Good question... personally I've been using hp laser printers over network for years and they work with everything I've used. Windows, Mac, Linux, Android phones and tablets without issue.
Linux 6.0 and higher supposedly support the device. The system boots using EFI so as long as you can get a compatible kernel and the right firmware, it should work.
Yeah, tough. I will wait to see if somebody more explicitly Linux-friendly makes a good workstation class ARM (or better, RISC-V even) laptop, because I've had nothing but problems with my Thinkpad Z13 Gen 1 Radeon/Ryzen and Lenovo is in my bad books.
I had a M2 MBP issued for work by my previous job and it really was excellent; performance and battery life. But I'd have a hard time giving up Linux, and I'd worry about installation of Linux on hardware like that always being second-class.
For me, don't mind mac so much as do wish I could stick to the linux/windows keyboard bindings I'm used to. Rosetta2 and docker have come a long way since i got my m1 as well.
There's two groups wanting Linux: One wants access to Linux tooling and will almost certainly have a good experience with WSL/2. The other group wants to not run Windows, so WSL in any form is mostly useless.
> Other tools, such as MySQL, node, git, etc. have no issue when running under emulation, at least in the context of development
I'm surprised the author is using these under emulation. Surely they'd run at completely native performance if you just rub the through WSL2? Linux distributions have had ARM based distributions for ages.
It may be because he couldn't get IntelliJ to work native in a way that let him access WSL, ("It also wasn't able to access WSL, meaning I was locked out of using Linux tooling all together.") So that tooling was awkward.
If he was stuck using windows, maybe best bet would have been to run an X server and run IntelliJ etc inside WSL itself.
Hi, author here, you are correct! WSL tooling was locked out for a long time due to IntelliJ not being able to access WSL until the native build came down. Unfortunately, WSL on ARM still has some bugs around the VM hard locking itself when waking from sleep. It's frustrating, because when it does work it's fantastic.
Have you tried a standard VM? It sucks that WSL2 isn't working right yet, but IntelliJ has some pretty extensive tooling for working over a (virtual) network. It's possible that the sleep issue plagues normal VMs as well, but maybe it's worth a shot?
You'd lose out on things like hardware accelerated graphics but reading your description it sounds like that's not really something you rely on anyway.
I do run ubuntu 22.04 in a VM, it works fine for things that aren't graphical and it has the usual performance penalties that come with connecting via a RDP connection, but it's a nice alternative when I need it. I haven't tried IntelliJ via virtual network to the VM though, that's a novel idea, I'll give it a go!
I'd be much more interested in the Linux situation on this laptop. Most of the UI applications I use casually or CLI apps are available on ARM (I have them running fine on a Pinephone, so the experience on a laptop will only be better).
But not sure how the dev workflow would be: Podman, IntelliJ, Sublime Text, etc.. And some casual stuff like Calibre, Thunderbird, Joplin.
Oh and how much would the battery life improvement be. I already get 4~5+ hours on my 5 year old Thinkpad, so an improvement to 6 hours is not really worth it.
How is the BIOS and openness on these systems? I always heard the arm laptops are way more locked down than the x86 ones.
Same where wondering if it performs the same/better under Linux. I imagine the battery life could be better under Linux based on my experience with Windows 11 on my own laptop, I mean, the fans don't even spin up that often under Linux unlike the near constant fan chorus Windows gives me.
Also, a computer without a single fan so I can comfortably use it on my bed? With possibility to replace it's SSD? It sounded a lot like my next laptop.
...yeah, I was wondering that too. This seems like a job for a Gentoo system, compiled from the ground up with "-march=native". Would love to know how that would turn out performance-wise, in comparison to Windows.
There's also KISS Linux, which I find quite approachable compared to Gentoo. I know, sounds totally backwards, but I think it's really cool how simple it is to manage your repositories and make your own packages. And yes, you recompile the whole system with `-march=native` during the install process :)
What does your `top` say? What does your browser's performance tool (e.g. `about:performance`) say?
I went from a constantly-overheated Linux laptop to a completely calm Linux laptop, on the same hardware, distro, and DE. Checking for unnecessary processes that eat CPU, spurious ACPI events, overly-heavyweight browser tabs goes a long way. Making sure the browser uses hardware support to play YouTube is another heavyweight.
It takes some time, but much less time than I had expected.
So the Mac has battery saving built in to the OS. You are doing a lot of reverse engineering to achieve something other software/hardware does automatically.
And to be specific, I suffer from bad hardware/software communication. On a linux laptop, no less (System76, which I would never recommend). Linux S3 sleep is not hardware sleep, I lose a lot of battery there. WiFi and Bluetooth connectivity either has to be battery draining or have poor performance.
The battery my laptop has itself is also too small, and could only get moderately good life if it had an optimized OS anyway. Again, a Linux-first laptop.
I am sure if I switched to a Lenovo with a big battery that I would be better off than I am now, so I'm also complaining a bit about my hardware. But the problem is still there.
With Apple M2 paired with an OS designed for it, with high end batteries and screens, I don't see any pairing of (laptop) hardware and software competing, on a fundamental level. Every other system is second class.
Edit: ACPI events is the key word here, and it is not worth anyone's time to wade through that garbage to get sane laptop performance. The ACPI stuff isn't even designed for end users to alter, and it is hacked together due to all the hardware variations out there. It is simply doing the best job it can, on average.
There's kind of an implicit lie when people say linux works on laptops at all. It works on MOST laptop configurations in MOST functions. But it BARELY works in others. You find this once you dig sufficiently deep enough.
I run the OS of my choice on hardware of my choice in a way of my choice (that is, not Debian, Ubuntu, or Fedora). The small amount of tweaking I have to do to get the experience tailored for me personally is completely worth it, besides other, less tangible benefits like "free as in freedom".
If I were fine with someone else making all these decisions for me, because making them myself is more painful than accepting someone else's not entirely comfortable decision, I'd go for an Apple device, no doubt.
Tweaking the OS to play nicely with particular hardware is key. Apple are very good at it. I suppose e.g. System 76 also tweak Pop OS to run especially smoothly on their hardware. Linux is very much ready for that: when I worked at Google (2011-15), I had a Linux laptop with a Google's internal variety of Ubuntu, adapted to a relatively few hardware models they used as desktops and laptops. It worked basically flawlessly, and my T420 had like 6 hours of battery runtime browsing and coding. All I had to customize was the GTK theme and such.
Maybe something like "tweak packs" that adapt Linux to some very specific widespread hardware could be a hit.
Based on my experience, I can tell you the grass is not greener on the other side.
Forced to use Mac for work, the previous Macbook Pro would lose charge _while plugged in_! The battery goes empty, it shuts off and I wait for it to recharge while off.
They recently changed mine to an M1. Battery is much better than the previous generation and could be good, so long as you're not using it for anything. As soon as the IDE/Zoom/Docker/what have you spins up, it loses charge. It is slower at loses charge while plugged in, but so long as you are not using it except for note taking, I would not trust away from a power source.
If I don't have a power source nearby, I turn everything off and switch my dev workflow to Sublime to prolong the battery.
I don't have to suffer any of these shenanigans under Linux. Granted, I have a Thinkpad, which has great Linux support, so that definitely helps.
Sounds like defective devices. Haven't had a single issue like that on any of our fleets. (mixed T2, M1, M2 devices, around 210 per fleet, around 7 fleets in 3 different countries)
Usage (varies over time...): software development (JetBrains, NetBeans, VSCode, Sublime, vim), google meet, slack, rancher desktop, docker desktop (being phased out), capture one, creative cloud, OBS, virtual desktops for local testing.
We have had self-discharge on HP ProBooks and EliteBooks but that was due to a bad USB-C implementation and was fixed with third-party chargers. Some older Dells had it too, but that was with dual power input (USB-C and classic barrel) and switching over to the legacy chargers didn't have that issue. Those run a mix of Windows and Linux.
I have had both experiences. The base system for Linux will be great for battery life (has been ruthlessly optimized for power savings by the biggest companies in the world), but some of the desktop software on top will brutalize your battery.
