Glad to read that you seem to have found a usable setup now.
Yeah, it’s a pity that Focusrite as the hardware vendor doesn’t care about Linux users. I suggest to complain to them, maybe at some point in the future they will start listening and providing their tool for Linux as well…
In case you still want to be able to use the Scarlett, there are still two ways to set up the routing as you want it:
Use alsamixer on Linux (it’s CLI and not perfectly intuitive, but it’s “complete” in the sense that you can set up true stereo on the headphone outputs, I did that on the 18i8 back in the day), or
Configure and safe the settings using the official app on Win or Mac (don’t know if Wine might be a solution here)… AFAIR, the Scarlett will remember all routings set up if booting to Linux afterwards.
Focusrite provided Geoffrey Bennett some support developing “alsa-scarlett-gui” after seeing his GoFundMe campaign was nearing its target and the program was progressing towards completion.
As I mentioned above I was a happy Focusrite user myself for a bunch of years. Yes, they are providing their devices as class-compliant audio devices which allows driver support to be implemented rather easily (mostly out of the box). But as a Linux user you are still second class citizen, as you need the support of volunteers implementing the drivers or providing third-part configuration tools. I still might have outdated information here, but all the Scarlett driver support on Linux was provided by people who were not employed by Focusrite in their free time. I would be happy to learn if I am wrong here and if this has changed.
Sorry if this is sounding harsh, but in my opinion it’s the hardware manufacturer who should provide support. Funding third-party devs is a great move, I didn’t know that, but it still wouldn’t have been necessary in the first place if they would just release their devices with first class support. (That would includes driver support + the availability of the configuration tool from day 1).
I think you are correct that none of the driver support built into the kernel was completed by anyone employed at Focusrite. However, them providing some amount of technical and financial support to volunteers is more than most manufacturers do. If they determined it would be profitable to officially support Linux, I think they would do it. They have likely determined the costs exceed the projected profits and have chosen instead to make some modest investments in the volunteer efforts, making their devices a product of consideration in the niche market of Linux audio without taking on the overhead of official support. If that is true, you can’t really fault them for doing what makes sense. No successful business is going to choose to develop a product that is projected to lose money.
But you can still ask them if they would reconsider releasing their configuration tool also for Linux. If enough users do this, maybe their projection tipps over at some point…
There was a developer a while back who released a Linux version of their audio plugin and promptly discontinued it a version or two later. Their reasoning was Linux downloads accounted for less than 5% of their revenue and over 50% of their support tickets. I know this is anecdotal given I don’t remember the details, but it rings true, doesn’t it?
While it never hurts to ask, I doubt even a 1000 emails from different users saying “I promise I will buy your device if you support Linux” would affect their calculations.
Porting the software has its own challenges, i.e. which distribution do you support? Of those 1000 people promising to buy a device, how many use that particular distro? If it can be packaged to be distro-agnostic, does anyone on your current team of developers know how to do that? Does this mean Focusrite would need to take over the kernel development for their drivers (I would think so)? If so, is there anyone employed there that knows anything about kernel development? Does your technical support staff know anything about Linux desktop usage, Linux audio configuration, and the differences in distributions, i.e. this one uses Pipewire version X, this one uses Pipewire version Y, this one doesn’t use Pipewire, this one uses an “audio” group, this one uses a “pipewire” group, this one uses a “realtime” group, this one uses KDE, this one uses GNOME, etc.? How many people who understand this stuff are going to have to be added to the payroll at Focusrite to develop the software and provide email/phone support to get their devices working as well on Linux as they do on Mac and Windows? How many resources are going to be diverted from Mac and Windows support where the bulk of the market resides to cater to this small group of users?
I have no direct insight on any of this, but from my observations over the years watching Linux users struggle to get native Linux audio applications working properly within a distribution, let alone across distributions, I can easily envision the support costs exceeding the money made from pushing a 1000 extra units each year.
Was also very recent as well (see a few posts above) where they donated money, hardware and engineering resources to Geoffrey. I can’t speak to how effective that is/was nut it definitely isn’t nothing.