Intel shows budget Android phone powering big-screen Linux Intel is showing what it calls "Big Screen Experience" at Mobile World Congress, an Android smartphone which runs a full Linux desktop when plugged into an external display. The concept is broadly similar to Microsoft's Continuum for Windows 10 Mobile, but whereas Continuum devices are towards the high end, Intel's project is …
"The concept is broadly similar to The Motorola Atrix - 2011."
along with a few other Motorola devices that could do the sane thing like the Razr. mine is sitting on my lapdock as we speak. Hardware now at a point its more realistic along with "the Cloud :-( "
I want an ASUS padfone with 4.7" (MAX as with the tablet dock who needs a bigger phone screen!!!)) screen multi core (4 or 8) cpu INTEL with a 10"+ screen tablet dock, attachable keyboard and a desktop dock. with a HUGE battery in the phone (2 days life in standard use MINIMUM 5000+mAh?) multiple USB-C connectors. and a 3,5 Audio Jack. + Micro SD (SDXC) card support for 512GB+ cards
running Windows 10 or Linux/Android combo. or Pure linux (user install options)
This post has been deleted by its author
Been using the Android phone with a wired connection to the bigscreen and Bluetooth keyboard for several years now. VnC to remote desktop provides DaaS on any OS. Online office apps work fine. With screen casting it went wireless in 2013. Hoping for the 4K Chromecast now. A nuisance to answer the phone though.
Are the Intel chips with Imagination Technologies GPUs still crippled on Linux?
Yeah, agree, MS has Continuum ready to roll. This Linux thing is just a lab demo... and doesn't look all that great based on the picture.
I see what Intel is doing here. Trying to entice the smartphone crowd to come back to Chipzilla instead of ARM by showing them that they will need the CPU power to compete with MS after MS, finally, brings out a proper smartphone (which it appears they will do this year with Surface Phone) and has Continuum out there. I think Intel is playing with fire though. If Redmond decides that Intel is no longer their preferred CPU and starts working with ARM more extensively and maybe IBM Power (which IBM has been begging MS to do forever), that hurts Intel a lot more than this hurts MS.... MS, internally with Azure etc, is one of the largest CPU buyers in the world. Forget about the ecosystem, they could just move a significant amount of their internal CPU to something other than Intel and Intel would feel the pinch. It is Win-tel (Win before tel) for a reason.
"If Redmond decides that Intel is no longer their preferred CPU and starts working with ARM more extensively"
Redmond don't have enough toes left to pull that off. Arm negates most of Windows advantage (the huge range of binary windows x86 software). Primarily OSS/Free sotware OSs can pull this sort of thing off as they are set up to recompile the stack often, Windows hasn't and still isn't.
The concept is broadly similar to Ubuntu Edge!
Ubuntu screwed up everything by asking for an unrealistic amount of money for their crowdfunding campaign. Subsequent smartphones have gotten off the ground with less than $32 million in crowdfunding!
> Not really, there's no MS Office suite for Linux
This is about 'mobile'. There is MS Office for Android, which is said to be better than the one for Windows Phone 8.x.
There are many Office suites for Android and/or Linux including WPS, LibreOffice, Google Office. If you require MS Office then use whatever works for you. Others don't need it or want it.
"Linux IS the biggest OS in the world already (just not currently on the desktop)"
Linux is a kernel. It forms part of many operating systems. As a collection, yes it's popular due to it being free. No single operating system (bar Android) is deployed at any real scale, and very few come with the support people expect (and exception would be RHEL, Android support is an utter abomination).
"No single operating system (bar Android) is deployed at any real scale,"
Well one of the challenges with Linux has always been fragmentation, which is something Red Hat obviously want to address :(
On the other hand, if you were to be slightly generous and count all deployed instances of Linux+Busybox as one flavour of Linux (which isn't a particularly unfair thing to do), you'd come up wth the same conclusion - Linux+Busybox already is the biggest OS in the world.
No offence intended to (and indeed, many thanks to) the good people of GNU etc, without whom this picture would be rather different.
> Well one of the challenges with Linux has always been fragmentation,
The same 'problem' can be aimed at, say, the car industry. Dozens of manufacturers (there used to be hundreds) making many different models that are changed each year, and then there are all the options and different colours. It is an absolute nightmare having to choose one, and even worse if you want replacement parts if something breaks.
Just think how much better it would be if there was just one engine/chassis and just a handful of different bodies built on that. Everything would just fit regardless of which model you bought.
And yet people seem happy with the current situation.
>No single operating system (bar Android) is deployed at any real scale, and very few come with the support people expect (and exception would be RHEL, Android support is an utter abomination).
Linux totally owns supercomputing, with 98.8% of the top 500 running it. It totally owns cloud - over 60% of the virtual machines in Microsoft's own Azure cloud are Linux, and almost all of every other cloud. It is the boss of public web hosting. It is used in the server room quite a lot more than you would think. And of course it runs every server at Google - who account for a significant share of all global server use. All of these things are at global scale.
As for support, the level of support Microsoft offers is legendary for its futility. You might as well call Dell. Unless you mean patch support in which case it is well known Microsoft abandons their platforms on a regular basis for their own business reasons, includes patches you don't want like GWX, issues patches that brick your machine terrifyingly often. They make most Linux distributions' swift and regular patch issuance seem absolutely boring.
"Linux is a kernel. It forms part of many operating systems"
Yes, but "no kernel, no operating system".
Weather it's Linux+Hurd, Linux+Something else, it's still Linux in my book.