I've experimented a lot (mainly on Fedora) and the biggest offenders are browsers, but also sometimes Gnome gets in a bad state and eats up a lot of CPU. It's often gjs eating CPU so it may be an app I'm leaving open or even an extension. I've tried to narrow it down but haven't fully figured it out yet.
But, if you keep only a small number of tabs open and close everything you aren't using, battery life on Linux can be really great.
> has been ruthlessly optimized for power savings by the biggest companies in the world
It has been ruthlessly optimized to save power for servers with greatly documented hardware structures.
It has not been optimized for Desktop OS, for Bluetooth, Sleep, WiFi, graphics card power saving etc.
I am a tab zero guy, for the record. I've been using computers since Windows 3, so I'm very sensitive towards intensive processes.
My wife can watch netflix on her (5 year old) Macbook for hours. I lose 13% battery life on one episode. I leave my laptop on sleep off the charger, and it is dead sleeping after a few hours. Linux users may hack their way to lower battery usage, but they still do not have it like Mac does.
You must simply leave it on the charger and turn it off when you are done. They are portable desktops.
A big difference in battery life comes down to hardware acceleration in your browser which depending on hardware,browser, and version may not be enabled.
i've clocked ~8 hours of media time with a T420s with an ultrabay battery and tlp/powertop reporting an average of 6-8w during the process. Regular battery + the extra ultrabay battery put the capacity up to about 7.7Ah; about the same range of battery capacity as a newish macbook pro.
a t420s is a very old core machine at this point. I would suspect the same conditions with a similar amount of battery with a modern processor could do a lot better; but i'm getting too old to sit in front of a screen for 8+ hours of media time, honestly -- and lately with USB-PD and power banks I have been skewing towards buying ultrabook style laptops and accessorizing via USB3 rather than with proprietary expansion slots.
I don't know what the media time on something like a laptop with an N100/linux + a powerbank, but I suspect it'd be quite a long time.
I use a pinebook pro (same company as the pinephone). Sublime works a treat. So does thunderbird, calibre.
Intellij doesn't have aarch64 Linux distributions, you can kind of hack it together it's completely unsupported. Not sure about Joplin but I expect it would be ok (I think it's electron based? VS code works ok too).
Re locked-down: pbp can run many different Linux variants, and boots with tow-boot bootloader. Hopefully that would support ThinkPad to, but I don't know.
When I last checked, the mainline kernel support for the 8cx was EXTREMELY limited. Having messed around with lots of ARM based devices on Linux in the past, I would not bother with this one. Especially given that rather unimpressive battery life.
I'd love a modern ARM based Linux laptop, but I also know that unless it's got mainline kernel support or a dedicated Linux friendly vendor (like Pine) behind it, it's lifetime and updates are going to be extremely limited. And that doesn't even consider graphics API support, which will have an impact on rather basic things, such as YouTube video playback.
Honestly, the only viable ARM laptop with good battery life and modern day performance is an Apple MacBook Air or Pro, which has a very impressive community project behind it. But even that is still incomplete and might peter out if some key figures burn out or become uninterested or preoccupied with for-pay work or whatever.
> When I last checked, the mainline kernel support for the 8cx was EXTREMELY limited.
Is that the case? I would've thought the devices present in 8cx are similar to the other Snapdragon 8 series which would probably be well supported upstream? Or do they linger in a GKI tree before landing upstream?
Maybe the exclusivity between QCOM/MS is mutual and they'll be able to partner with vendors like Canonical in the future.
Personally - I would love an XPS13 ARM linux laptop, hopefully Dell decides to jump on the bandwagon. I have owned i5/i7 XPS13s for several years now and am looking forward to ARM if they can make one that performs well.
Some 8xx series are well supported, but mostly 835 and 845. QC does release better quality sources for mainlining AFAIK, but I haven't seen any efforts related to newer chips land.
I think some of the aarch64 laptop have landed in the Linux kernel source tree.
I am currently using a C630 laptop wich could be considered a predecessor to the discussed model ( with an anemic 4GB of RAM), and following up some botched Windows update (Windows recovery put the system in a worst place). I decided to take the plunge and install Linux.
Hardware support is rough, and depends a lot on firmware file you would need to salvage from the windows partition.
So far without these files, among the most necessary stuff I would like to have.
- Wifi, I a coping with a wifi dongle, wich mean I am out of one of the 2 usb-c ports
- Sound ( usual linux problem, worked around the issue, with another dongle, or Bluetooth audio)
- External display support (worked on windows)
Otherwise, linux experience is leagues ahead of Windows, in terms of boot time and responsiveness.
Battery life is great for my use ( mainly write python and C code ). And far as battery life improvement, your mileage can vary, I haven't timed mine yet but can try and will report in a week about it (maybe you ca give me aome pointer to simulate the usual load your would put it through).
- podman work great
- sublime text run as smooth as it is on x86-64
- bios is an UEFI and I dont think it required a signed bootloader, although updating the setting to boot grub was not as straightforward.
- I am mainly using Wayland, with most of the caveats associated to it (Firefox does flicker a lot, so I am mainly using chromium)
On the most annoying side, I don't know if there is a way to configure the lid action, as it put the laptop on standby.
I would recommend, if only for the gain in thinness and if you are traveling a lot with your laptop.
Thanks for that! Probably in a couple of years, as my laptop is still serving me well enough for now. It's also pretty light and still has good battery, so not a ton of benefit there.
If the battery life was doubled (~8+ hours), that definitely would be more intriguing.
But ouch, hadn't had WiFi or sound issues on Linux since the 2009~2010 era, will not be fun to get back to that, though hopefully in a few years as the drivers mature, this would become less of an issue.
> On the most annoying side, I don't know if there is a way to configure the lid action, as it put the laptop on standby.
Even when an external display is connected? That's got to be annoying!
I have largely the same feelings on this. If you haven't used WSL on Windows, the workflow is actually very nice. You can install and run GUI apps in the Linux environment. I usually use VS Code with remoting for WSL though. My current job and personal setup aren't in windows, but have used it in the past and it's very usable, if not ideal.
For now, sticking with my macbook m1 air for personal use. Though I really don't like the differences in the kb layout and hotkeys from nix or windows.
* Battery Life. ~6/7 hours max as it currently (linux-x13s 6.4). There is a lot of room for improvement, which will happen as more people tune/tweak.
* Performance. Honestly, the only noticeable issue is lzma2 compression. It's ass slow, even when using 8 threads. Other then that- Gnome (Wayland), CLion, Firefox, and Rust have no issues. I can't think of any performance issues when comparing it to my HP 14t-ea000, which is a 11th gen i7.
* Stability. Firefox occasionally crashes and I've not done enough digging to file a ticket on it yet. The WiFi Drivers crash on suspend (https://bugzilla.kernel.org/show_bug.cgi?id=217239) but it's being worked on. The GPU crashes on startup but recovers. Going to file a kernel report on this one when I get around to it. I'd put it at 95% of what I'd expect on my x86-64 laptop.
Camera doesn't work (yet), as well as some of the other GPU features. BIOS is limited, and boils down to full access to the security options as well as the to-be-expected Lenovo keyboard options. There's a "beta" Linux boot option, but I didn't need that when I first installed linux. Actually, I'm not sure what I'd want exposed in the BIOS that's not already there. Memory Overclocking?
It runs cooler than my HP, which runs on the hotter side. Port selection is limiting, but having a 5G Modem is cool. Arch has almost all of their packages cross compiled to arm64, and I've not run into an issue where there was a package I needed that wasn't there.
TL;DR. A Very functional laptop. I've started using it as my primary laptop, but still carry my HP with me if I'm traveling somewhere. Give it a few months for the kernel issues to be ironed out and it'll be a very nice laptop.
I can't speak to the GPs reasons, but Chrome has a large enough user-base that anyone building for the web is going to want to at least test their work in it.
As if those will come any time soon. The big ARM scare of 2017 is already over. ARM is not competitive. By the time they catch up in the server and laptop markets, RISC-V will have eaten their lunch. All they have left to do is run marketing campaigns against RISC-V and watch their inevitable decline.
