Hacker News Reader

Comment by pyrophane | Hacker News Reader

Keep in mind that probably the most important spec when considering a new laptop is one that is often not directly stated: the processor series.

I'm not talking about i3/i5/i7, but rather U/Y/H. This letter determines the TDP (thermal design power/point) them machine is designed to run at. The TDP will govern the setting for the base clock speed, and, just as importantly, the throttling behabior under load.

Processor series TDPs are Y: 4.5W, U: 15W, H: 45W.

The new MacBook Air appears to have a Y series processor, like the MacBook, which means it will be aggressively throttled to keep power consumption and heat generation low.

Practically, that means that the new Air will not be capable of running sustained workloads much above its base clock speed, which makes it unsuitable for many programming-related tasks.

The Pro is still a much better choice for programmers. The 13 is suitable for many things, but the 16, with the H series processor, is really preferred for computationally intensive work.

You can get away with this machine if your workflow primary involves a text editor and remote servers, but otherwise I would still opt for the pro.

View on HN

The CPUs in the new MBA have 9W TDPs.

OEMs can still control load performance, it's not entirely determined by TDP.

Anandtech has a good article: https://www.anandtech.com/show/13544/why-intel-processors-dr...

Depending on OEM settings and cooling capacity, it's possible for Intel CPUs to indefinitely run at greater than their base clocks and TDP.

Totally. To back this up with a real world example, Apple screwed up the throttling of the i9 MacBooks originally, causing them to be over throttled. A software update that pretty much amounted to an MSR write on boot fixed it.

https://www.macrumors.com/2018/07/24/throttling-fix-2018-mac...

Why does Intel let OEMs play with that? It's to give OEMs a knob they can play with to balance perf and price for their segment. It lets OEMs cheap out on cooling (or go with new form factors like the tiny GPD models) and go with the low minimums, or have a full system that's capable of higher TDP like Apple's tend to be.

Macbook Air has extremely bad cooling (on purpose, because it's thin and quiet). Don't expect it to boost well.

"Boost well" probably means different things to different people.

On my fanless 12" MB with a 7W TDP m7-6Y75 (1.3 Ghz base, 3.1 GHz max turbo), running a Prime95 torture test stabilizes at the following for the CPU via Intel Power Gadget: 9.5W, 1.95 GHz, 90°. (Laptop feels warm to the touch - it only ever feels "hot" under GPU load.) Normal all-thread load stabilizes around 2.5 GHz.

This is underestimating modern processors and overestimating programming compute resource usage.

Your post may apply if we're working with 4K+ resolution video files, rendering and other activities, but the modern programmer, even compiling binaries, will be fine on a MacBook Air.

How do I know?

I use a 2 generation old Macbook as my primary personal development machine. I write Go, Rust, Java and TypeScript using the common tool chains for all those languages.

You haven't been forced to deploy something that builds a local K8s cluster and builds and runs 10+ containers just to get a microservices app up and ready for development, I see... (only partially /s).

That's a good point, and outside of my personal development workflow. Condolences!

It doesn't even take that much. Docker Desktop on MacOS is widely known to be terrible and have extremely poor performance. Running webpack-dev-server with filewatching through Docker nearly grinds my 2017 Macbook Pro to a halt. Obviously most node/webpack/JS stuff can run natively on Mac but it's not uncommon to see dev stacks that dockerize everything because that's how the production stack works.

I often have up to 10 VSCode editors, an IntelliJ window and two browser instances open.

Add some server processes, UI watchers/builders, and my Thinkpad starts to freeze up. In an ideal world, I would have a low-latency remote desktop going, but even in 2020 it seems like a pipe dream.

I’ve been using paperspace cloud machines with parsec for low latency Remote Desktop. Originally used it for gaming. It’s remarkable how good it is for everything else too.

5ms encoding lag, 10ms network lag, totally workable.

