Hello, and my first words are : Happy New Year 2020 from France to all the VMware folks.
Now are my 2 questions : a) is there a release of VMware Fusion that allows me to see and use Thunderbolt hard drives connected to my iMac27 from a Linux or Windows VM as I do with USB attached devices ?
b) Further, what is the minimum version of OS/X that supports such a VMware Fusion release if it ever exists ?
Thank you for the response.
BTW, is there any plan to develop such an interface since Thunderbolt devices are far to be dead in the Apple world ?
Fusion doesn't comment on future features, so no way to know.
Virtualizing thunderbolt is more complicated and USB, and the number of actual devices is still pretty low, so I'd guess that it's not a high priority item. There's tons of USB-C, but not many true TB3 devices.
Apple is kicking third-party apps (even Fusion) out of the kernel (by deprecating the Kernel Extensions feature in later releases of Monterey and beyond), so it probably will never happen unless Apple releases some APIs for the Fusion team to use --(like it did with hypervisor.framework and vmnet.framework). If it Apple starts missing its numbers bad enough and wants to sell some of those US$129 Thunderbolt 4 cables (for 1.8m), it had better get started on that stat..
Well, fusion would have to build support as well, and given that we don't even have real tools for Windows yet, I'd guess that's way down the priority list - even if Apple provided the API's.
I'd speculate Fusion's days are ordinally numbered backwards with an 8-bit integer because the market is flooding with freeware solutions built using SwiftUI on Apple's (vastly inferior) hypervisor and vmnet (with broken bridged networking) framework (or VirtualBox's user mode bridged networking system) and paying m-series MBP customers will probably be able to buy a version of Windows 11 ARM directly from Microsoft that installs into macOS with a single click, before too long (only question is if Microsoft will roll their own UX or if they will bail/buy-out Parallels and augment it with official drivers).
Remember that Fusion is part of the larger ecosystem, and unlike Parallels, doesn't have to survive as a standalone business.
That said, I do worry about the investment level given the slow pace of development.
I could see Microsoft eventually providing Win 11 ARM running natively on ARM SystemReady compliant platforms - it runs already. I am not as certain as you are that the work will be done to bring a native-booting Apple Silicon version of Windows 11 to market. ARM PCs and Apple Silicon Macs are a drop in the bucket in the grand scheme of things for Microsoft. What incentive does Microsoft have to provide a native implementation on Apple Silicon? Certainly not volume of Macs? And how are those ARM PCs selling?
You could also make the argument that Apple has no incentive to do this either - it hurts their platform and developers by making the Mac "just another interchangeable Windows PC".
The other implications in your argument are definitely more troublesome to VMware. Both Apple and Microsoft are forcing the hand on using their OS-native hypervisor capabilities. Apple by deprecating kexts and forcing use of their hypervisor and Microsoft by making functionality such as Memory Integrity and WSL2 reliant on Hyper-V technology. You have to build to those capabilities if you want to run on those platforms. VMware seems to have the mindset that everything has to be compatible with ESXi. That's fine for VMware but maybe not so good for people wanting to run virtualization on the desktop.
With Hyper-V being free and WSL2 being released, VMware Workstation is even under attack today. When Apple improves their native hypervisor stack (and I believe they will), I unfortunately agree that Fusion and Parallels will be non-issues in the marketplace.
Apple sells about 6 million MacBooks per quarter (maybe a bit high right now because of the Intel transition), which is a significant market for third-party developers, even Microsoft. Apple and Microsoft are cross-licensees, investors, friends. Not sure if Apple would give Microsoft a break on the 30% cut to market/distribute a package-installer in their Mac App Store, but Microsoft certainly could sell off Microsoft.com directly to Apple Silicon Users. I'm thinking Microsoft would purchase Parallels to avoid having to run a project to create the instrumentation that surrounds the operating system and hypervisor. The only way I see Apple balking is if they perceive this as a threat to adoption of Metal, but at the same time, Apple has not had a tremendous amount of success attracting game studios to Metal anyway.
I'm going off-topic to the original post - to which I apologize.
I consider Apple and Microsoft as "frenemies" - they'll collaborate when they feel it benefits both of them, and compete when they don't. Yes, they cross license and invest.
Microsoft "could" do a lot of things. To date they have not made any move to easily provide or support Windows 11 ARM on anything but Surface and OEM ARM PCs. Ball's in Microsoft's court.
Yes, 6M Macs per quarter may be sold, that does not mean that third party developers providing their software on the Mac platform feel it's as profitable as coding for Windows. Last quarter that "6M Macs" represents less than 10% of the total shipments - Windows being the rest. "Show me the money".
Perhaps this is why Apple is making their push for macOS compatibility with iOS applications and their switch to Apple Silicon - where the economies of scale favor the Apple ecosystem and where developers can leverage them by developing both for macOS and iOS/iPadOS.
Since when has Apple been seriously focused on the gaming developer?
Yup, all about scale. Is metal technically impressive? hell yes. Is there sufficient market to justify porting AAA leading edge games to a niche platform? Nope. That's why I run shadow.tech for gaming - especially since virtualization is actually worse on M1 than it was on Intel (at least on Intel you could run an eGPU).
I suspect we won't see Microsoft make it any easier to virtualize windows, and have zero expectation they'll ever make it bootable - they don't even have decent support for it on their own hardware!
As for the apple framework, my biggest fear is that apple does enough to kill real solutions, but not enough to make it viable for real workloads.