Linux is reaching nearly 4.5% market share, which is a substantial amount of people. The platform is growing steadily and it would a smart move for Neural to be early in having full support. And there’s tons of people that want to switch to linux but haven’t yet for one reason or another. (Linux revolution will happen eventually )
I’ve not yet managed to get the QC working as an interface on linux. Using a VM running Ubuntu Studio. Only a glitchy mess, pure noise in fact.
It would be great to have Cortex Control on Linux. That being said, the QC is Class Compliant and will work on Linux without any additional driver installation, of course sans Cortex Control.
Same for me. I won’t buy your stuf unless it has Linux support.
What do you guys get for giving microsoft and apple exclusivity? It’s 2024 and portable code is an easy thing.
I’m a fan of Linux and I wouldn’t mind seeing support for it. However, currently it is only 4% of the market and that is an all-time high. Easy to see why many apps and devices don’t support it. Maybe if Linux ever grabs more of the OS market share more companies will consider supporting it.
Maybe if Linux ever grabs more of the OS market share more companies will consider supporting it.
Chicken and egg problem ? (also 4% is the Desktop market share, if you consider cellphones, iot or datacenters, then Linux becomes overwhelming)
It just isn’t financially viable to invest manpower in Linux support
Writing cross platform code has no reason to be more expensive than writing non portable code.
Especially for smaller companies.
https://www.modartt.com/ ← tiny company with Linux support. Proves it can be done.
I use their piano simulator daily on a circa 2012 Imac running ubuntu 24.04
(along Ardour DAW https://ardour.org/ Hydrogen drum machine and sooperlooper looper)
Apple would just tell me to dump the hardware and purchase newer one. Linux is good for climate as well as for your wallet.
I also run the same software suite on a raspberry pi as a portable setup that I carry to gigs.
Try to do that with windows or osx
In my opinion, windows and osx dominance is the result of a combination of lazyness, ignorance and lobbying. But what do I know…
It will always be more expensive , due to the very fact that it is cross-platform… It is not about writing code in itself, it is about testing code (i’m not talking about unit test or integration test) and maintaining code on the target platform… the more platform you have to test, the more time you need (or a bigger team) in the end it cost more to the company than focusing on the two most commonly used platforms for end-users…
But the most important thing is that a Driver is not a ‘Cross Platform-able’ thing. It is the opposite : a very platform specific piece of software that act as a ‘bridge’ between an OS low level architecture and an application running on this OS…
Thus, a driver is not a ‘traditional’ software piece of code than can be coded in a ‘cross platform’ way… Most drivers are coded in C or C++ , two of the most ‘portable’ efficient programming languages, but still : a driver often need low-level assembly code or platform specific sections that are not portable anyway… In the end, the C/C++ ‘portable’ part is only used for the most trivial sections of the driver… and again, it is more costy to develop/test/maintain a driver for 3 platforms than 2 platforms …
So, in the end, it would cost NDSP a good amount of money to support Linux, for no real return in investment. This is as simple as that.
It will always be more expensive , due to the very fact that it is cross-platform…
You’re right, doing something is always more expensive than doing nothing.
But the most important thing is that a Driver is not a ‘Cross Platform-able’ thing.
That’s the wole point of a driver, making the code that sits above it portable.
The fact that the audio functionnality of the Cortex, which uses a USB class compliant driver is working on Linux proves my point (and maybe the company hasn’t spent that much ressources in testing it on Linux and it is nevertheless working).
So, for Cortex Control, what are we talking about ? midi over USB, serial over USB ? No need to dable in C++ let alone assembly for that: kernel developers have done it for you. In fact, all that would be needed is the description of the protocol between the cortex and its host computer and someone (me?) would be able to write an app. I am under the impression the main challenge in our case could be the graphical toolkit. This could be sidestepped in a first time by writing a command line app.
I think that NDSP has enough “construction areas”. Lots of people complain about missing functionality on the QC.
And it seems Control is not stable on some Apple computers.
I think Linux support for Control would be nice in the future. But maybe it’s not time for it yet.