You can improve this substantially on MacOS using docker-sync (the issues with Docker performance on Mac are primarily related to the file system)

I wrote a little about this; it's for Rails but the same ideas apply:

https://medium.com/@bdcravens/fixing-docker-for-mac-and-rail...

docker-sync helps a little bit, and so does limiting Docker to 1 CPU core, but I still routinely get spurts 100% CPU from Docker that despite supposedly being on 1 core still grind my machine to a halt.

That's my work pretty much except not k8s but it does spin up between 4 and 8 containers (depending on which system) some of which require 4GB of RAM or more (don't ask..).

On the flip side standard issue laptops are now 16" Macbook Pro's with 32GB of RAM which is pretty hardware.

Just because there are needs for high-end CPUs for some developers it doesn't mean that all programmers need them, as OP's comment implied.

I don't get this argument. Run local for basic testing, else use it as an ssh machine. I work with super computers. The power of my development machine usually doesn't make much difference, I spend a lot of my time ssh'd anyways. So if something has a nice screen, good battery life, and a comfortable keyboard then I'm good. As long as it works for presentations and can compile my latex files, I'm good.

I don't see Airs as development platforms. They are portable platforms.

In fairness, as a Rust dev my computer sounds like it's a jet engine anytime I compile. I wouldn't even think about using a slimmer CPU. Not that you're wrong, it's just where my head is at haha.

Well my "programming laptop" spends all of its CPU these days on corporate garbage such as splunkd, which regularly melts down, some kind of corporate managed software updater which is terrible, Zoom, and trying to render dashboards from Grafana which is among the most flagrantly wasteful javascript hacks ever devised. It doesn't help me much that vim and screen are efficient.

programming != running a text editor.

A lot of usecases involve running a replication of the server environment locally in resource intensive containers

I'm lucky to have development access to a small slice of a very large vmware cluster.

I'm surprised no one else operates this way.

The performance gains are immense.

Or how about just having a workstation at work? E.g. the Dell XPS 15 is a very popular laptop, but an equivalent cost workstation is around 8x faster for sustained well-parallelizable loads. Use a KVM switch with your work monitor, or use it as a headless server, do whatever. Want to work on machine learning stuff? Throw in an RTX 2080 Ti, off you go, never worry about cloud costs. Are you I/O bound? RAID0 off a couple NVMe drives.

It's immensely flexible and I'm surprised it's not more common.

It's more cost upfront, which is often a show-stopper. "Why can't you spin a beefy instance on EC2 when you need it for these 2-3 hours of intense work? It's so much cheaper!"

Also, a cloud VM, if it breaks, is replaced "for free" and "instantly"; fixing a physical machine has a cost and is a delay. Unless you're big enough to keep a spare (probably makes sense for many dozens of machines), it also looks a bit uneconomical from the business's POV.

Never mind that your cloud costs in a year end up being comparable to said workstation.

There are other people who operate that way. But lots of them use AWS or some other provider for that instead of local resources.

Sometimes latency issues. I work from home office.

Don't we all :)

That was my experience at previous employers, but not my current one. We have some shared test environments in the cloud, but individualized test environments are discouraged. They're in the process of rolling out more powerful workstation laptops to provide us with enough horsepower to do testing locally.

That's why I'm tempted to build a custom workstation at home: so that I always have access to the compute resources I need to do my job.

I'm using a 4-year-old Thinkpad with a dual core U-series i7 and 32GB or RAM. The results are...adequate.

A typical compile for me takes about twice as long compared to colleagues with newer H-series processors, but the individual projects are pretty small so that often isn't a huge deal.

However, once you try and run nearly a dozen docker containers for local testing, things slow down quite a bit. Multitasking while those tests are running is very difficult.

I thought the same thing, but then I got a MacBook Air and couldn't believe how slow it was in practice. I didn't know if it was thermals or memory bandwidth or something, but Java compiles which used to take 5 minutes would suddenly take 30 minutes and it just wasn't even really tractable to use.

Generally speaking If you need any heavy computational workload for more than a few min the MacBook Air doesn't fit. The 3.4Ghz Boost only works in seconds and it long run it could only sustain at about 2 Ghz.

Not to mention previous MacBook Air were Dual Core only.