> Chrome is slow and frustrating to use, with delayed animations and noticeable input lag, it's such a frustration that I entirely gave up on Chrome in favor of Edge which runs natively
I use native Chromium on my Windows-on-ARM laptop (a couple of generations older than the one reviewed). It runs really well and the only drawback is missing DRM stuff so some videos won't play. YouTube almost always works, but most other non-youtube videos don't. For those I can fall back to Edge. I'm looking forward to Google releasing a native build of Chrome.
Like the author, I also do a lot of Kotlin development and use intellij and pycharm a lot. This sounds like an awful laptop to do that on. Kotlin is great but it's compiler is rather slow and you end up waiting for that a lot. So, faster is better.
I noticed about a 2.5x build speed improvements when I upgraded my cheap Samsung i5 laptop (a temporary thing I used for a while) to the M1 14" Mac Book Pro. On a typical day, I run full builds quite often so this really matters to me. The difference between waiting 5 minutes or 1.5 is really important to me. During this build, 250 integration tests run with 20 threaads maxing out all of the cpu cores.
This windows laptop is being described as a lack luster i5 sounds like it would be a hard no for me. That and all the other issues (the usual meh touchpad, the underwhelming GPU, etc.) makes for an overpriced laptop that just isn't worth having. You can probably get something way nicer for the money. Including a mac.
Silly question ; am I the only person who still uses a desktop / tower machine? Feels like I'm a huge minority these days, but when it comes to needing processing power and juice it still feels it provides budget and upgrade flexibility no laptop can match.
I notice a lot of people using desktop in the office an having a personal laptop.
But also there has been a shift to cloud instances that provide the flexibility you say at the price of renouncing ownership of the machine. Especially for GPU-based computations.
If the desktop is insecure, it is always possible to put it behind a secure firewall implemented with a small computer, e.g. a NUC-like computer or one of the many cheap, small and fanless computers with multiple Ethernet ports that are available now exactly for implementing firewalls, routers or other similar network appliances.
Of course, configuring correctly a secure firewall requires some expertise, but it should not be too difficult to find online resources or someone willing to help.
I have done exactly like this for accessing remotely my desktop during the last two decades, without any problems whatsoever.
It's much like anything else... I run my desktop under Linux, keep it upodated, have SSH on a non-standard port with only public/private key access enabled for my account. I access it over wireguard running on my home server. Wireguard is the only port available through my firewall.
So I can connect to my home network over wireguard, and SSH to my desktop. VS Code + Remoting extensions work great for this... almost like editing locally, but running remotely. The integrated terminal in VS Code is invaluable when doing this.
I have a nice big workstation with heck load of cores, memory and a 4090. It's the only way to fly. Laptops are great for emails but toys when it comes to dev machines.
I noticed a lot of people switching to desktop during work from home. They have this gaming PC just sitting there, and it turns out to be way faster than there stationary MacBook.
Personally I use a desktop at a standing desk from 9-5 and a 13" laptop around the house/on the deck.
I'm likely going to upgrade from the Dell XPS 13 to a MacBook Air, but the desktop is a higher priority. She could use some upgrades.
My personal machine is a desktop, and I have a laptop that I use for light portable web browsing occasionally. But for work I've come to expect to be given a laptop. It means you can easily WFH, take it to a meeting, or hot desk, without needing to RDP/VNC. And neither I nor my company would want me doing work stuff on my personal desktop. I do know some developers who work by remoting to a VM instead (often as part of a compromise to let developers have admin/root access for development tooling)
I use an massively overbuilt i9-13900K tower everyday and a laptop on the go. I remote into the pc when I need major grunt.
I do know a ton of people who live out of a laptop tho. I can sorta see the appeal depending but the lack of display real estate and how low the screen is and all is just a nonstarter for me for any long duration work.
And plugging up to a dock or what passes for a dock these days and external monitors and keyboards etc is a flaky annoying PITA.
You are definitely not. Writing this from my work machine, passively cooled Ryzen desktop, which has been serving me well for the past 5 years. And I don't miss much having a laptop. I work from home anyways.
I think the demographics depends heavily on industry. I know many people doing 3D work and they all use desktops these days.
I actually built a desktop tower, and then built a laptop to use it from bed! Because I found that regular laptops were totally unacceptable. I tried to cope with a gaming laptop but could not. It built up so many issues over less than 2 years that it became worthless.
Anyway, I get all the "portability" of a laptop (with my display, keyboard, and trackpad(!) in bed), but all the performance of an honest-to-Arceus desktop, with a full desktop 12400F (boosting up to 5.2GHz on all 6 cores at a cool 130W) and a full desktop RTX 3060 12GB (I regret not getting the Ti for performance reasons, but for VRAM reasons this is acceptable since I do ML work).
> Luckily (or perhaps shadily), you will rarely ever pay the full price for the Thinkpad x13s. As is tradition, Lenovo seems to be perpetually running deals, taking up to 40% off (at time of writing, 45% time of purchase), bringing the laptop to a much more reasonable $1479.60.
If something has a $2500 price tag but always has a deal where you can take it only for $1500 then the real price is $1500, not $2500.
Lenovo is not the only vendor who uses these tricks.
I have not looked recently at the Dell site, but in previous years, for many years, every time when I have looked at Dell Precision mobile workstations, they always claimed that the price that what listed at that moment was more than one thousand dollars less than the "normal" price.
May i ask why ? I don't use Macs because my day to day requires X86 and linux. But pounds for pounds if someone doesn't mind MacOs. The new Mx series of macs is just a better value than most high-end laptop theses days
I hate Apple, its products, and its ecosystem. The developer tools suck and macOS provides no advantage over windows other than an annoying user interface (if you're into that sort of thing)
I don't think the UI is annoying so much as different from what you may be used to. For me the annoyance is a few different key bindings. I like some parts of Windows, Linux (Gnome, KDE) and Mac... I have my Budgie desktop setup with a mix of the three.
>The new Mx series of macs is just a better value than most high-end laptop theses days
?
I just recently bought a laptop and when I was comparing benchmarks, I didn't see Apple anywhere near the top of the list for CPU single or multithreaded. Not that I remotely care about CPU speeds, my number 1 focus was VRAM. Made it an easy decision since you basically pick something with Nvidia and high VRAM.
This is ARM. The cpu isn't the only thing. Memory and disk io are also very good. Also portable/laptop and desktop aren't the same. Mx also has good accelerators for x86 and encoding tasks, so it will depend on what you do.
Next gen AMD mobile is also looking very good. AMD 7000 series mobile is really nice, but battery life isn't quite there compared to Apple.
We're discussing laptops here. Performance per watt is possibly one of the most important metric to the average consumer even if they don't know it by name.
Hey there, author here, for me personally it's about the windowing and file explorer. I spent 6 months trying to get used to it at a job, and even bought a M1 Macbook Pro last year to try, but I just couldn't find any combination of plugins and programs that made it feel right. I know, it's not a great reason, and I have no hate for anyone who yums my yuck, but at the end of the day it's what stops me from using MacOS.
> I just couldn't find any combination of plugins and programs that made it feel right.
This: I recently moved to Linux. I was happy on Windows, because with the right combination of plugins and programs it felt just right + I could get everything I wanted hardware-wise.
Last time I checked, this is not possible on Mac. You get a good battery life, but I don't care about that - I'd rather have a Zen3 AMD but I'm happy with my current laptop.
With Linux, I don't have to fight my OS to get everything the way I want it to be, I can get a nice 4k OLED touchscreen on my Laptop with an internal LTE/5G module so I don't have to carry a phone just for getting online, etc.
BTW I use arch (lol) and I think I'll stay - mostly thanks to Arch because before that I didn't like Linux due to Ubuntu questionable choices that made me prefer Windows before.
BTW you said " I don't have the 5G enabled version as it wasn't available when I bought mine, which I do regret, but if it's even half as good as it is on the SPX then it'll be totally fine. "
If you want to help to make 5G work, get in touch. I'd need to know if you have internal WWAN antennas- if not, you can add them. It's just about installing a M2 module. At the same time, you can replace the current 512G NVMe by a 2T like from Sabrent or WD SN730 (though the latter has some firmware issues when using ZFS due to the power states)
Hey, that's be really interesting to see. I haven't opened up my x13s yet, but I'll do that soon. In the middle of a move, so I'm not sure what box the tools are in, but once I find them I'll let you know :)
I think it's perfectly valid reason. At the end of the day a laptop is a tool, and your experience using the tool trumps any intrisic quality of said tool.