I’d like to believe that Linux could happen. So weird that companies still don’t try to go cross-platform-first in 2024. That sounds so old fashion to just favor one OS over another.
Linux market share keeps on rising, the only thing holding me back is no Cortex control and NDSP plugins, all the other stuff I can live without or have native support, if NDSP jumps on the wagon I’ll be switching in a matter of weeks!
Still would be awesome to have Linux support. Cortex Control installs just fine under Bottles (Wine), but it does not recognize the USB device. Yes, Linux is just at about 4% (which is still around 60 million systems), but Linux is also closer to Mac than Windows, which should make adding support a bit less complicated. And as both Mac and Windows are already supported, cross-platform complexity is likely not the issue.
You could also opt to open source this and ask the community for help, hack I’ll pitch in some of my own time as well, also Linux usage among developers is at 28%. Or start a crowd funding, I’ll pitch in 100 euros to start.
This is literally the only program I have no replacement for to move away completely from Windows to Linux.
I’ve been using Linux systems both personally since 1995, and professionally since maybe 2001, and I agree with you that open sourcing + providing a Linux build especially for a device like the QC which itself is actually running (a pretty old version of) Linux would be a great move.
As you mentioned the biased figures of OSes usage, they always misses the fact Linux is ubiquitous in the server/cloud world and pretty well placed in the embedded world (as for example we could say the QC), not event speaking about supercomputers…
Just to say that I fully agree with you about a potential move to OSS, nevertheless I know that companies are often reluctant to open source part of what they consider their “secrets”. The thing is the secret is both the software and the way the software is written ;-), tested, etc, and sometimes there are actual secrets.
And in this case, and in order to avoid endless discussions about how Linux is represented among the end-users, I prefer to propose a move to a web-based interface! For the company it has numerous pros:
A unique code base for Windows, Mac and Linux (and potentially others, I do not forget our BSD derivative friends).
Of course more users happy and a new axis of communication
No users using old versions, a much better control over UX
Core company secrets actually kept
I know companies especially in the music world who have done this bold move (Singular Sound did it with their BeatBuddy few years ago… Now the BeatBuddy editor is fully web-based and works reliably on ALL platforms). Now with the current web standards and browsers this works perfectly, and could even work offline.
I just drop that there ;-), in the hope someone from Neural DSP reads it one day.
Having the editor as web based would suck so much. I understand why you would want that, but it’s way less practical. Anyone that tours wouldn’t want to deal with a web based editor. You can’t always have access to great internet on tour.
I know a big plus for FOH engineers is the ability to edit their boards offline on their laptop. If the software for their boards was stuck to having internet access I know it wouldn’t be popular. Same thing would apply for editing the QC.
The push for programs like Pro Tools to always be online to check licenses, etc also sucks a ton. I really hope companies stop the reliance on constant internet access for editors or licensing for their programs.
I am totally against programs constantly requiring an unneeded internet access, on that I fully agree with you. I don’t know about pro-tools move or actually any windows/mac desktop software network/login requirements as I moved away from those toxic ecosystems decades ago, and for me Pro Tools is something called Ardour, an open source alternative available to all systems, that I use for years now (by the simple fact you mention Pro Tools, I know you are not a Linux user… at least not for that).
But nevertheless, even though I would always prefer a standalone desktop app, and will not argue that a webapp is better than a desktop app, yet modern webapps can nowadays provide a desktop-like experience (honestly sometimes indistinguishable), and when developed correctly can even work offline. There are even levels of “webitude” for applications from pure web-app to let’s says electron-based webapps. The fact an application needs a constant network connection is not (only) tied to the “type” of application but more surely to the way it is designed and developed…
Depending on the underlying development framework/language used by NDSP, porting to a new platform (Linux) could be more or less costly (for example if written in Java it would be almost nothing…) and NDSP will for sure choose it’s strategy (if any will to fulfill that request) based on it’s original code base…
)
, an open source alternative available to all systems, that I use for years now (by the simple fact you mention Pro Tools, I know you are not a Linux user… at least not for that).