Yeah, I'm writing C++ on a 2013 Macbook Air. This would be a _huge_ upgrade for me.

I agree with you :) but let's not tell the employers who gladly buy the top-of-the-line models for their programmers ;) I am quite happy with my new 2020 MBP.

Generalizing "programming" is a fools errand at this point. For some, their IDE alone would warrant a pro-spec machine. Some people need to virtualize their development environments. Some people need to virtualize several environments interconnected. And for some people, programming is just a simple terminal session with tmux running - VIM on one screen, shell in another, tests in another.

I totally agree that you can't fit all programmers into one bucket. I think we should be able to describe some general patterns in programmer workflows such that we can give concrete advice to programmers about what kind of hardware they would need.

A perfect example of this the recent Level1Techs video about compiling an Unreal engine game using the AMD Threadripper 3990X[1]. Some concrete advice:

1. Often the heaviest aspect of a programmer's workflow isn't compiling the code, but running automated tests using VMs and stuff

2. compilers tend to benefit from large cache sizes and good single-threaded performance. Multi-threaded performance only makes a difference in limited scenarios.

I've also heard that storage performance, and in particular storage latency, can make a big difference on compiler performance, but the difference is starker on Windows than on Linux because NTFS is a lot slower than many Linux file systems (EXT4, XFS, etc.).

The thing is, we as programmers don't have a good mental model for how the software tools we use (IDEs, virtualization/docker, compilers) scale with modern hardware, and very few technology reviewers are producing content that would help us understand.

[1]https://www.youtube.com/watch?v=VQa6r6Ci1jg

Depends on what IDE you use. Webstorm pushes even my octacore pretty well. Sure, some people prefer to work like it's still 60s tho.

Even Chrome brings my machine to a halt. I've been running Docker recently and 8Gb of ram is instantly gone. Sometimes I think most folks are simply more patient with their computers, and will wait a few seconds for operations to finish, while I get a feeling something is taking too long and start messing around with ps aux, killing processes and such...

You can lower docker resources in a settings. I use it on my air.

That is peanuts. Have you tried opening a browser and surf the world wide web?

Compiling routinely pegs all 12 cores on my top-spec MBP for 5-10 minutes at a time. YMMV.

What about large excel files with a bunch of pivot tables and v/xlookups?

I picked up the new 2018 Air right after the redesign, 16gb and 512 ssd model. I have had no problems with xcode doing iOS dev (even using simulator on occasion), vscode and nodejs/redis/mysql working fine, also no problems running windows in parallels desktop, it’s super fast and quiet!

Great machine, only wish I had waited for the true tone and keyboard redesign but those are minor, have grown to love the keyboard.

Same here. No problems doing development. The Achilles' heel? Work conferencing with Google Meet. Kills the poor machine.

I did the same but got 32GB. It is a fantastic machine, but it is even better with the blackmagic eGPU. Heavy recommend on that if considering a home dock setup.

Can someone explain chip speed and relevance in a way I'd understand pretty please? I dialled out of this stuff years ago - at a point when bigger numbers just meant better, but now they just seem the same?

I'm still using a 2013 MBA as my primary machine day to day. It has an i7 and 8gb of RAM. I've been waiting for the keyboard change before I upgraded, but today I see my options are i5 and i7 (as 2013). I'd get 16gb of RAM, but literally have no clue about the chip stuff. I'm assuming we've moved on somewhat from 2013?!

Any help much appreciated! Thanks in advance!

For Intel chips i3/i5/i7 have generally represented feature ranges, so to really identify the chip you have to find out their product code or at least processor generation. Once you have that you can use ARK to distinguish between them. Here's an example of an older 2017 era desktop processor of i5 and i7: https://ark.intel.com/content/www/us/en/ark/compare.html?pro...

Complicating things recently is the introduction of the Xeon W and i9 lines, but for the most part the Xeon ones have yet more a new features and i9 are higher core counts to compete with AMD.

