This is in the works, but currently held up due to dependencies on tooling further upstream.

Because we load R as a library, we need to wait for R itself to be ported to Apple silicon before RStudio can be fully ported.

Our build infrastructure will need to be updated to the latest version of Xcode to get support for cross-compiling to arm64.

The only downside of this approach is that it will make the binary enormous (as we will ultimately ship x86_64 and arm64 versions of Chromium). Thankfully it is possible to build arm64 binaries on x86_64 hardware.

We should do this as a universal binary, supporting x86_64 and arm64 without requiring a separate download.

This issue tracks the work needed to add native support for Apple's new ARM-based silicon to RStudio. And just to give a little context, I'm on the M1 MBA with 16GB RAM.R and RStudio work today on the M1 chip, albeit with compromised performance, via Rosetta2, an emulation layer.

Again, I use it for very basic CRUD requests, so that's all I've tested it with. In terms of CPU, I've allocated four cores to the VM, but I haven't seen it use any more than 50% of one core. The VM consumes ~5GB RAM of the 8GB I've allocated to it. Performance-wise, I've found it to be quite snappy (to be fair, I've just tested the basics).

The install script ( here) from the vagrant project was quite helpful in installing and configuring the DB. I used keyboard to navigate the OL8 installer, and selected the 'server' installation mode with headless packages. From what I've read, hardware acceleration doesn't work either, so my recommendation is to stay away from the UI as much as you can. Mouse didn't work for me, so the GUI is pretty much useless. The setup, configuration and installation is completely manual and will take a significant chunk of time, but the outcome is pretty satisfactory. I've had success running Oracle Linux 8 (x86_64) with UTM, and then installing Oracle DB 21c EE in the VM. I've been struggling to find a solution to this for the past couple of days, and finally have a halfway-decent solution.

Rstudio for mac m1