I'm still going into a docker for linux on windows dev but want to play with using WSL2 for this cycle (deploy is to linux machines / container images). Mostly just need to figure out tooling setup (pycharm / vs code)
They could in theory implement a sound device for the virtual kernel though I cannot see any information that they have actually done that from a quick search.
Some people have worked around this issue by setting up pulseaudio for windows and then connecting pulseaudio via network from WSL out to Windows. Works for both WSL1 and WSL2.
While this is a nice development, I personally still require running VMs and Hyper-V's GUI/console support is so limited I'm going to be forced to stick with VirtualBox for VMs.
Fortunately, Cygwin is fine for basic shell/utility stuff and using Cygwin also means that I can continue to use ssh-pageant (talks to putty's ssh agent) which isn't possible under WSL.
Only semi-snarky answer: The song and dance of drivers, and not having to reboot.
Plenty of mainstream PC configs are going to be a hassle to get to functional parity on the Linux side. You can complain "it's deliberately crippled hardware/bad drivers" but odds are, the guy buying the fleet PCs are not checking if they're Linux-friendly.
If I just want to run some command-line dev tools, I can probably get there faster by installing WSL than trying to start with a Ubuntu ISO and setting up the nVidia drivers.
When you start looking at what Microsoft can offer now re: Microsoft365, Dynamics365, Power Platform, Sharepoint, Azure, wsl, massive sales dept. with years of experience, etc., why would the enterprise ever go with anything else? Something tells me Azure will end up kicking Amazon's ass in the end.
19 comments
[ 3.2 ms ] story [ 61.3 ms ] threadI'm still going into a docker for linux on windows dev but want to play with using WSL2 for this cycle (deploy is to linux machines / container images). Mostly just need to figure out tooling setup (pycharm / vs code)
Interesting to see how WSL2 feels now that Linux things are actually running in separate VM.
Some people have worked around this issue by setting up pulseaudio for windows and then connecting pulseaudio via network from WSL out to Windows. Works for both WSL1 and WSL2.
Fortunately, Cygwin is fine for basic shell/utility stuff and using Cygwin also means that I can continue to use ssh-pageant (talks to putty's ssh agent) which isn't possible under WSL.
Is there any advantage over simple multibooting?
Plenty of mainstream PC configs are going to be a hassle to get to functional parity on the Linux side. You can complain "it's deliberately crippled hardware/bad drivers" but odds are, the guy buying the fleet PCs are not checking if they're Linux-friendly.
If I just want to run some command-line dev tools, I can probably get there faster by installing WSL than trying to start with a Ubuntu ISO and setting up the nVidia drivers.
Give full credit to them for the capabilities while fully documenting the limitations compared to alternative choices.
Why spend $4.000 on new mac for development if you can get $2.000 windows 10 machine that will give you same/similar/better? performance.