In broad strokes in the laptop space the majority of enhancements since 2013 I would say is higher RAM capacity (that Apple as a manufacturer has not to pursue compared to Lenovo, etc.), still yet lower overall power consumption to improve battery life, DDR3 to DDR4 ram, bigger L1/L2/L3 caches, and normal improvements to embedded graphics to support bigger onboard and external displays, and PCI-E enhancements for drives as NVMe-style drives are what I would consider the new standard.

Thank you!

Same question here. My 2011 Air is basically unusable now, so I've been waiting for this machine. i3, or $300 more for the i5 (which was my first inclination). Then you get into the i5 machine and they offer an i7 for another $150.

My work won't be processor intensive, just a personal machine. But given how long I kept my old one, I intend to keep this one for a long time, and I'm not sure where the best price:performance tradeoff is. I could argue going low-end and replacing it sooner, or high end, and keeping it forever.

Great to hear others getting same mileage as me - my late 2012 Air bought in March 2013 still going strong with only a battery replacement done so far. 4Gb RAM, 128 Gb SSD. Used daily for browsing, watching videos online, building IOT software, playing with unikernels. The only time it hasnt performed was when used last year as a work machine while waiting for macbook to arrive - and that was down to the collection of corporate stuff required. Once got new work machine, i removed all the work stuff, and the Air went back to being the solid daily hack it has always been. In the same time my friends with other laptops have had several cycles.

It's largely turned into a doorstop after the battery was dying a year or so ago. I bought a $70 battery from a random seller on advice from a friend who had good luck with fit and instructions. Replacement was a breeze, and it seemed great at first, but pretty quickly started just dying on me. I use the laptop so rarely, it just seems like, no matter what the SOC was last time I used it, next time I pick it up (usually a week or more later), it was dead and needed power.

that said, about a year or two ago was also when it just started feeling too slow to be usable (beachballs all the time). I blame web bloat, mostly :-/

I bought it new, so I definitely got my money's worth out of it.

Purely in terms of the Options of MBA 2020.

i3 - Dual Core i5 - Quad Core i7 - Quad Core and slightly higher Clock speed.

That is it. Apple has kept it fairly simple most of the time.

Worth nothing the i7 is only very very very slightly faster as you are limited by thermals. I dont think it is worth the price. If you ever need the processor power then the current ( and future ) MBA isn't for you.

And in all honestly the CPU performance between your 2013 MBA and this MBA isn't all that different if you are getting the Dual Core. After all your 2013 CPU had 15W of thermals to play with, this newer CPU only has 9W peak, ( and 7.5W average ).

But the overall package of MacBook Air still many times better. Screen Quality, Ports, SSD Speed. Speakers. Not sure if they have updated the WebCam, if not it will be worst than your 2013 MBA.

Well I'm on a 2016 macbook air, 8gb RAM and doing web programming everyday. 3 phpstorm windows, sometimes Photoshop open, more than 20 tabs in Chrome, iterm, figma, paw... Honestly it runs pretty well and I was waiting for this update.

Assuming PHP you just described "a text editor and remote servers"

Apparently PHPStorm and Photoshop aren't "text editors" one would imagine.

PhpStorm = Big, Java-based IDE. It's not some simple text editor.

PHP Dev often means some sort of local server running too.

big complex IDEs are almost never compute bound.

compilers are often compute bound (certainly on SSDs).

web servers are almost never compute bound.

You severely underestimate JetBrains IDE.

any IDE that is compute bound should be thrown out, especially for a language that is never compiled.

One has to wonder about the carbon footprint of millions of developers running these monstrous IDEs.

Yeah, but that's a lot of us. Obviously there are some programmers out there doing computation-bound work, but an awful lot of us aren't. If you're doing web stuff or making apps, then basically any computer manufactured in the last 5-8 years is going to be fine. I do iOS side work on a 2012 quad-core Mac Mini. It's totally fine.

It's OK to geek out about hardware, if that's your thing. But it's not really necessary for most people -- even programmers.

I think you’ll be happy with the update, especially if you were happy with the previous machine. Remember you can try it out for 14 days risk free, so if you have any doubts you have time to test it thoroughly before committing. https://www.apple.com/shop/help/returns_refund

How are you going to return it if all Apple stores are closed due to COVID-19 ?