I can relate, the main reason i do not use Mac's as personal device ( i have had some issue to me in my professional life) is MacOS. It always existed in a sort of middle ground that doesn't suits my needs : Doesn't have the broad compatibility with windows (in my case gaming, CPU/GPU and tools like intel vtune) and it's a unix, but close to enough to linux where most of professional work revolves around (lack of virtualization, not elf format support, native toolings are different etc... etc...)
The hardware is amazing, the price is great, but the software is just not for me.
And it's sucks, because the landscape of high-end windows laptop is just plain embarrassing.
I am currently driving a surface pro 9 : Garbage battery life, runs hot on anything but web browsing and worst of the all the screen (120 Hz minds you) have HORRIBLE ghosting.
The only saving grace of windows laptops right now is that if one is patient there are some sweet sweet discount deals to be had.
Seems like a fairly valid reason. I used Windows for the first time since early XP recently (in a VM; wanted to check that an interview question was doable in it without undue pain). I was just astonished at how bad it felt, and couldn't wait to shut down the VM. It's unsurprising that others have the same thing in the other direction.
I used to as well. But the M series just blows away every laptop - every Pc - I’ve ever had in the last 30 years… at some point it is an ideological proposition.
My position is everything but ideological: I will consider a Mac as soon as 1) the hardware offers an OLED touchscreen and internal LTE (everything else is secondary, with DDR5 I've even relaxed my requirements for actual ECC) 2) Linux can run as well as it does on my Thinkpad or Surface.
Until then, I refuse to use a Mac, because I don't like the OS or the limited hardware choice.
Personally I think those hardware limitations are pretty silly. IMO an OLED screen is irrelevant unless you are gaming or something, and touchscreens on laptops are of pretty limited utility. And an internal LTE model would be an expensive waste if you have an iPhone, because it pairs to your iPhone instantly.
The Linux limitation is totally fair, although kind of irrelevant here (as it doesn't run well on the Qualcomm chips either).
I'm here with you. I just refuse to use Windows, but I do recognize and empathise with people who feel this way about Mac or Linux. I would rather use something because I love it, rather than not use it because I hate it; but it's totally normal to find something off-putting.
The 14-inch MacBook Pro pricing quoted in the article seems wrong.
Looking at the US Apple Store right now, it has the 14-inch MacBook Pro with M2 Pro (10-core) cpu + 32GB ram and 1TB storage as US$2599, rather an US$3099.
Interesting, when I wrote the article the prices were what I could generate on the Apple website, but you are correct that they are not the same now. I should have taken screenshots! Thanks for the heads up, I'll update the article to reflect that :)
Yeah, after I kitted things out with full upgrades, the macbook does seem to be the better option, especially if you like macos. The rosetta 2 support and docker have come a long way and most of what I use at least now has aarch64 support.
What's all the fuss about glass vs plastic touchpad in the article? I never had a glass touchpad on my laptops but I had glass touchpads on all my smartphones (the screen.) I see why the screen of the phone can't be plastic (colors, scratches, more?) but I never felt that the plastic touchpads of my laptops did not do their job as nicely as the screens of my phones.
It reminds me of the glass vs plastic phone backside arguments. Glass is more fragile than plastic and will not absorb shocks. It is also heavier and harder to replace (most plastic backsides can be removed with a guitar pick).
I was astonished at the amount of ink, or pixels wasted by this ridiculous and completely manufactured distinction on the part of inept tech "journalists" and bloggers who had to find something non-technical to identify a pecking order among phones in order to attract readership who didn't understand the more subtle distinctions between features that affect actual usage of the phone.
The only other ridiculous argument that had an even worse effect on the net usability of phones, as well as the indistinguishable limited variety that we have today, is the ridiculous obsession with thinness. And then everyone puts a case on it anyway.
Thin, thin, thin, and a glass back on top of that. I'm not usually sympathetic to conspiracies but it sure seems like the interests of phone manufacturers ( especially one in particular) who want to sell a new throaway device every year were suspiciously well served by both of these absurd abitrary obsessions of the tech press.
People say "vote with your feet" and "just don't buy the stuff you don't like", but you can't exactly buy nothing when there are no options you like, because what if you need a laptop anyway? So you end up having to buy something "thin and light" which is a signal to the OEM that people want more thin and light even if all they wanted was a laptop.
Unfortunately the phrase vote with your feet is too often misapplied either by the hopelessly idealistic at one end, or by the toxically cynical at the other.
particularly those who are well aware that you have two feet to work with while their side has paid influence wrangling a stampede of cattle in the other directiin.
I personally like light phones and thin or small are two ways to make them light. I prefer small to thin if I have a choice, but the sub 6" Android phones can be counted on the fingers of a hand nowadays (with fingers to spare) so I can't be too picky. 200 g seems to be the new normal. Ugh!
It just feels nicer to use. Not more functional to use, but it does feel different to your fingertip.
Like the author mentions, it's not a dealbreaker, but is a calling-card of this particular machine's build/materials/specs not living up to its pricetag.
I'm going to assume that, not being an Intel processor, it won't have Intel Management Engine. So where does this device land privacy-wise? Can this device safely be recommended for highly privacy-focused people who might otherwise choose "respects your freedom certified" [1] legacy hardware? Or does it have something similar to IME with similarly negative privacy implications?
Given that this is a Lenovo laptop, one who is truly concerned about privacy may want to avoid it regardless [0].
Not to mention none of it is open source, from the firmware to the hardware, so I hardly see this as a suitable replacement for people looking to run FSF-approved computers.
I find that battery life extremely uninspiring. I wonder if that's because of Windows, JetBrains or a combination of both.
I run a x86-64 laptop with Linux AND decicated graphics (Ryzen 6850u and Radeon 6800S) and get 6-10 hours of battery life while developing.
Granted, that's after quite a bit of tweaking - I've configured my monitor to go to 60hz, limit the CPU to only 4 cores/8 threads and 2200MHz while on battery. I write Rust with rust-analyzer so the CPU is constantly in use, and my browser of choice (Firefox) is far less optimized for battery life than any chrome-based browser is on Windows.
This gives me a laptop with no concessions with regards to compatibility and it's probably running loops around this ARM based chip, even when running at my tuned-down power settings.
I'd love to have a Linux compatible ARM laptop, but if they don't give you real world double digit battery life, why bother?
Meanwhile I've had my laptops plugged in for 99.999% of their life.
Sometimes I'll unplug to bring it to an overnight christmas party or camping to watch a movie at night. Then I'll probably end up plugging it in when I arrive.
Just remembered, one time I brought my laptop to a real estate meeting, and it was so boring I did work during it. But the meeting was over in an hour.
I'm sure some people like having their computer not plugged in, but a docked laptop is so nice.
"high-end" arm windows machines never made much sense to me. It seems that the main appeal is longer battery life, but looking at the recent CPU from AMD, and the rumors of the next CPU from intel.It's seem clear to me that x86 is making more progress in reducing power consumption than arm is making in performance. And i wouldn't be surprise that in a couple of year we get very efficient x86 cpu's from both AMD and intel.
AMD is getting here, certainly... the latest 7430HS mini PCs still use 6-8W at idle though, which is quite a bit more than ARM still. I've been using an m1 air as a personall laptop and when traveling rarely even think about having to plug it in... for my limited use with the screen around 60% brightness, I can use it all day and then some. Mostly text/email or remote editing (wireguard, ssh, vs code).
I agree, Mx laptop are still miles ahead in term of performance per what. But i would venture to suggest that a lot of it is more about the platform (OS, BIOS,chipset etc... etc...) then the CPU per-say.
Full disclosure - I do work at Microsoft on Windows. One of the work devices I have is a Surface Pro X - and it has been generally a good experience. I don’t really write code on the device (I work on Hyper-V), but for browsing, Office, Remote Desktop, WPA etc it’s good.
