1. 62
  1. 36

    Tangentially related, it seems like macOS on M1 swaps very enthusiastically, to the point of possibly bringing SSD life well under a year: https://twitter.com/marcan42/status/1361151198921826308

    (I didn’t want to post a twitter thread as a submission, but seems like it may be of interest.)

    As marcan42 points out in the thread, this is clearly an OS software issue so it should be patchable in software as well.

    1. 11

      Interesting. I’m up to 5.4TBW; kernel_task has written 69.5GB (!) in 3 days 1 hour of uptime. One to keep an eye on; thanks for sharing.

      1. 8

        This is madness, I’ve only managed 15 TBW on a Samsung 970 EVO 1TB on an extremely heavily used laptop in 28 months, under 2% the drive’s warrantied TBW, and keenly aware I’ve been hammering the drive at various points. That also includes 2 h00j VMware VMs

      2. 16

        Not a big problem, just swap out the SSD of your MacBook when it’s dead. Oh wait, it can’t be replaced :/ Another reason to support Right To Repait.

        1. 5

          This might be controversial, but I think you’re just watching what happens when miniaturization and integration happens. The SSD on these is basically directly connected to the FSB, and that contributes to the performance of it. How do you make that replaceable effectively?

          Your ALU used to be a discrete, replaceable component. Then it became integrated. Then L2 cache. Should it stop, especially if integration can make things more reliable (i.e RAM slot failure)?

          1. 5

            I think “things that are consumables” such as batteries are those things that absolutely must be replaceable. SSD’s fit that category because they actually wear out over time.

            But I think you raise good points about other discrete components, not being able to upgrade my RAM sucks, but if it’s more reliable, performant, cheaper and uses less power than alternatives, then it’s a compelling choice.

            1. 5

              I agree that this is miniaturization and integration, but I’d argue it’s not strictly necessary for performance.

              AFAIK the M1 RAM is LPDDR4X-4266 and you can buy DIMMs[*] in this specification as well. The SSD is NVMe and as far as I know there’s nothing special about the signalling compared to an off-the-shelf NVMe SSD.

              integration can make things more reliable (i.e RAM slot failure)

              I don’t have any numbers to back this up, but my gut feeling is that in the average upgradable laptop the number of lifetime failures that require replacing the RAM is going to be equal or higher than the number of lifetime failures that require replacing the RAM slot - so there’s a gain and a loss here.

              I’d suggest it boils down to three things:

              • Integrating everything on one board (or one package in the case of the RAM) is cheaper to design, manufacture and test.
              • Integrating everything makes the product smaller and slimmer, and portable device consumers love slim products (as do Apple industrial designers, it would seem).
              • Upgrading or repairing laptop internals is not something the majority of laptop customers plan to ever do (unfortunately), and there is no other regulatory pressure requiring this (which brings us back to Right to Repair).

              [*] EDIT: I originally thought you could buy SO-DIMMs in this spec, but maybe only DIMMs. I think it’d be technically possible to have a small size & replaceable standard for these, but maybe the industry is going with soldered RAM to the extent that it doesn’t exist.

              1. 1

                I wonder how much putting RAM on as an MCM lets them run LPDDR at those speeds/latencies.

              2. 1

                especially if integration can make things more reliable (i.e RAM slot failure)?

                And yet, the only failing RAM I had in a machine in the last 10 years was a MacBook Pro with on-board RAM. If the machine actually had a DIMM slot, it could’ve been replaced without replacing the whole logic board. (Since the MacBook Pro was just two days old, they replaced the whole system, of course.)

            2. 2

              This comment should be a separate post by itself. Thank you for the heads up!

              1. 2

                Not a problem on mine somehow (918GB writes in 6 weeks).

                1. 2

                  I’m second thinking getting an M1 now, maybe I’ll wait for this to be fixed. Hopefully, in time for the new macbook pros. :p

                  Still my current Linux laptop is 4 years old, and has <10TB TBW on its nvme. I haven’t used it a lot in the past 6 months but it has been in used daily before that. So, 918GB in six weeks still seems like a lot.

                  1. 2


                    Just checked my 3.5 year old MB12, it had 27.5TB writes over 182-ish weeks, which is roughly 0.9TB/6W. So yeah, it’s normal.