Lucky you, because yours still has a U processor, not a Y one.

I'll wait for some benchmarks then. Maybe there's not much improvement. I see Apple is comparing last model that had 2 cores with this new 4 cores and saying there's a 2x improvement.

If you are running something computationally intensive isn't a desktop more suitable?

45W will get your laptop pretty warm and noisy and will drain your battery.

I'm on a 7200u (i5, 15W) and I could really use the extra oomph (rounding things, it'd be 40% extra performance assuming a cubed increase in power consumption vs performance improvement -- which is pretty close to what I've seen in practice) of it being 45W.

Been looking at the newer gaming notebooks, and I'm waiting the "real world confrontation" of 10th generation i7s (specially those that have AVX512) vs Renoir (lots of cores, AVX2 on a single instruction and (not as much as desktop) lots of cache), since I run some numeric heavy code and it's not quite clear how they'll stand up to each other.

Why do I run this on a notebook and not a desktop? Because I'm always on the move, and I can pack the notebook and go anywhere I have to and be able to work with or without internet connection (which is sometimes the case).

You might be interested in the ThinkPad P1 (Gen 2). It has the option of not including a dedicated GPU, which will save a considerable amount on battery power, while still packing in a 6-core i7 H-series processor.

But I am also interested in the upcoming processor generations.

Thanks for the suggestion. I was willing to wait a bit more to see that standoff: What will get me more bang out of the processor, the "super" architecture of Zen 2, or the coveted AVX512?

The P1 comes with a 9th gen i7, I assume they'd eventually get a gen 3 with 10th gen i7 and then I'll be able to compare.

> What will get me more bang out of the processor, the "super" architecture of Zen 2, or the coveted AVX512?

Bychance, have you seen the article from puget systems on how to use MKL with Ryzen[1]? It explores some of the different trade-offs between AMD and Intel for numeric computing.

[1] https://www.pugetsystems.com/labs/hpc/How-To-Use-MKL-with-AM...

Yeah, I'd seen that, great link! Waiting to see something like that with the mobile processors.

> 45W will get your laptop pretty warm and noisy and will drain your battery.

So put it on a stand and plug it in? Problem solved.

> isn't a desktop more suitable

No it isn't any more suitable, because you can put your laptop on a stand and plug it in to achieve the same thing.

It's not the same thing by far: Laptops have much worse heat dissipation than desktops, they overheat, and your high-powered CPU gets throttled down.

Some people need to carry their computer around, so the laptop has to be a compromise between stationary computing power and portability. The MacBook Pro strikes this balance quite well all around. Of course there are bigger and heavier laptops, if you require even more performance and are not limited by portability.

I find it a bit opaque that they do not advertise exactly which series/gen they are putting on the machine. Can't find it even on the Tech Specs page.

They don't advertise their memory models or disk manufacturer either. Probably typical customer does not care. Wait for teardowns if you're interested in those details.

I'd say anyone buying a pro careas at least about the generation.

I'm a FE developer (Mainly React/Node/React Native Development) and I'm still using a 2012 13" Macbook air (8Gb) as my daily driver at home. There's a noticable difference between it and my work (granted still relatively old) 2017 MBP 17". Running builds and an entire test suite maybe take twice as long but it's overall very usable and I see no reason yet to upgrade.

but it's overall very usable and I see no reason yet to upgrade.

HiDPI. Once I had used a HiDPI screen for a while, I can't really tolerate LoDPI displays anymore.

I agree. I'm an a 2012 15" MBPr, and and I can see myself using this laptop for a long time still.

>You can get away with this machine if your workflow primary involves a text editor and remote servers, but otherwise I would still opt for the pro.

I bought a 2018 15 inch MBP with 6 core i9, 32 GB of ram and vega GPU with the idea that it would be my one stop development machine. The system is terrible, especially considering the outrageous price - it does have sufficient compute power but the thermals are insanely bad - it's constantly turning the fans to 100% and I can't even tune the power usage - when I run a VM/emulator along the IDE people in the office start turning their head in my direction because of the fan noise.