My biggest annoyance is that Motu don’t have ARM64 drivers for their audio interfaces, so I can’t use the one I have for microphone input. I’ve tried asking - but it doesn’t seem like it’s even on their roadmap, which is unfortunate. I’m pretty sure they license the Thesycon driver which does now support ARM64 [0] - but I guess they need to either update or just build the ARM64 version… The inbox class driver doesn’t work because Motu use implicit feedback.
Its 2023 and I still don't understand the point of CPU speeds, even as a dev doing both single and multithreading. AI is continuing to put priority on GPUs.
In the worst case, I wait an extra fraction of a second with a bad CPU.
With a bad GPU, you are basically excluded from the latest technology.
I'm sure ARM really wants to advertise that CPU is relevant, but it continues to be nearly irrelevant when you buy something a laptop with a dedicated GPU. (Its good enough)
I suppose it depends on what performance bracket you are targeting. Scripting and interpreted languages, JIT next and VM after that (i.e. JVM, CLR), native code after that and optimised low-level code after that would all have very different reactions to 'good CPU' vs. 'bad CPU'.
A 'slow' process (±10s) taking 1 second longer won't seem that much slower, but a fast process (±100ms) taking ten times as long does definitely become problematic.
As for AI, it's mostly irrelevant what your local machines does, unless you are an end-user (i.e. you are running a model from a vendor on a local accelerator for local processing instead of remote processing). Development happens on remote farms and on-line usage also happens on remote farms.
Try encoding video in real time or NP hard optimization problems (e.g. generating bitstreams for FPGAs) instead of running something as trivial as a database and a Webserver.
I'm sitting here with "only" 12GB of VRAM and I can't train even the bigger RWKV models, and that's a freaking RNN, it's not even a transformer! And that's highly disappointing for someone who's been working as a contributor to one of the inference libraries, because the fruits of my labor aren't even going to benefit me that much. Bleh~
I (somewhat) lie, all my experience working on rwkv.cpp does gain me knowledge and also contribute to the community. I'm just sort of sad that I don't have enough VRAM to finetune a bigger model to fit something like a chat format, or learn a certain vocabulary or whatever.
After some years of a 15" 2016 MacBook Pro (perhaps the worst laptop Apple ever made) at work, the M1 Max that replaced it was a revelation. I have no use for a fast GPU (in fact one of my pet peeves about the 2016 MBP was that they eliminated the no-discrete-GPU option that prior 15"s had, resulting in far worse thermals for no gains for many users). Your milage may, of course, vary, but some of us appreciate a fast CPU and have no use for a fast GPU (the M1 Max is, I gather, tolerably fast in this department, but the poor thing never gets to do anything other than displaying GUIs in my case...)
RE ML stuff, do people actually do the heavy lifting for that _on laptops_ much?
CPU stuff, let's see... For a dev, maybe running their own program, compilers (sometimes with a lot of optimization passes), interpreters for python/ruby, an IDE with indexing over the whole project, running the OS itself, tests, browser tabs... None of that runs on a GPU! (except part of the browser rendering). CPU is fairly important.
That's ecause the most important component of a laptop is not its CPU; it is the --==[[ keyboard ]]==--.
A laptop without a keyboard with decent key travel and tactile feedback simply deters you from using it. People who do not write much (code, documents, etc.) - can perhaps make do with point-and-click-oriented laptops. We cannot. That's why are currently condemned to choose between 10-year-old CPUs and I/O ports, but with decent keyboards; and fancy modern CPUs in laptops that are inconvenient to use.
Yeah i have to call BS on this narrative that somewhat the keyboard is the most important thing for developer. Writing code is not writing documentation or writing books : Most of time spent not on typing letter... Debugging, reading code, thinking about a code etc... are more important. I have yet to see a case where typing speed is a limiting factor for coding.
Loads of non-dev information workers type way more than dev...
It's true that if a laptop doesn't have a monitor, you can't use it to read code. But - you don't need an ultra-perfect color reproduction, super-low refresh rate etc. on your monitor to read code. So if your laptop has a TN rather than IPS panel - that might be consideration but nothing like decent keyboard or not.
Please keep in mind that humans have a huge variety of anatomy, movement, and strength. While I don’t doubt that high travel keyboards are more comfortable and painless for YOU specifically, I can assure you that your experience does not speak for everyone. Extrapolating your keyboard preferences to every body shape is like expecting an Olympic gymnast to be comfortable in a pregnant baryatric person’s pants.
Please consider that most people don’t care about long key travel, and that some people such as myself actually strongly prefer short key travel as it reduces pain for me. Also, have you used this keyboard or are you judging it based on one zoomed out marketing image?
With the current specs, pricing and usefulness this actually seems like a bad deal. Not a useless deal, and in the Windows space perhaps the only real deal, but a bad deal nonetheless.
What it can be is a good starting point for the industry to get moving. Not sure if this runs SystemReady firmware or if this is using Microsoft's UEFI stack, but either way it could finally get vendorland moving.
I hope that's not the case... as most developers are not USers, and while they may make a decent living, I doubt most developers would consider blowing thousands of dollars on a Mac (regardless of whether they like it SW/HW-wise). And who buys Microsoft laptops? That's weird.
Yes, but this a english speaking forum, from a mainly US based incubator, so the audience is self-selecte... Nothing against dev from other part of the world.
> I doubt most developers would consider blowing thousands of dollars on a Mac
Mac are consistently the most selling laptop... so clearly someone is buying them.
People forget that it is not just the processor that consumes power. The reason the new MacBooks are so efficient is not just because they are ARM-based, it is also because they have incredibly energy efficient screens, WiFi, RAM, SSDs, etc. They can achieve this through their sophisticated manufacturing process, with highly integrated components directly soldered onto the board, a fact that is widely (and correctly) criticized for making repairs and expansion nigh impossible, but it is also true that this is much more power efficient than what you can achieve with a more modular design.
I have a Thinkpad L14 AMD, it consumes well below 10W for casual browsing/reading mails under Linux. I would wager that maybe 1/3 of that is consumed by the CPU. So even if you had a CPU with 0W consumption, I would just increase my battery life by a third, at least for that kind of usage.
Is it really that hard to design sockets that don't waste such a significant amount of energy? Honest question. Do you really save so much by soldering something without any type of ribbon, socket, or other removable connector in between?
For RAM, yes. Soldering it down right next to the CPU allows for lower voltages and faster bus speeds. For SSDs, it's more difficult. It surely makes no difference for standard PCIe, there's really no excuse to solder that one to the board. I'm actually not sure how exactly SSDs work in a MacBook M1/M2. I think it's plain NAND storage soldered directly onto the board, with the controller in the M1/M2. I have no idea about power savings here, but it sure is fast...
For RAM I think even the trace length on the motherboard matters but this is just a reason for why the sockets are usually near the CPU, not a reason why you would have to eschew sockets altogether.
I'm drawing a blank on what it was called now, but way back in the day, in response to the Intel Apple laptops having far better battery life than apparently equivalent Wintel machines, Intel launched a programme to improve laptop platforms. IIRC, at the time, some laptops had up to about 10W worth of inefficient/non-essential (or completely unusable; controllers for parallel ports but no actual parallel ports for instance) chippery on their motherboards.
I'd assume all the low-hanging fruit is _long_ gone there, though.
To be fair, Apples anti repair bullshit is not just soldered components. I'd buy them if it were just that.
It's their refusal to release necessary software and tools to allow for repair, their design that is unnecessarily hard to open and repair (there is 0 reason to have any glue in a notebook ever) and them fucking over anyone in their supply chain making it impossible to even buy replacement components. They do this so they can quote you outrageous repair costs so you buy a new device instead
> Chrome is slow and frustrating to use, with delayed animations and noticeable input lag, it's such a frustration that I entirely gave up on Chrome in favor of Edge which runs natively.
This is kind of surprising, because Windows for ARM has been around for _7 years_ (I'm ignoring ye olde Windows RT here, because everyone else did). I got an ARM Mac Mini a couple of months after they launched; at that point, while there was still a good bit of stuff with poor or no native support (most annoying: IntelliJ; don't think it got native support for about six months), native Chrome was already available.
I suppose the difference is that third parties don't see Windows for ARM as a serious product, but there's a definite chicken and egg problem there.
