Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
What Intel software will you miss after the Arm transition?
- Thread starter Jyby
- Start date
- Sort by reaction score
They didn't talk about this during the announcement, but I am worried that transition to Arm will lead to the MacOS becoming more locked down like the iOS.
More specifically, the ability to use software that didn't come form the App Store. Most of the software I use did not come from the Mac App Store.
I am unsure if that will happen, or maybe happen down the road, but hopefully never.
The other concern I have is with software that is no longer updated by the developer.
Maybe with Rosetta II this won't be a big deal, but Apple dropped support for the original Rosetta relatively quickly, so this might happen with Rosetta II also.
Yeah Apple locking down the OS wouldn’t be great. But then I think they would lose their proudly held Unix certification etc. People couldn’t use the Mac to develop etc. I know lots of people who use Macs to do embedded development and I think Apple only uses Macs to develop its hardware. Apple products are mad with Mac.. But to your point Windows 10 S is Windows Store only
But for the Intel software do you have software you use that you are afraid will be left behind?
Last edited:
They demoed Photoshop, which isn't available on the App Store.
That worries me too, but they made the point of showing Parallels and if they dropped Virtualization support that would immediately kill such an app unless Parallels wrote their own emulator. Virtualization is a big deal for developers - I use it all of the time for testing software on old OS versions, so I can't see them killing something like that at least for several years. Kind of like how we got over a years warning before 32-bit apps got dropped.
Edit: The platform state of the union confirmed they were showing ARM Linux, not emulated x86 Linux. And they worked with Parallels to make that possible. They also confirmed (which should be obvious) that they're shipping full copies of the x86 binaries to make Rosetta possible. Which means they will want to get rid of in the future to save space. It's likely they will probably do the same thing as last time, where it's included by default initially, then optional in a later OS before they eventually remove it. I hope VMWare or Parallels comes up with a solution for x86 support and it performs decently. I know Codeweavers has Crossover (WINE) for Android that does x86->ARM translation, so I hope they port that to Mac too. But I feel bad for them after the hassle they went through last year writing their x86->x64 translation. Edit 2: Apparently I was mistaken about Crossover for Android supporting ARM. WINE can support it with QEmu but Codeweavers does not directly support ARM themselves. I suspect this will need Rosetta, and hopefully they don't abandon Mac support in a few years.
Last edited:
Don't worry Dave, Apple will bring out their own version of Fusion called "Fusions". The first version will be kinda of buggy, later versions will be really great. Then it will go into decline. And then it will be uncermoniously abandoned by Apple...
It's hard to say which software I will miss - because likely I won't be aware of the incompatibility before its too late. It's the small/indie devs with limited budgets and time that will be lost in the translation.
Will SheepShaver (MacOS9 virtualization) make the cut? How long will it take GPG Suite to migrate? What about old classics like Zoombinis (will it remain playable once Rosetta 2 support is dropped)? How about gems like TrueCrypt or VeraCrypt? SpamSieve is a must-have - will it remain? The list goes on and on. And it's completely possible that some software will work initially due to Rosetta 2, only to not work in the future once that support is dropped (i.e. when Apple deems the transition to be complete).
Will SheepShaver (MacOS9 virtualization) make the cut? How long will it take GPG Suite to migrate? What about old classics like Zoombinis (will it remain playable once Rosetta 2 support is dropped)? How about gems like TrueCrypt or VeraCrypt? SpamSieve is a must-have - will it remain? The list goes on and on. And it's completely possible that some software will work initially due to Rosetta 2, only to not work in the future once that support is dropped (i.e. when Apple deems the transition to be complete).
I used to run Windows in a VM because there were a few pieces of software I needed that were Windows-only. Then that was resolved. Now I worry about it happening again but there is so much in the Open Source and other worlds that I don't worry about it that much. I need tools for work but I'm sure that my workplace will get the tools working before migrating.
There's one notes application that I like but I think that will also get ported.
So I'm not that worried about it at this time.
There's one notes application that I like but I think that will also get ported.
So I'm not that worried about it at this time.
How are they going to allows cross-platform applications using X11 for the GUI to run on macOS, many of which are not specifically designed for macOS. This includes numerous scientific and academic software projects.
