the perf per watt, which allows that performance without also turning the laptop into a furnace/jet or destroying battery life.
Because the point of a laptop is to be portable? And a large chunk of that is being small and light? I could haul my full desktop and monitors around with me, but it's not worth the tradeoff. If I can get a computer that does 70% of what my desktop can do at a tenth of the weight and size, that's worth the money.
Super critical for those that travel a lot, whether for work or personal needs. I've lugged some boat anchors around; having extra space for the necessities, or just being able to work in economy class on a long flight, are premiums I'm willing to pay for.
Is ignoring M4's existence really 'wrong' given the fact that it isn't in macbooks yet? Are the benchmarks they mention possible on the iPad? Do the iPads benchmark similarly to the macbooks with the same chip?
Does Apple also have to somehow properly benchmark against the not-yet-available highest end models of their competitor's chips for their numbers to not be wrong?
It looks like it hasn't even been a month since the M4 was publicly announced, I think it's fair that that's just too soon for the benchmarks to make it into these early comparisons.
> Is ignoring M4's existence really 'wrong' given the fact that it isn't in macbooks yet?
I don't think so, but it is a risky position to take. Presumably plans to publicly lean on that benchmark were in flight before the M4 was announced, so they went ahead with it anyway.
I think it's risky because: assuming the M4 delivers the performance we'd expect from another Apple Silicon evolution - and releases in non-mobile devices this year - your marketing headline effectively becomes "hey, our best is only better than our competitor's last generation [and our competitors are already better]"
That's a risk they all take. AMD and NVIDIA (or Intel) have the same thing happen all the time, one of them has to go first, and usually there's not much of a gap between their announcements, and thus each risks being overshadowed by the other or sometimes even forcing them to cut prices or cancel a product as a result because the first one's announcement happens to make it completely pointless.
The difficulty here is Microsoft is staking a fair bit on this. To some extent, their ability to make ARM their future is dependent on this push succeeding in getting people to swap to it.
AMD/Intel can always just lose gracefully and go again next year. For Microsoft trying to decide how much they're investing in ARM, going all in on "we're better than the Macbook Air" when there's a good chance Apple is going to blow them out of the water in a few weeks is a much bigger gamble. It's possible Apple's announcements aren't that impressive in some dimension, but even if it's in the same range, it could really dampen Microsoft's push here.
I think it's the opposite, something must have been wrong with M4 and Apple managed to spin exactly "our best is better than competitor's last generation" around the reality by releasing iPad early. We will probably see it with Apple either doing M4X for MacBooks, or not.
I had a similar thought, maybe they should have compared it to MBP, but at the end user is looking at how they can spend their money in the best way possible. For $999 you can choose between MBA 8-Core/8GB/256Gb or Surface 10-Cores/16GB/256GB. It seems like Microsoft does have a better deal here. It is similar weight to MBA, similar size, better battery life, and seems like more performant.
But there are numerous issues, that might be discovered later:
1. Similarly to the MBP because Surface will be more powerful than MBA, after you install all the default apps you need, and various Electron apps (like Slack) and updater like Microsoft Office, Google Chrome, will keep running in the background, you will never see 20 hours on the battery. This is the reason why I like MBA comparing to my MBP M2 Max, just because this chip cannot drain battery that fast.
2. Microsoft did a smart thing, announcing those laptops a few weeks before WWDC, but planning to ship them only after WWDC. So apple cannot play back and release M4 MBA on WWDC and show how much faster it can be comparing to Surface. Just because there is nothing to compare.
3. The quality of the Surface devices seems like it was always questionable.
Anyway, I am really excited about ARM powered Windows laptops, for multiple reasons. At the end I hope to see Bootcamp again, maybe games will be optimized for ARMs, and we will be able to play games on MBP with Bootcamp :D
Lifetime Apple user since before there was even a Mac. I just hope these are great pieces of hardware, whether in one area or another Macs or these perform better. Good quality, affordable hardware available for any OS platform would be great for users.