> There are some Asus laptops that seem like they'd fit the bill, but they're just so clunky and large, which gets in the way of the portable dream.
Does anyone know which laptops Devin is referring to?
I'm one of those odd guys who love big laptops (I've used 17” models for over a decade now) and I don't care about the weight at all. But I'd love it if I could trade some of those models' performance for more battery life (a reliable 8 hours would be sufficient), and they're never designed for that.
Hey there! I don't remember what laptop I was specifically thinking of when I wrote that (Apologies, I started writing this back in February and got distracted for a number of months) but there is a non-zero chance I was thinking about the Asus ZenBook Pro Duo. Sorry I can't be more specific about past me's thought process :(
167 comments
[ 5.1 ms ] story [ 209 ms ] threadBased on this: https://blog.rymcg.tech/blog/linux/thinkpad-x13s/ GPU support seems to have issues, which is a shame. That was written a few months ago, though.
I had a M2 MBP issued for work by my previous job and it really was excellent; performance and battery life. But I'd have a hard time giving up Linux, and I'd worry about installation of Linux on hardware like that always being second-class.
I'm surprised the author is using these under emulation. Surely they'd run at completely native performance if you just rub the through WSL2? Linux distributions have had ARM based distributions for ages.
If he was stuck using windows, maybe best bet would have been to run an X server and run IntelliJ etc inside WSL itself.
You'd lose out on things like hardware accelerated graphics but reading your description it sounds like that's not really something you rely on anyway.
But not sure how the dev workflow would be: Podman, IntelliJ, Sublime Text, etc.. And some casual stuff like Calibre, Thunderbird, Joplin.
Oh and how much would the battery life improvement be. I already get 4~5+ hours on my 5 year old Thinkpad, so an improvement to 6 hours is not really worth it.
How is the BIOS and openness on these systems? I always heard the arm laptops are way more locked down than the x86 ones.
Also, a computer without a single fan so I can comfortably use it on my bed? With possibility to replace it's SSD? It sounded a lot like my next laptop.
...yeah, I was wondering that too. This seems like a job for a Gentoo system, compiled from the ground up with "-march=native". Would love to know how that would turn out performance-wise, in comparison to Windows.
GKISS if you want glibc.
God, I have experienced the exact opposite with Linux on laptops. Battery management is so bad I am considering moving to a Macbook.
I went from a constantly-overheated Linux laptop to a completely calm Linux laptop, on the same hardware, distro, and DE. Checking for unnecessary processes that eat CPU, spurious ACPI events, overly-heavyweight browser tabs goes a long way. Making sure the browser uses hardware support to play YouTube is another heavyweight.
It takes some time, but much less time than I had expected.
And to be specific, I suffer from bad hardware/software communication. On a linux laptop, no less (System76, which I would never recommend). Linux S3 sleep is not hardware sleep, I lose a lot of battery there. WiFi and Bluetooth connectivity either has to be battery draining or have poor performance.
The battery my laptop has itself is also too small, and could only get moderately good life if it had an optimized OS anyway. Again, a Linux-first laptop.
I am sure if I switched to a Lenovo with a big battery that I would be better off than I am now, so I'm also complaining a bit about my hardware. But the problem is still there.
With Apple M2 paired with an OS designed for it, with high end batteries and screens, I don't see any pairing of (laptop) hardware and software competing, on a fundamental level. Every other system is second class.
Edit: ACPI events is the key word here, and it is not worth anyone's time to wade through that garbage to get sane laptop performance. The ACPI stuff isn't even designed for end users to alter, and it is hacked together due to all the hardware variations out there. It is simply doing the best job it can, on average.
There's kind of an implicit lie when people say linux works on laptops at all. It works on MOST laptop configurations in MOST functions. But it BARELY works in others. You find this once you dig sufficiently deep enough.
I keep my laptop plugged in. It's pathetic.
No this is completely down your hard hardware
If I were fine with someone else making all these decisions for me, because making them myself is more painful than accepting someone else's not entirely comfortable decision, I'd go for an Apple device, no doubt.
Tweaking the OS to play nicely with particular hardware is key. Apple are very good at it. I suppose e.g. System 76 also tweak Pop OS to run especially smoothly on their hardware. Linux is very much ready for that: when I worked at Google (2011-15), I had a Linux laptop with a Google's internal variety of Ubuntu, adapted to a relatively few hardware models they used as desktops and laptops. It worked basically flawlessly, and my T420 had like 6 hours of battery runtime browsing and coding. All I had to customize was the GTK theme and such.
Maybe something like "tweak packs" that adapt Linux to some very specific widespread hardware could be a hit.
Forced to use Mac for work, the previous Macbook Pro would lose charge _while plugged in_! The battery goes empty, it shuts off and I wait for it to recharge while off.
They recently changed mine to an M1. Battery is much better than the previous generation and could be good, so long as you're not using it for anything. As soon as the IDE/Zoom/Docker/what have you spins up, it loses charge. It is slower at loses charge while plugged in, but so long as you are not using it except for note taking, I would not trust away from a power source.
If I don't have a power source nearby, I turn everything off and switch my dev workflow to Sublime to prolong the battery.
I don't have to suffer any of these shenanigans under Linux. Granted, I have a Thinkpad, which has great Linux support, so that definitely helps.
Usage (varies over time...): software development (JetBrains, NetBeans, VSCode, Sublime, vim), google meet, slack, rancher desktop, docker desktop (being phased out), capture one, creative cloud, OBS, virtual desktops for local testing.
We have had self-discharge on HP ProBooks and EliteBooks but that was due to a bad USB-C implementation and was fixed with third-party chargers. Some older Dells had it too, but that was with dual power input (USB-C and classic barrel) and switching over to the legacy chargers didn't have that issue. Those run a mix of Windows and Linux.
I've experimented a lot (mainly on Fedora) and the biggest offenders are browsers, but also sometimes Gnome gets in a bad state and eats up a lot of CPU. It's often gjs eating CPU so it may be an app I'm leaving open or even an extension. I've tried to narrow it down but haven't fully figured it out yet.
But, if you keep only a small number of tabs open and close everything you aren't using, battery life on Linux can be really great.
It has been ruthlessly optimized to save power for servers with greatly documented hardware structures.
It has not been optimized for Desktop OS, for Bluetooth, Sleep, WiFi, graphics card power saving etc.
I am a tab zero guy, for the record. I've been using computers since Windows 3, so I'm very sensitive towards intensive processes.
My wife can watch netflix on her (5 year old) Macbook for hours. I lose 13% battery life on one episode. I leave my laptop on sleep off the charger, and it is dead sleeping after a few hours. Linux users may hack their way to lower battery usage, but they still do not have it like Mac does.
You must simply leave it on the charger and turn it off when you are done. They are portable desktops.
i've clocked ~8 hours of media time with a T420s with an ultrabay battery and tlp/powertop reporting an average of 6-8w during the process. Regular battery + the extra ultrabay battery put the capacity up to about 7.7Ah; about the same range of battery capacity as a newish macbook pro.
a t420s is a very old core machine at this point. I would suspect the same conditions with a similar amount of battery with a modern processor could do a lot better; but i'm getting too old to sit in front of a screen for 8+ hours of media time, honestly -- and lately with USB-PD and power banks I have been skewing towards buying ultrabook style laptops and accessorizing via USB3 rather than with proprietary expansion slots.
I don't know what the media time on something like a laptop with an N100/linux + a powerbank, but I suspect it'd be quite a long time.
typo: meant 7.7aH
Intellij doesn't have aarch64 Linux distributions, you can kind of hack it together it's completely unsupported. Not sure about Joplin but I expect it would be ok (I think it's electron based? VS code works ok too).
Re locked-down: pbp can run many different Linux variants, and boots with tow-boot bootloader. Hopefully that would support ThinkPad to, but I don't know.
https://www.jetbrains.com/idea/download/download-thanks.html...
https://www.jetbrains.com/rider/download/download-thanks.htm...
I'd love a modern ARM based Linux laptop, but I also know that unless it's got mainline kernel support or a dedicated Linux friendly vendor (like Pine) behind it, it's lifetime and updates are going to be extremely limited. And that doesn't even consider graphics API support, which will have an impact on rather basic things, such as YouTube video playback.
