One year after our Kickstarter ended, we're still hard at work at Console.
First, let's take a look at some successes we've had. Hundreds of thousands of people have installed Console OS KitKat on PCs and 2-in-1's. We've had a massive uptake in support for 2-in-1's with Intel Core processors, which account for the vast majority of users.
In addition to that, we have the better part of a hundred thousand registered users. People willing to take action when we need them to. We know you want us to start taking actions that leverage that kind of following. And we will, check the end of this update to see how.
In all, not bad for a startup that raised (and has run) for over a year, on less than a hundred grand.
But we've also met some challenges. While we met our goal to ship Console OS KitKat in December on most targeted devices, the Lollipop Logjam™ has been our toughest challenge yet. Lollipop on Intel never formally supported our target devices fully, and worse, Intel never finished KitKat drivers. Support for Android on Core processors remains in limbo at Intel. And we've had to take over the bootloader and rewrite it from scratch to pass Google Verified Boot restrictions - restrictions that our partners were barred by Google from sharing with us during our Kickstarter campaign.
Read on to see how we're addressing those challenges, and what the road ahead holds for us.
Since Windows 10 just shipped - we wanted to take a second and address this first.
Windows 10 is out, and yes, it is safe to install Console OS KitKat builds before or after Windows 10. The two won’t impact one-another, regardless of if you use the Microsoft-offered USB/ISO clean installer, or upgrade in-place via Windows Update.
An Update on Android Security
With the rise of the Android Stagefright bug, we want to let users know we take this stuff seriously, even on betas. Our testing shows that Console OS KitKat is not affected by these issues. This is particularly because Console OS does not run directly on a cellular network, so it cannot receive MMS files automatically.
That said, we will evaluate Google’s new monthly AOSP security fixes. When Google releases them, we will examine and port in the payload if it provides a security benefit going forward.
A New Name
Console OS, or ConsoleOS? We know, we get that a lot too. It doesn’t help that hashtags create chaos there too. In one social media post alone, we often refer to it as Console OS, and close with a #ConsoleOS. It also doesn’t help that some companies have started smashing OS into the brand now. It also doesn’t help that Chrome OS and Mac OS shrugged this off, whereas SteamOS and watchOS don’t.
Going forward, we’re going to put this to rest. We plan to rename things to ConsoleOS. We’re going to get rid of the space, that’s all. By branding things ConsoleOS, it will be easier to search for online, and in social media.
While we realize this will take some getting used to, we will update documentation and even the Kickstarter site. This change will take effect when we ship Lollipop source code.
Speaking of Source Code…
We have started the process for open-sourcing, well, just about everything. The entire Console OS stack will be open-sourced and serve as a superset for the Android-IA kernel - with over 30 PCs and PC tablets staged out of the gate.
Now, here’s the catch - most of our target devices still have issues. We have all the necessary building blocks and drivers (finally!) but there are last-mile bugs to squash. We've gone into great detail on some of these topics in the Kickstarter comments. After we open-source the stack - that discussion will move to our Wiki and Forums, both of which we have received performance upgrades (and now load twice as fast).
We will be wrapping development in the open, along with AOSP builders and the AOSP community. This will include a public rewrite of our installer and bootloader (formerly called IAGO), which we have taken over from Intel and will be maintaining under a new name.
We’re going to put your Kickstarter dollars to work, offering incentives and bounties to AOSP builders in the community, to help us finish the fight. That process should go live late this month - if all goes well. We’re hard at work on it. We will offer nightly builds for device targets, just as soon as we feel it’s viable to do.
Finally, we're happy to report that our analysis of Android M is that it will be a much less painful transition - what we're doing today, will pay it forward to Android M when it (and drivers for it) become available. Unlike the massive restructuring in KitKat to Lollipop, Lollipop to Android M is a more progressive improvement.
Finally, An Update on Core
Intel has done something really unique over the past 18 months. In fact, we don’t think this really has ever happened before.
Over the past 18 months, Intel has shipped three totally unique generations Intel Core processors, each with new generations of Intel graphics technology. Refreshed 4th Generation “Haswell” - 5th Generation “Broadwell” - and now 6th Generation “Skylake” processors have all debuted over the past year and a half. Each have a different GPU generation (Gen7, Gen8, and Gen9, respectively). Typically, Intel’s “tick-tock” cycle yields a launch every 9-12 months.
But in the past 18 months, some chips have been released with as little cadence as three months (for example, Broadwell-DT to Skylake-DT was June to August, not even 90 days). Now you can see why we haven’t even tried to bring Console OS to Broadwell. We have to skip it for Skylake to stay alive, with the resources that we have.
6th Generation Core (formerly "Skylake") is a big deal for us. For the first time, Intel is unified around a single architecture. The 14 nanometer, 9th Generation Graphics silicon in Skylake will jointly run the gambit - from smartphones (via Atom derivatives), all the way to supercomputers. Until now, Atom and Core have been on different tracks - and that’s finally coming to an end somewhat. This is really, really good news for us - because Atom and Core are now cut from the same cloth, and ideally, the same drivers. Ideally.
As we have mentioned in the past, Android on Core development has been frozen at Intel. There is no Lollipop driver set for Haswell, or Broadwell. With hundreds of thousands of downloads of Console OS - we think Android on Core’s demand is quite proven.
So here’s what we’re going to do. Over the past few months, we’ve prepared our case for bringing Android back to Core. Things at Intel are quite busy - the Intel Developer Forum is a week away. We're going to be there all next week for meetings.
We’re going to wait for the big show to end, and then we’re going to make our case internally. We’re going to present to Intel the compelling 6th Generation Core products that we’ve articulated - products we think you really, really would enjoy. Products we think you would want to buy, and products that return Android on Core processors to mainstream development.
And, we’re going to ask for something simple - an up or down vote from Intel to enable us to make these awesome products using Android and Core happen.
If yes, we’ll hammer out the details, the roadmap, and the money, and get this stuff in motion. And if the answer next month is no, we’re going to present to you what we have brewed up - our vision and those articulated solutions. If it comes to that, we'll turn to you, and we’ll ask you to make the case for us.
But until then, and we are now getting daily emails and tweets - we ask for a bit of diplomacy and patience. We appreciate that people are emailing, posting on social media, and want to see Android come back to Core. You backed us because you believe in that - so hang tight while we make that case one… last… time.