And the performance still isn't even close to a mid range desktop machine which would have cost 1/4 the price.

I'm seriously considering selling my MBP buying this Mac Book Air top config (I need an OSX machine to develop iOS/OSX apps, otherwise I would go for Lenovo X1 Carbon) and building a desktop which will be always on VPN for heavy lifting + it can replace my gaming console.

The 2019 Pro completely fixed the thermal issues, it is actually considered to have pretty impressive thermals now.

https://www.youtube.com/watch?v=WxocVricANg

And the performance still isn't even close to a mid range desktop machine which would have cost 1/4 the price.

I have a MacBook Pro 2018 and Intel NUC running Linux with exactly the same CPU. In typical C/C++/Rust builds, the Linux NUC is perceptibly much faster than the MacBook.

Didn't dive much deeper, but I'd guess the difference is thermals (despite the NUC having a small enclosure) and much higher system call overhead in macOS.

That said, I still like having a Mac as well, since there are so many great applications that I use frequently. But a sweet spot (as you suggest) is a reasonably powerful MacBook for the things that macOS is good at and a powerful Linux workstation or server for compute.

Unlike most other Y-series devices (including the MacBook), the MacBook Air has a fan. This allows it to dissipate more heat and sustain higher clock rates.

Yep, so many people don't think about that when they look at laptops and it's so annoying.

FWIW, the Y CPU in the Air is set at 9W instead of the usual 4.5W

This is why there is a huge difference between 8-gen or greater Intel CPUs on mobile. As of 7th gen, the only mobile CPUs that had quad-core were the HQ-series 45W TDPs which required a totally different cooling design. Then, with 8th gen, all the regular low-TDP CPUs got quad core and the 45Ws started offering six-core.

For example, the only ThinkPad of that generation's T-series that had one was the T470p which had a significantly different design in terms of batteries, thunderbolt features, etc. to accommodate the significantly larger cooling system. With the very next year's model, all the models got quad-core with their usual low-voltage/low-power (and low battery usage) design.

The nice thing about this one is the lack of the touch bar.

MacbookAir always seemed like a tablet with keyboard to me. I have seen ordinary people complaining that they can't multitask apps there, that they have to run one app at the time. I think Air is not supposed and never was to be used as a development or video processing machine.

Where do you even find this info? Even on the Tech Specs page for the macbook air on apple.com it says "3.0GHz 6-core Intel Core i5 Turbo Boost up to 4.1GHz 9MB shared L3 cache" and doesn't give the model number.

Practically, that means that the new Air will not be capable of running sustained workloads much above its base clock speed, which makes it unsuitable for many programming-related tasks.

Does that still holds when using the power adapter?

OP's comment was about thermal dissipation, which, if anything, would be a more difficult problem with the power adapter (since the processor doesn't need to be throttled to conserve battery, for example). So, yes.

Thanks, I guess the question then is, for how long can the mid and max clock rates be sustained.

So for peak load, it’s an upgrade compared to the previous generation MacBook Air. But for sustained load, it seems to actually be a downgrade, as the previous gen still had processors of the U series.

as the previous gen still had processors of the U series.

No? The base model had a i5-8210Y?

https://en.wikipedia.org/wiki/MacBook_Air#Technical_specific...

I am sorry, with previous generation I meant the non-retina version of 2016, which indeed had U processors.

I've been issued a MBP at work and now thinking of picking a 5 year old MBP to work at home, since switching two OSes is overwhelming at this point. Is this a good idea?

Yes. What kind of work do you do that prevents you from taking the MBP home?

> The Pro is still a much better choice for programmers.

I'm not totally sure I agree. I typically do a lot of programming on my MacBook Air. It's simply a box I use to ssh to something else more powerful to do the actual runs.

Yes, if you're running computationally intensive stuff on your MacBook Air you're going to be disappointed, but if what you're mostly doing is typing into a terminal screen then it's probably perfect. And probably even over-powered? :)

That's exactly what he said at the end of his comment.

Learned something new, thanks!

Reply on news.ycombinator.com