Honestly, the only viable ARM laptop with good battery life and modern day performance is an Apple MacBook Air or Pro, which has a very impressive community project behind it. But even that is still incomplete and might peter out if some key figures burn out or become uninterested or preoccupied with for-pay work or whatever.
Is that the case? I would've thought the devices present in 8cx are similar to the other Snapdragon 8 series which would probably be well supported upstream? Or do they linger in a GKI tree before landing upstream?
Maybe the exclusivity between QCOM/MS is mutual and they'll be able to partner with vendors like Canonical in the future.
Personally - I would love an XPS13 ARM linux laptop, hopefully Dell decides to jump on the bandwagon. I have owned i5/i7 XPS13s for several years now and am looking forward to ARM if they can make one that performs well.
I am currently using a C630 laptop wich could be considered a predecessor to the discussed model ( with an anemic 4GB of RAM), and following up some botched Windows update (Windows recovery put the system in a worst place). I decided to take the plunge and install Linux.
Hardware support is rough, and depends a lot on firmware file you would need to salvage from the windows partition.
So far without these files, among the most necessary stuff I would like to have.
- Wifi, I a coping with a wifi dongle, wich mean I am out of one of the 2 usb-c ports
- Sound ( usual linux problem, worked around the issue, with another dongle, or Bluetooth audio)
- External display support (worked on windows)
Otherwise, linux experience is leagues ahead of Windows, in terms of boot time and responsiveness. Battery life is great for my use ( mainly write python and C code ). And far as battery life improvement, your mileage can vary, I haven't timed mine yet but can try and will report in a week about it (maybe you ca give me aome pointer to simulate the usual load your would put it through).
- podman work great
- sublime text run as smooth as it is on x86-64
- bios is an UEFI and I dont think it required a signed bootloader, although updating the setting to boot grub was not as straightforward.
- I am mainly using Wayland, with most of the caveats associated to it (Firefox does flicker a lot, so I am mainly using chromium)
On the most annoying side, I don't know if there is a way to configure the lid action, as it put the laptop on standby.
I would recommend, if only for the gain in thinness and if you are traveling a lot with your laptop.
If the battery life was doubled (~8+ hours), that definitely would be more intriguing.
But ouch, hadn't had WiFi or sound issues on Linux since the 2009~2010 era, will not be fun to get back to that, though hopefully in a few years as the drivers mature, this would become less of an issue.
> On the most annoying side, I don't know if there is a way to configure the lid action, as it put the laptop on standby.
Even when an external display is connected? That's got to be annoying!
For now, sticking with my macbook m1 air for personal use. Though I really don't like the differences in the kb layout and hotkeys from nix or windows.
It runs cooler than my HP, which runs on the hotter side. Port selection is limiting, but having a 5G Modem is cool. Arch has almost all of their packages cross compiled to arm64, and I've not run into an issue where there was a package I needed that wasn't there.
TL;DR. A Very functional laptop. I've started using it as my primary laptop, but still carry my HP with me if I'm traveling somewhere. Give it a few months for the kernel issues to be ironed out and it'll be a very nice laptop.
I'll try running firecracker later and see if it'll boot.
I'm like the author, I consider the M MacBooks very impressive but I dislike the OS itself.
And it was very unexpected that Chrome underperform so much, without it it's a complete deal breaker for me.
But I think the author is using Chrome under emulation, so bad performance is expected.
I use native Chromium on my Windows-on-ARM laptop (a couple of generations older than the one reviewed). It runs really well and the only drawback is missing DRM stuff so some videos won't play. YouTube almost always works, but most other non-youtube videos don't. For those I can fall back to Edge. I'm looking forward to Google releasing a native build of Chrome.
I noticed about a 2.5x build speed improvements when I upgraded my cheap Samsung i5 laptop (a temporary thing I used for a while) to the M1 14" Mac Book Pro. On a typical day, I run full builds quite often so this really matters to me. The difference between waiting 5 minutes or 1.5 is really important to me. During this build, 250 integration tests run with 20 threaads maxing out all of the cpu cores.
This windows laptop is being described as a lack luster i5 sounds like it would be a hard no for me. That and all the other issues (the usual meh touchpad, the underwhelming GPU, etc.) makes for an overpriced laptop that just isn't worth having. You can probably get something way nicer for the money. Including a mac.
But also there has been a shift to cloud instances that provide the flexibility you say at the price of renouncing ownership of the machine. Especially for GPU-based computations.
Since my laptop is basically just a thin client used for ssh, vnc, and a web browser, it’s super easily replaced.
Cant people detect when their offline machines are being/have been hacked?
Of course, configuring correctly a secure firewall requires some expertise, but it should not be too difficult to find online resources or someone willing to help.
I have done exactly like this for accessing remotely my desktop during the last two decades, without any problems whatsoever.
So I can connect to my home network over wireguard, and SSH to my desktop. VS Code + Remoting extensions work great for this... almost like editing locally, but running remotely. The integrated terminal in VS Code is invaluable when doing this.
Personally I use a desktop at a standing desk from 9-5 and a 13" laptop around the house/on the deck.
I'm likely going to upgrade from the Dell XPS 13 to a MacBook Air, but the desktop is a higher priority. She could use some upgrades.
I do know a ton of people who live out of a laptop tho. I can sorta see the appeal depending but the lack of display real estate and how low the screen is and all is just a nonstarter for me for any long duration work.
And plugging up to a dock or what passes for a dock these days and external monitors and keyboards etc is a flaky annoying PITA.
I think the demographics depends heavily on industry. I know many people doing 3D work and they all use desktops these days.
Anyway, I get all the "portability" of a laptop (with my display, keyboard, and trackpad(!) in bed), but all the performance of an honest-to-Arceus desktop, with a full desktop 12400F (boosting up to 5.2GHz on all 6 cores at a cool 130W) and a full desktop RTX 3060 12GB (I regret not getting the Ti for performance reasons, but for VRAM reasons this is acceptable since I do ML work).
If something has a $2500 price tag but always has a deal where you can take it only for $1500 then the real price is $1500, not $2500.
I have not looked recently at the Dell site, but in previous years, for many years, every time when I have looked at Dell Precision mobile workstations, they always claimed that the price that what listed at that moment was more than one thousand dollars less than the "normal" price.
If I couldn't find a good Thinkpad, I would get a Surface.
?
I just recently bought a laptop and when I was comparing benchmarks, I didn't see Apple anywhere near the top of the list for CPU single or multithreaded. Not that I remotely care about CPU speeds, my number 1 focus was VRAM. Made it an easy decision since you basically pick something with Nvidia and high VRAM.
Next gen AMD mobile is also looking very good. AMD 7000 series mobile is really nice, but battery life isn't quite there compared to Apple.
Probably wasn't normalized as perf per watt.
Of course, that is a marketing metric, not something to genuinely care.
Better battery life + less heat = more comfort.
This: I recently moved to Linux. I was happy on Windows, because with the right combination of plugins and programs it felt just right + I could get everything I wanted hardware-wise.
Last time I checked, this is not possible on Mac. You get a good battery life, but I don't care about that - I'd rather have a Zen3 AMD but I'm happy with my current laptop.
With Linux, I don't have to fight my OS to get everything the way I want it to be, I can get a nice 4k OLED touchscreen on my Laptop with an internal LTE/5G module so I don't have to carry a phone just for getting online, etc.
BTW I use arch (lol) and I think I'll stay - mostly thanks to Arch because before that I didn't like Linux due to Ubuntu questionable choices that made me prefer Windows before.
If you want to help to make 5G work, get in touch. I'd need to know if you have internal WWAN antennas- if not, you can add them. It's just about installing a M2 module. At the same time, you can replace the current 512G NVMe by a 2T like from Sabrent or WD SN730 (though the latter has some firmware issues when using ZFS due to the power states)
I think it's perfectly valid reason. At the end of the day a laptop is a tool, and your experience using the tool trumps any intrisic quality of said tool.
