Businesses above a certain size overwhelmingly use pcs unless they are in specific arty sectors. M1 won't change that, or at least not rapidly.
I think Microsoft have 5 or 6 different lines of business that bring in billions or more of revenue: windows, cloud, gaming, office/productivity and surface. They are the most diversified of the big players.
Totally agreed. I'd also want to add something for this:
> Businesses above a certain size overwhelmingly use pcs unless they are in specific arty sectors. M1 won't change that, or at least not rapidly.
Besides the fact, which many HN users working for FAANGS or startups don't encounter, that big enterprises are both rigid and also quite locked into the Microsoft/Windows/Office mono culture, big companies tend to have very elaborate purchasing processes.
Things with just 1 vendor tend to be completely ignored, if they're meant to be used throughout the company. There is a certain irony with my previous paragraph, but Microsoft stuff is usually grandfathered in, the rules for new acquisitions don't apply to them. On top of that Microsoft has super solid backwards/forwards compatibility, migration paths and an amazing history of working with enterprises, as much as they are hated especially by the Open Source community for other things.
So you can expect ARM Macs to be adopted and used, but most likely frowned upon by IT departments and CIOs if too many people want them. With Intel they can have a bidding process or at least a product comparison between Dell, HP, Lenovo, etc. With Apple it's... just Apple. Also Apple's B2B side is super underpowered and their history is very chequered due to their frequent breaking changes.
I have seen that business orgs within an enterprise with their own P/L usually end up doing an end-run around IT and just buy MacBooks from suppliers or directly from Apple.
Funnily enough, both times I saw this happen (at a large bank and later at a utility company) the only thing we needed from IT was network access and a Windows VM license so that we could run the god awful company payroll app that required IE11 to work. Once a week dozens of very highly paid devs and SREs would boot up a Windows SOE VM, login to the domain, start the payroll app, enter in five rows of data and shutdown. What a waste of time and money.
We kind of have this setup for those who want Linux as their desktop OS, it's provided by our IT and it's great.
If you want a Linux workstation, IT will provision you a Windows VM. It has email, Office, and our payroll app. With most stuff on Teams you can ignore it most of the time and get on with your work. IT's happy, we're happy.
Second that. Can I have Apple serviceman in 4 hours at my doors after I call that something is broken? Will that serviceman arrive with the new computer just in case the old one cannot be repaired? Dell, etc. has that and this is something IT in every company wants.
Yeah, Dell definitely has enterprise support all worked out, and can use the economies of scale on that to provide what are usually rock solid systems which they partially assemble themselves. If you're shipping a few 100k units of a desktop spec, it pays to make sure you can control costs be removing problematic components, and in their case, they often write their own firmware so they can quickly work around bugs.
Dell may actually be the PC provider closest to what Apple does, since I believe they source components for their own hardware in some cases (or use white label in others, but with their own firmware as I noted).
That said, I think the only thing that prevents Apple from competing in this space is desire and experience providing that service to enterprises at scale. It's not like they don't have the money or manufacturing pipeline, and lack of experience solves itself after a while.
>Things with just 1 vendor tend to be completely ignored, if they're meant to be used throughout the company.
I disagree. I work in such a company and they are always looking for one and only one vendor for their millions of PCs. It used to be Dell, now HP but they completely outsource procurement, maintenance, repairs and disposal of the computers to that one full-service provider.
Apple wouldn't be considered because they don't offer this full-spectrum service.
I think what he means with "needs more than 1 vendor" is that there exist a competitor to Dell (or now HP) that provides roughly the same service they can switch to if they are not happy with their services. With Apple the only supplier is Apple. There is no other supplier you can switch to if the relationship with Apple gets to the point that you can not work with them anymore.
This ignores the operating system from the equation. Basically there is no effective replacement for Windows or macOS (or Linux). Whichever platform you pick you are stuck with due to legacy/having to retrain hundreds (if not thousands) of workers if you want to switch.
I’ve contracted in lots of large enterprise, and software engineers who prefer (and are provisioned) Macs are everywhere. I’ve even seen a number of .NET teams who bootcamp with them. I’m contracting in a reasonably large bank right now, and most of the people in my office use MacBooks.
It’s true that large enterprise is slow and cumbersome. But they really don’t care too much if they’re buying you a MacBook, or a surface, or a Dell. Their main concern is whether they can properly manage all their devices, which isn’t too hard to do these days. Whether a large enterprise has MacBooks or not depends a lot more on whether or not they’ve figured out how to use JAMF than it does anything else.
There is no such large enterprise. You need software engineers at the very least to operate an ERP, and any organisation that doesn’t rely heavily on an ERP to operate is not a large enterprise. Any large enterprise will also have an infrastructure team, comprised of people just as likely to have strong preferences about their operating systems.
Technical and design staff may make up a relatively small portion of the headcount for some of them, but the Windows-only enterprise is becoming increasingly less common. The primary reason being that the tools they need to support other operating systems are simply more capable and accessible than they have been in the past.
I don't know in what kind of circles you mingle, but the software engineers I know that are working on ERPs are VEEEERY different to the ones that are working in FAANGS and startups. Think (Notepad++ and JDeveloper) vs (vim/emacs and cargo).