                    1. 1

                      I’ve had a 2019 MBP from work for almost a year now, and I’m at 65.8 TB written. I don’t think this is an M1 problem so much as a macOS problem (if indeed it’s actually a problem).

                      1. 1

                        Yes it’s certainly an OS issue.

                        Could be some combination of usage patterns with memory configuration. Like I don’t do npm or use any heavyweight IDEs, maybe they provoke the system to swapping out more.

                      2. 1

                        FWIW smartctl claims 27 TB written on my mid-2012 MBA11. I’m no expert, but I think my wearout is all zeroed out. Can’t upgrade past 10.15, not sure if OS matters.

                    2. [Comment removed by author]

                      1. 2

                        I get 0.26 MBps: (918*1024)/(6*7*24*3600).

                    3. 1

                      I have had this experience with macOS (VM) in general, testing my memory profiler’s OOM detection (https://pythonspeed.com/fil). it seems much more aggressive about swapping than Linux, to the point where I needed to come up with a different heuristic.

                    4. 15

                      Could have been titled “The Worst Experience I’ve Had With Emacs” no? Seems like everything else is amazing.

                      1. 19

                        I admit the title is clickbait, pedantically it is my best experience with apple silicon too!

                        1. 18

                          Why the compulsion to use a clickbait title? I have a feeling I know but I’d rather you say than me.

                          As for the actual article: ARM processors in general seem to be the way of the future. The other day for fun I was looking up processors made in USA and Canada and so many companies are making ARM processors. Whoever is developing software for them is going to see the rewards in a decade.

                          1. 7

                            This post was a fluff piece that I wrote on a whim and didn’t expect it to be anywhere much as popular as it is. I chose a clickbait title purely because I thought it would be amusing to.

                      2. 11

                        runs a lot faster than it has any right to

                        Consistently my experience with my M1 mbp. I am still floored by how good it is, several months on. When I go out with it, I don’t bother bringing any kind of power cable. I was worried I might regret the decision as someone who relies on her primary laptop for development (both work and hobby), but it’s honestly been so good.

                        One word of advice: If you’re thinking of getting an M1 macbook, fork out extra (both cash and probably waiting/delivery time) for the 16gb of RAM. I did so and my experience tells me it was a very correct decision.

                        1. 7

                          How does Nix fare on the new hardware and operating system?

                          1. 3

                            It is only amd64 at the moment, but it’s worked fine in my testing.

                          2. 3

                            I do like the idea of this new macbook and will probably get one. I don’t like the fact that a reasonably trivial observation, that is not especially different from the previous 27* similar observations is submitted to lobsters, is the most upvoted story on the front page at the time of looking, on a site that [for me at least] has been more about edifying CSey things than consumer electronics news.

                            I have no solution, I’m just writing this incase anyone else who reads it agrees and decides not to upvote this kind of thing next time. Nothing personal against Cadey, a lot of what they submit is less transient in subject matter (and transience or otherwise is obviously just my own value judgement anyway). I am just a small voice in the news-vs-self-improvement culture war on lobsters.

                            *exaggerating for effect.

                            1. 3

                              I didn’t expect this post to be as popular as it is. I thought it would end up being one of those that gets 5 points and then flitters off into the ether. I just wanted to write about a tool that I use that ends up being beneficial for me.

                              I try to make more evergreen content (ie, content that ages well), however this ended up being one of those little bits of filler fluff that ended up way more popular than it really had any reason to (CloudFlare is reporting that there’s been 27,000+ hits to that page in the last 72 hours, which is more than twice what my post on Saturday about FreeBSD had at around 8,000+ hits since it was posted).

                              It feels like a reasonably trivial observation because it was really a trivial observation circled around my astonishment at how fast this thing builds Rust code without getting warm.

                              I’ll try to see if I can do more comprehensive deep-dives in the future.

                              1. 2

                                I must emphasise again, I wouldn’t for a second wish to suggest this is a reflection of you or what you wish to write about - your website, your party, your canvas - and I wouldn’t want you to modify your behaviour as a result of my own particularly personal preferences about what I’d like lobste.rs to be (vs say HN). Please carry on, I like your site, if you want to do a post about comparing PCB-mask colour schemes on a custom mechanical keyboard, then that’s glorious and please do without even a moment’s hesitation on the basis of one audience vs another.