As a long-time Mac user, I'm actually eager to switch to Windows: more hardware innovation happening with more options in that ecosystem (e.g. in dual screen laptops like Asus Zenbook Duo). Windows has a few rough edges but my usage is dominated by the browser and terminal, which are similar on both Mac and Windows. Great to see these ARM-on-TSMC benefits of performance and battery life land in the Windows ecosystem.
Are there benchmarks that test usability? I don't care how long it takes to encode a 4k video. I care about how long the UI hangs while doing basic tasks or how many work-arounds I have to look up on Reddit to get my Bluetooth headphones pairing reliably. Looking at you, Windows...
It doesn't matter, Windows is insufferable. Like a giant advertisement pretending to be an operating system. They need to start again from scratch or just simply give up and ship Ubuntu
Windows, since 7, appears to suffer from a "law of conservation of awesomeness": any improvement in one area must be counterbalanced with at least a commensurate drawback in some other area.
When your advertisement is all about your enemy which does not, unlike old intel version, cannot even run yours and hence is totally different market sort of … one should be worried about you.
How many workloads are running in a compiled language though? I'm at a Java shop right now, and I realized I can save a ton on node costs by just running those existing workloads on ARM.
The less thing I want to use is MacOS[0], the only real alternative for me is a Linux flavor, which may be the case soon enough if keeps trying to pushing Copilot and other AI crap down my throat.
That's... a 4 year old writeup that's mostly complaining about things that have since been fixed and were generally considered flaws at the time even by people who liked the OS/Hardware, and some of them are specific flaws to a specific model of laptop at the time.
You're free to not like it, but if you're still holding on to that list of reasons, it's a bit like abandoning Windows because of the W10 rebooting issues and disliking the original Surface. There's been a lot of change and there's other options.
You can't just dismiss it as being invalid because it is 4 years old. I counted 9 of the 15 items as still present issues on current macbooks. Most of the things not solved are design decisions. It's not like people needed more evidence that apple is slow to adopt, or never does adopt, quality of life improvements that are present in their competition.
"It’s also important to note that, unlike Apple’s MacBook Air, Microsoft’s new Surface Laptop isn’t fan-less, allowing it to squeeze more performance out."
And there ladies and gentlemen is the catch... What a totally laughable and unfair comparison. Wake me up when it's fanless
There's nothing unfair about fans. They're not new technology and apple is allowed to use them too. And they do, in a way that is barely noticeable. If you do it right it's fine.
Apple DOES use them and Microsoft chose not to compare their product to the MacBook Pro clearly because it wouldn't have benefited them to do so
"There's nothing wrong with fans" -- sure nothing wrong with them as long as you don't mind the extra noise, energy usage, point of failure and an entry point for dust and other crap into your laptop.
Give me a break! Passive cooling is the future for personal computing.
But as you point out, apple does have fans and in the m-series laptops they're done in such a way that they're nearly silent, or off. So, they really do not contribute meaningfully to noise, energy, or failure. They help with the burst workloads that a PC may encounter. Designing a PC to require the fan always be on is a sign of inefficient design, but designing a PC to specifically avoid a fan is to just let your competition beat you.
TOPS (Trillion Operations Per Second) is a meaningless score without including the precision. Are they INT4, INT8, INT16 or FP16? Microsoft's qualifications for a Copilot+ laptop require 40 TOPS at INT8: https://www.theregister.com/2024/05/21/qualcomm_windows_micr.... The new Snapdragon X Elite can do 45 TOPS at INT8.
Apple's M3 has 18 TOPS, and their M4 has 38 TOPS, but the M3 was measured at INT16 and the M4 at INT8: https://www.anandtech.com/show/21387/apple-announces-m4-soc-.... Cutting the precision in half lets you do ~twice the amount of work.
The Snapdragon is very impressive, but it's silly to see TOPS claims from companies repeated by journalists without adding some context.