And as far as enterprise IT, there are a ton of Windows infrastructure folks, and again, there's little overlap between those and FAANG types. Plus infrastructure folks frequently hate Macs, for example (the stuff to manage fleets of Macs is vastly underpowered compared to stuff used to manage fleets of Windows PCs).
Plus even in those companies, the ERP & infra folks are maybe 1-5-10% of the workforce and they're generally in supporting roles (frequently they're temps or consultants), not in core business roles. Nobody listens to them :-))
And they generally want to get rid of them, they're seen as an expense.
I do agree that Windows-only places are becoming less common, but in big enterprises it's far from agreed that going outside the Microsoft ecosystem has a good ROI.
I think you’re being a bit judgemental and overly stereotyping people here.
In any case, the reason enterprise adopts these technologies isn’t directly because of a technical ROI. It’s because giving people the tools that they want to use is generally seen as beneficial. Providing good UX to internal users is becoming just as trendy in enterprise as DevSecOps and Service Oriented Architecture is. There’s plenty of fairly obvious business reasons why enterprise would care about these things. One of the more important ones is that these organisations are competing FAANGs and startups to hire talent. If you walked around the bank office I was working in today I doubt you’d be able to tell it apart from a typical well-funded startup.
This to me is the interesting spectator sport for M1 (I have no intention of buying Apple hardware) - what are the competition going to do in response?
Based on all the reviews and benchmarks, the new generation of Apple laptops are going to outperform anything AMD/Intel, while having twice the battery life. These are not 'small difference' only of interest to geeks, these are large user visible differences that could make everything non-Apple look inferior.
I don't see x64 suddenly getting competitive on this front any time soon, so are the 'traditional' laptop vendors (e.g. Lenovo) going to have to pivot to ARM as well? Possibly segmenting so that 'gaming' or 'workstation' are different market sectors that stay on x86. Even if they do switch there's going to be lag for development, and there's no guarantee that the software landscape will be as set up for it as on Apple.
As someone whose purchases are 'second hand thinkpad, install Linux' I'm all for it. If you pick the right distro, Linux has supported ARM well for a while.
First, I think the "rapidly" is key here. Momentum / current market share for enterprise and gaming is strongly in the x86 / Windows space.
Second, the M1 successors have to cover a lot more use cases first. There are certainly use cases where M1 is winning single core performance, and some for parallelism, and nearly absolutely in efficiency and battery life. There are still plenty of cases where software either only works on Windows, or it's where many users are currently using it, and accustomed to using it. There are other cases where "M1 for x" doesn't exist yet - high-end workstations, for example. We have rumors and expectations for this to come out in over the next two years, and assuming the competition stands still, you might start to see cases where the time savings on heavy computation favors the Apple Silicon options heavily enough to overcome inertia.
But I would suggest people look at Intel and AMD in the Windows space for a precedent. AMD multi-core performance has been dominating Intel for about 3 years now... guess who owns the market share? It's still Intel. The momentum definitely takes years to overcome. AMD is at something like 20% of Windows notebook share after about one year of having dominate mobile chips. See this - https://www.pcworld.com/article/3588154/amds-notebook-pc-sha...
> Momentum / current market share for enterprise and gaming is strongly in the x86 / Windows space.
Sure, but they're not just going to sacrifice the consumer market entirely. The industry as a whole will need to respond.
> There are other cases where "M1 for x" doesn't exist yet - high-end workstations, for example.
For now. In a year, I wouldn't bet on it.
> AMD is at something like 20% of Windows notebook share after about one year of having dominate mobile chips.
AMD beat Intel by a little in efficiency over the last year (and for the mainstream laptop market, after a certain point, that's mostly what people care about). M1 is a _dramatic_ jump forward, there. "Here's an AMD and an Intel laptop for 1k each, the AMD one is a bit faster and has slightly better battery life" is a different proposition to "Here's the same AMD laptop from earlier, plus a $1k MacBook Air. The MacBook Air's a bit faster, and has twice the battery life".
The Ryzen 4000 laptop chips outperform Intel's as much as 45% in benchmarks, while getting 33% better battery life.
The M1 is roughly competitive with the Ryzen 4000 chips (depending on variances in TDP, cooling, etc.) but arguably gets as much as double the battery life.
No argument that it's an advancement of CPU technology, but I think you might have some magical thinking if you think momentum doesn't matter.
It's not like suddenly 100% of buyers look at only benchmarks where the M1 leads, or only at battery life, or only at software that runs on the M1 rather than all the software they use.
If users were only looking at CPU benchmarks, AMD would have a lot higher market share, and the same will be said about the M1 after a year as well. Will Apple users replace Intel laptops in record numbers? Probably! But the conversion of Windows users will be slow regardless of hardware.
On the ARM front, MS launched the Surface Pro X last year that runs on an ARM chip they developed with Qualcomm. It is way behind M1 in terms of performance and battery etc but its the same kind of idea - fast ARM based ultraportable with emulation for non-arm apps. So they do at least have a 'dog in the race', even if its a slightly lame dog at this point. I imagine they will be throwing increasingly more resources at ARM in the future.
Folks here don't care, if their Business Software runs whatsoever or is whatsoever faster as long as it is cheap on a per unit basis. There is no such thing as ROI in Excel tables for controllers in my company. ;)
I think Microsoft have 5 or 6 different lines of business that bring in billions or more of revenue: windows, cloud, gaming, office/productivity and surface. They are the most diversified of the big players.