Will I be able to run command line tools from terminal (a python script for example)? Is python or Rstudio going to work at all?
Is it possible to connect to a server via Terminal and launch a non-Mac applications with its GUI?
What happens to the Unix based language?
Will I be able to run command line tools from terminal (a python script for example)? Is python or Rstudio going to work at all?
Is it possible to connect to a server via Terminal and launch a non-Mac applications with its GUI?
What happens to the Unix based language?
Active Trader Pro (ATP) from Fidelity Investments, their real-time trading tool. This is a Windows executable and up until about six months ago was still 32-bit.
It runs on macOS, packaged in a customized version of CrossOver. Fidelity retail investors were warned at Catalina's release that ATP was not compatible (at the time) and that Mac users should stay on Mojave. Eventually ATP was updated as a 64-bit Windows application and the Mac version followed a while later.
The Mac version of ATP (running in CrossOver) performs decidedly poorly compared to the same executable running natively on a Windows box regardless of whether it's Mojave or Catalina.
ATP is so well optimized for Windows, I prefer running it at home on a $170 cheapo Wintel PC rather than my Mac mini 2018 that was >10x more expensive. Likewise, I prefer running ATP on a $745 Acer notebook PC rather than my $1650 MacBook Air 2019 when I'm on the road.
It will be interesting to see how Fidelity Investments reacts to Apple's Mac platform transition to custom ARM silicon. One thing I know for sure: ATP on Wintel isn't going anywhere.
The current iPadOS Fidelity app is nothing like desktop ATP, it's closer to a dumbed down version of the website.
Naturally, one can accomplish many trading tasks using a web browser or the mobile apps (iOS, Android), etc. but the latter two aren't the same as the full-blown desktop tool. If one can play fast-action videogames in a desktop web browser, then I'm guessing that there's an avenue for ATP to make it to a browser but we haven't arrived there yet.
Active Trader Pro is probably the most glaringly obvious piece of Wintel software that I would miss at this time.
It runs on macOS, packaged in a customized version of CrossOver. Fidelity retail investors were warned at Catalina's release that ATP was not compatible (at the time) and that Mac users should stay on Mojave. Eventually ATP was updated as a 64-bit Windows application and the Mac version followed a while later.
The Mac version of ATP (running in CrossOver) performs decidedly poorly compared to the same executable running natively on a Windows box regardless of whether it's Mojave or Catalina.
ATP is so well optimized for Windows, I prefer running it at home on a $170 cheapo Wintel PC rather than my Mac mini 2018 that was >10x more expensive. Likewise, I prefer running ATP on a $745 Acer notebook PC rather than my $1650 MacBook Air 2019 when I'm on the road.
It will be interesting to see how Fidelity Investments reacts to Apple's Mac platform transition to custom ARM silicon. One thing I know for sure: ATP on Wintel isn't going anywhere.
The current iPadOS Fidelity app is nothing like desktop ATP, it's closer to a dumbed down version of the website.
Naturally, one can accomplish many trading tasks using a web browser or the mobile apps (iOS, Android), etc. but the latter two aren't the same as the full-blown desktop tool. If one can play fast-action videogames in a desktop web browser, then I'm guessing that there's an avenue for ATP to make it to a browser but we haven't arrived there yet.
Active Trader Pro is probably the most glaringly obvious piece of Wintel software that I would miss at this time.
Last edited:
None of that will change in macOS 11. They haven't removed the Terminal and macOS is still a unix based operating system. Python and X11 are open source, so something like homebrew (once updated to support the new OS) can still be used to install it.
The only thing this really changes for open source stuff is likely a need to recompile for ARM. If you already used a package manager like brew you were already compiling stuff yourself.
All of the software I miss is PPC, and already unusable. I actively stayed away from little one-off apps that would eventually become abandonware starting in 2007 or so. Seems like that was the right idea.
Motorola chips weren’t proprietary. They were just different. 68K and PowerPC chips are still in use in embedded applications (cars, ATMs, etc).
This time, Apple is designing the chip from the ground up, albeit using ARM technology as a base.