                                It was more merely a comment addressed to anyone else who is, like me, inclined towards a less frothy front page and more something that is more likely to reward a thirty minute coffee break from coding with some new ideas and perspectives on how we go about solving our problems. I fully accept that the people speak with their upvotes but I think it’s worth maintaining a quiet energy to push back against tyranny of the majority in place of something less sensation and more thoughtful - c.f. US politics recently - even if it seems a bit futile*. I am happy to be this quiet (or even downvoted) voice at the bottom of an occasional thread, I just think it’s useful for someone to say it periodically.

                            2. 3

                              While the new MBP is probably the best performance laptop out there, my older desktop has much better per thread performance on my loads (CPU/memory bound, dataset size >> L2 cache size).

                              Heat dissipation still helps.

                              1. 2

                                I think Cocoa Emacs may need a bigger rewrite to become future proof. It will probably survive macOS 11 but the Cocoa integration code really is a huge complicated 20? year old hair ball.

                                1. 2

                                  Would be interesting to see if the new pgtk (pure gtk) integration works, since gtk3 does support macOS. It’s still WIP of course, and I’m not saying it should replace cocoa, but still…

                                  1. 2

                                    As a macos user, there is nothing about gtk3 or gtk4 that would make me want it to be used in gui emacs on macos.

                                    Why would I trust a work in progress over something that actually uses the platform? Why is macos special in this regard and windows being left out? Why not just update cocoa emacs instead of make emacs an even worse experience on macos? Will gtk improve the spawned process speed on macos?

                                2. 2

                                  I’m so keen to see what apple does with the second and third generation chips, it felt to me like this first generation was intentionally held back in specs to get them into the hands of people (devs) who really need them, but make them hopefully unattractive to most buyers. The results so far are pretty astounding, and Rosetta 2 being more than usable is a huge win, I hope Apple give a talk about how it works one day (when it stops being a killer app).

                                  1. 1

                                    I was able to compile a native M1 emacs GUI by tweaking this a bit: https://gist.github.com/naokton/762bea1b1a4a6301bc986d57b6a3c223

                                    Seams to be running pretty well :-)

                                    1. 1

                                      Something that really surprised me was how well Dolphin runs when you use a native build. I’m able to play Gamecube and Wii games at retina resolution and the MacBook doesn’t even get warm to the touch. The amd64 version of Dolphin uses some Just-In-Time compilation that Rosetta can’t emulate at all, however the aarch64 one runs a lot faster than it has any right to. It must be easier to translate binaries between RISC processor types or something. You have to build Dolphin from source when you do this, however it’s worth it.

                                      Dolphin’s AArch64 JIT is definitely not as sophisticated as x86_64 but is surprisingly mature! You’re running the same engine that people installing Android get.

                                      1. 1

                                        I’m getting to the point of wanting to replace my 2014 Macbook Pro, but I’ve been so turned off by the keyboards over the past years that I’ve been holding off. I had to get the fans replaced on this machine about a year ago, but it’s been good otherwise.

                                        I’m glad to see they fixed the keyboard. I couldn’t bring myself to spend around $2000 on a laptop that I would hate typing on. I might wait a bit to see how things go with the M1 before I grab a new one, but I’m glad to read positive reviews.

                                        1. 3

                                          The keyboards have been fixed very well since the 2019 16”, if you need Intel.

                                          1. 3

                                            The post-butterfly keyboards have definitely been fixed, but I don’t think they’re as good as the 2014-5 era Macbook Pro. I had to pull out my old 2015 MBP a couple weeks ago and immediately thought, “wow, i really miss this keyboard”. So you might want to try one out in a store before you buy, if possible.

                                            disclaimer: I’m comparing to a 2020 16” Macbook Pro, which I’m not sure is identical to what’s in the M1 macs. There may be variations in the post-butterfly keyboards?

                                            1. 3

                                              I have both - a 2015 MBP and an M1 Air. They’re a little different, yes, but I don’t know if one’s better than the other. Once I’ve got used to one for a while, the other is slightly surprising, but that’s all.