I can relate, the main reason i do not use Mac's as personal device ( i have had some issue to me in my professional life) is MacOS. It always existed in a sort of middle ground that doesn't suits my needs : Doesn't have the broad compatibility with windows (in my case gaming, CPU/GPU and tools like intel vtune) and it's a unix, but close to enough to linux where most of professional work revolves around (lack of virtualization, not elf format support, native toolings are different etc... etc...)
The hardware is amazing, the price is great, but the software is just not for me.
And it's sucks, because the landscape of high-end windows laptop is just plain embarrassing. I am currently driving a surface pro 9 : Garbage battery life, runs hot on anything but web browsing and worst of the all the screen (120 Hz minds you) have HORRIBLE ghosting.
The only saving grace of windows laptops right now is that if one is patient there are some sweet sweet discount deals to be had.
Seems like a fairly valid reason. I used Windows for the first time since early XP recently (in a VM; wanted to check that an interview question was doable in it without undue pain). I was just astonished at how bad it felt, and couldn't wait to shut down the VM. It's unsurprising that others have the same thing in the other direction.
Until then, I refuse to use a Mac, because I don't like the OS or the limited hardware choice.
The Linux limitation is totally fair, although kind of irrelevant here (as it doesn't run well on the Qualcomm chips either).
Slower than a 3 years old M1...
It's 2023 people..
Looking at the US Apple Store right now, it has the 14-inch MacBook Pro with M2 Pro (10-core) cpu + 32GB ram and 1TB storage as US$2599, rather an US$3099.
Seems outright cheaper than the Thinkpad x13s.
But glass is considered premium.
The only other ridiculous argument that had an even worse effect on the net usability of phones, as well as the indistinguishable limited variety that we have today, is the ridiculous obsession with thinness. And then everyone puts a case on it anyway.
Thin, thin, thin, and a glass back on top of that. I'm not usually sympathetic to conspiracies but it sure seems like the interests of phone manufacturers ( especially one in particular) who want to sell a new throaway device every year were suspiciously well served by both of these absurd abitrary obsessions of the tech press.
People say "vote with your feet" and "just don't buy the stuff you don't like", but you can't exactly buy nothing when there are no options you like, because what if you need a laptop anyway? So you end up having to buy something "thin and light" which is a signal to the OEM that people want more thin and light even if all they wanted was a laptop.
particularly those who are well aware that you have two feet to work with while their side has paid influence wrangling a stampede of cattle in the other directiin.
Like the author mentions, it's not a dealbreaker, but is a calling-card of this particular machine's build/materials/specs not living up to its pricetag.
[1] https://ryf.fsf.org/
[0] https://en.m.wikipedia.org/wiki/Lenovo#Security_and_privacy_...
I run a x86-64 laptop with Linux AND decicated graphics (Ryzen 6850u and Radeon 6800S) and get 6-10 hours of battery life while developing.
Granted, that's after quite a bit of tweaking - I've configured my monitor to go to 60hz, limit the CPU to only 4 cores/8 threads and 2200MHz while on battery. I write Rust with rust-analyzer so the CPU is constantly in use, and my browser of choice (Firefox) is far less optimized for battery life than any chrome-based browser is on Windows.
This gives me a laptop with no concessions with regards to compatibility and it's probably running loops around this ARM based chip, even when running at my tuned-down power settings.
I'd love to have a Linux compatible ARM laptop, but if they don't give you real world double digit battery life, why bother?
Sometimes I'll unplug to bring it to an overnight christmas party or camping to watch a movie at night. Then I'll probably end up plugging it in when I arrive.
Just remembered, one time I brought my laptop to a real estate meeting, and it was so boring I did work during it. But the meeting was over in an hour.
I'm sure some people like having their computer not plugged in, but a docked laptop is so nice.
Do we really need more than 6 hours of never being plugged in?
I've heard of a usecase, bus trips with buses that don't have 120V outlets... Its pretty niche.
My biggest annoyance is that Motu don’t have ARM64 drivers for their audio interfaces, so I can’t use the one I have for microphone input. I’ve tried asking - but it doesn’t seem like it’s even on their roadmap, which is unfortunate. I’m pretty sure they license the Thesycon driver which does now support ARM64 [0] - but I guess they need to either update or just build the ARM64 version… The inbox class driver doesn’t work because Motu use implicit feedback.
[0] - https://www.thesycon.de/eng/usb_audiodriver.shtml
In the worst case, I wait an extra fraction of a second with a bad CPU.
With a bad GPU, you are basically excluded from the latest technology.
I'm sure ARM really wants to advertise that CPU is relevant, but it continues to be nearly irrelevant when you buy something a laptop with a dedicated GPU. (Its good enough)
A 'slow' process (±10s) taking 1 second longer won't seem that much slower, but a fast process (±100ms) taking ten times as long does definitely become problematic.
As for AI, it's mostly irrelevant what your local machines does, unless you are an end-user (i.e. you are running a model from a vendor on a local accelerator for local processing instead of remote processing). Development happens on remote farms and on-line usage also happens on remote farms.
I can picture wanting to do nsfw stuff locally...
But also this is a fortune 100 company that has the airgapped computer with the confidential stuff. Its big bucks.
I (somewhat) lie, all my experience working on rwkv.cpp does gain me knowledge and also contribute to the community. I'm just sort of sad that I don't have enough VRAM to finetune a bigger model to fit something like a chat format, or learn a certain vocabulary or whatever.
RE ML stuff, do people actually do the heavy lifting for that _on laptops_ much?
This is kind of my point.
What actual CPU stuff is done in 2023?
Its why Apple's marketing focuses so much on battery life. For a year, they won a competition that no one had cared to win.
... but, why would I use such harsh rhetoric?
That's ecause the most important component of a laptop is not its CPU; it is the --==[[ keyboard ]]==--.
A laptop without a keyboard with decent key travel and tactile feedback simply deters you from using it. People who do not write much (code, documents, etc.) - can perhaps make do with point-and-click-oriented laptops. We cannot. That's why are currently condemned to choose between 10-year-old CPUs and I/O ports, but with decent keyboards; and fancy modern CPUs in laptops that are inconvenient to use.
Now look at this laptop's keyboard:
https://p3-ofp.static.pub/fes/cms/2022/03/03/ww9s3cycoxj6gbs...
they keys barely even protrude from the surrounding surface!
And debugging we do with our brains mostly :-)
Please consider that most people don’t care about long key travel, and that some people such as myself actually strongly prefer short key travel as it reduces pain for me. Also, have you used this keyboard or are you judging it based on one zoomed out marketing image?
What it can be is a good starting point for the industry to get moving. Not sure if this runs SystemReady firmware or if this is using Microsoft's UEFI stack, but either way it could finally get vendorland moving.
What kind of person is writing that article? Sounds like someone who's identity is first and foremost a rich guy from the US.
I hope that's not the case... as most developers are not USers, and while they may make a decent living, I doubt most developers would consider blowing thousands of dollars on a Mac (regardless of whether they like it SW/HW-wise). And who buys Microsoft laptops? That's weird.
Yes, but this a english speaking forum, from a mainly US based incubator, so the audience is self-selecte... Nothing against dev from other part of the world.
> I doubt most developers would consider blowing thousands of dollars on a Mac
Mac are consistently the most selling laptop... so clearly someone is buying them.
> And who buys Microsoft laptops? That's weird.
Agree. Big mistake, but this one is on me :)
I'd assume all the low-hanging fruit is _long_ gone there, though.
yes? That's the entire reason why SoC exists.
This is kind of surprising, because Windows for ARM has been around for _7 years_ (I'm ignoring ye olde Windows RT here, because everyone else did). I got an ARM Mac Mini a couple of months after they launched; at that point, while there was still a good bit of stuff with poor or no native support (most annoying: IntelliJ; don't think it got native support for about six months), native Chrome was already available.
I suppose the difference is that third parties don't see Windows for ARM as a serious product, but there's a definite chicken and egg problem there.
Does anyone know which laptops Devin is referring to?
I'm one of those odd guys who love big laptops (I've used 17” models for over a decade now) and I don't care about the weight at all. But I'd love it if I could trade some of those models' performance for more battery life (a reliable 8 hours would be sufficient), and they're never designed for that.