It seems obvious to me (not a developer) that Apple moved everyone to x64 because they didn’t want to translate all the legacy x86 instruction set with crud all the way back from the early 1980s. They wanted to focus on x64.
Since Codeweavers already wrote the 32-to-64 bit translator, hopefully the port to Rosetta 2 still works for a few more years. Maybe then Quicken can write a Mac app with feature parity and I can stop using Windows apps entirely.
let’s see if Homebrew will get any support on ARM at all... isn’t it cheaper and easier for people to simply switch to Windows/Linux/Ubuntu?
Also, I don’t know if it’s easy to recompile to ARM from compiled application for x86. To what extend are people willing to work on this?
There is absolutely zero reason to be concerned about the Unix and homebrew ecosystem on an ARM macOS machine. 99.999% of the software in homebrew is already ported to ARM and the only real difficulty will be for the homebrew architecture to be able to handle multiple architectures.
All the applications and tools you mention have been running on ARM Linux for about a decade now. Getting them working on ARM macOS will be trivial.
Same boat here. ARM macOS is the end of the road for me. It’s terribly sad and I dread migrating to Windows or Linux, because I really do prefer macOS.
But first and foremost I need to be able to do my job, and my job requires that I be able to build and run x86 docker containers. I need to start figuring out my exit strategy...
So I guess my answer is this. “macOS” is the software I will miss the most after the ARM transition.
Thank you for the explanation.
Hopefully we can continue to use ARM based Macs in the future without difficulties.
This is correct.
All of the Linux crappiness that I endured in the late Nineties on Wintel hardware I can still experience on ARM hardware.
Back then I ran Red Hat Linux 5.2 and 6.0 or Gentoo Linux on Intel hardware (Pentium II, Pentium III, Asus motherboards, Matrox Millenium graphics cards, etc.).
Today it's Raspbian and LibreELEC on an ARM-powered Raspberry Pi 3.
And desktop Linux still sucks rocks. Poor device drivers (especially notebook PC power management), piss poor end user documentation, and excessive system administration load. Dependency hell is still a real issue. And the desktop environment UI looks like it was designed by a 10 year old.
F***ing Raspbian can't update itself even with a live WiFi network connection; I still need to plug in Ethernet. At least with LibreELEC I can avoid most of that dependency hell by downloading a complete image image with wget then rebooting to complete the upgrade.
So yeah, all of the major *nix applications are all available on ARM.
Desktop Linux on ARM sucks just as bad as Linux on x86/i64. The same experience is pretty much guaranteed.
Trust me, I still think Linux is a great operating system for servers, ATMs and smart doorbells, etc.
I run ATP and ToS on a 2008 Dell Studio XPS now. I was running it on my 2014 MacBook Pro before the Pandemic. This 2008 Dell Studio XPS runs ATP, ToS, Slack, Firefox, Thunderbird, iTunes, VLC and performs quite well. It's amazing what you can do with some old CPUs. The CPU in my MacBook Pro is about 75% faster in both single and multicore modes but performance of ATP is comparable between the two systems. It takes a fairly long time to startup on both system but I think that's due to Fidelity server latency. ToS is similar - though ToS has these software updates that can take quite a while.
I would hope that WINE gets ported. WINE just translates the Windows API calls I believe so it should be possible. Someone would have to do the work though. Fidelity could also just decline to support macOS on ARM and tell their customers to get a Windows or Intel Mac to run it.
I find ToS to be far superior to ATP but my main trading account is with Fidelity so I use both. ATP is written in Java so it's not as performant as it could be. I used to run QuoteTracker which was a small executable that ran in a very small memory footprint. TDA bought the product and didn't keep it updated and it atrophied. They were focused on ToS. Jerry Medved wrote it in .net; it wasn't portable and it used orders of magnitude more RAM. I had to dump it as it apparently had a memory leak problem which he couldn't debug. He thought that it was an interaction between the OS and his software. I moved to ToS - steep learning curve but it solved my problems.
It would take a fair amount of work to port I think for Fidelity. My impression is that they are not that responsive to bugs and feature requests.
[automerge]1592878482[/automerge]
Linux is a great environment.
For server-side development.