xkr47 wrote:Hi Geoffrey,
Amazing work! Thanks for being patient and devoting your time for this! Been planning to buy an usb interface for some time and this lets allows me to get the bigger Scarlett versions (thinkign of 18i8 right now) instead of having to consider other brands.
Thank you, and you're welcome! I'm no audio aficionado, but I think the Scarlett interfaces are pretty cool, and the input routing that you can do with the Linux driver is really useful. Now we just need a GUI for it!
Do you have a https://patreon.com/
account or similar so we could support your awesome work?
Oh, no need for that really, but thanks for the offer. I'm just happy that people are trying it out and providing feedback. Though, if someone wants to throw a Gen 3 Scarlett my way
then I'll certainly do my best to write a driver for it!
Well spotted! That 17/Jul/2019 post is the most recent version, although there'll be another one soon as requested by Takashi. The differences from the version on github are really minor which is why I haven't asked here for people to retest yet.
I wanted to ask about the NVRAM save - which is a great idea! - it seems your driver will save settings to NVRAM after no mixer settings have been changed for 2 seconds? Is it okay to save "so often"? I am wondering if the NVRAM can wear out too soon if it is used too much.. IFF this is the case, would it be enough to save say every 5 minutes? Ok so the computer might crash etc and it would be nice to not lose settings in this case. One option could maybe be making it a module parameter (with some default value) in which case it can also be edited "live" later through /sys/ if I've understood correctly. So if your computer crashes a lot you can pick a smaller interval
Or maybe you already investigated this from the windows mixer behaviour and that's where the 2 seconds came from
Hehe, yes I'm emulating the Focusrite Control behaviour. 2 seconds after a change with no further changes, the new settings will be saved.
xkr47 wrote:Again, thanks a million for your work and good luck in the final polishing of the driver!
You're welcome, and thank you again!
xkr47 wrote:I guess the dkms driver mentioned here would also be great for a lot of people not being able to use the latest and greatest kernel just yet because of e.g. distribution limitations etc.
Yes, I agree that would be great, even for people not on the latest and greatest kernel, because it looks like the driver should make it in to Linux 5.4 which if I understand the release cycle correctly is not due until Nov/2019.