Most of the userland stuff making up Unix-like OSs come from the same root philosophically, and in many cases from the same programmers.
I'm not concerned so much with the vener on top, but more with the bolts and nuts underneath.
I think you assume that unless you can see a GUI, it's not real. All servers with virtualised Linux environments are real (as real as running software can be).
Or are you saying that Ubuntu, Mint and whatever server versions are used are not Linux because thay have minor differences?
In that case Windows 2000, NT, XP, Vista, 7, 8 and 10 are not Windows. Not to mention Windows 3, 95 and 98. They are all different OSs.
In that case Windows 2000, NT, XP, Vista, 7, 8 and 10 are not Windows. Not to mention Windows 3, 95 and 98. They are all different OSs.
Windows < v4 is not an OS, but a co-operative multitasking desktop atop DOS.
Windows v4, aka "Chicago" or Windows 95 was an OS.
Windows NT is a different OS, that formed the basis of Windows 2000 and everything newer in the desktop Windows world.
Then there's Windows CE, another different OS, that deserves to be forgotten.
Yes, Ubuntu and Mint are different OSes. Forked from the same code base, and closely related, but they are different, in the same way that identical twins are still individuals, in spite of their similarities.
> Yes, Ubuntu and Mint are different OSes.
By the same means then Windows NT 3.1, 3.5, 3.5 Server, 4, 2000, Server 2000, XP Home, XP Starter, XP Pro, XP Ultimate, Server 2003, ... are all different OSes. By your measure then XP SP1, SP2, SP3 (multiplied by Starter, Home, Netbook, Pro, Ultimate) would also be 15 different OSes.
This is what I want, basically a PC in a mobile phone form factor, with a master OS & slaved radio phone system, that I can wipe completely & load a new OS at will
If I can run some flavour of L*nux, completely open sourced, then I don't need or want any apps
But no one wants this, there's no demand, say the manufacturers
But no one wants this, there's no demand, say the manufacturers
Two reasons for that. Reason one is that only the technically proficient will want to use something like that. The market would possibly sustain one or two models at best but not enough to satisfy the corporate greed involved.
Reason two is because they say so.
" Reason one is that only the technically proficient will want to use something like that. The market would possibly sustain one or two models at best but not enough to satisfy the corporate greed involved. " ---
Indeed, parent basically described the old n900, which was a Linux PC with a phone slaved to it. I loved mine, lots of nerds loved theirs (so much they tried to resurrect it with the neo900), but the wider world went "meh", until Apple came along with the iPhone.
What we want, is not what the public wants. The n900 didn't even manage to sustain one model in the market, let alone more. And that was back in 2009, when the competition wasn't as fierce as now, and the "app market" was still not totally captive by Apple/Google.
Maybe with persistence, marketing and refinement it could have been number 3 in the mobile OS options list, but Nokia couldn't financially sustain it, and after Elop got a hold of Nokia, a Linux based phone had no chance of surviving.
Now, I think the best we can hope for is some sort of hybrid like this. Still not sure of the security implications. There are so many apps on my phone, and I don't trust any of them not to be buggy or malicious, that I refrain from logging into sensitive places.
I have actually taken to carrying a second phone, running Cyanogenmod without any apps just for SSH and other sensitive stuff.
At this point I have been pondering starting an OSS project to take AOSP, or Cyanogen mod, and rip out all the stuff down to the bare essentials to run the phone and wifi, then build a GNU Linux distro on top of it. No Apps, no Android compatibility, but as close to a Linux OS as you can get, something akin to my old N900, or if I can't get the phone bits to work, my old N810.
What you mean like the Blackberry Priv? large screen, secured Android, physical keyboard, or if you are slightly more brave and aren't a slave to Candy crush etc the Passport is an excellent piece of kit. I have one and can't right now think of a phone other than potentially a PRIV that Id swap for and even that I can't see the cost/benefit bit making any sense for me as the PP does everything I need.
>Why bother with the Android part? Dump google and run the phone with Linux.
FFS! I've used a Linux desktop application on an Android phone (Inkscape) and it is a horrible experience. It doesn't matter how good the underlying OS is, if the UI is unfit for the Human Input method being used, it will be an exercise in frustration. Install it now if you don't believe me.
UIs are important. Those proponents of Linux who don't acknowledge that fact won't do their cause any favours. So, if you really want Linux to do well, promote good UI design. Here's the thing though: it is time consuming to get right.
"UIs are important."
Quite. The best approach seems to be that undocked you have a phone interface, docked you have a desktop interface. One of the things about Unix-based systems is that the UI is an additional layer on top of most of the rest of the system and the interface between the layers is clean enough to swap UIs as needed. Of course if you then try to run an application that needs the desktop interface when undocked you're in a hole of your own digging.
Having said that Ubuntu decided that what they really needed was an app-centric interface on the desktop to prepare the way for use on the phone and got it out even ahead of W8. I don't think it's proved as popular as the more traditional desktops.
>If I can run some flavour of L*nux, completely open sourced, then I don't need or want any apps
Can you expand upon that? I can't work out what you actually want to use your phone for. Take away all the applications, and you'll have nothing. No dialler, no SMS client, no gallery, no browser...
I have a nicely polished pebble I found. I'll send it to you. No charge. It is 100% secure and quite ergonomic, though it can ruin the lines of a lighter jacket.
Dump google and run the phone with Linux.
That's what Ubuntu are working towards with their 'convergence' thing link to ubuntu.com (the thing they originally tried to crowdfund under the name of 'Edge' -- link to indiegogo.com -- until some struggling desktop wannabe pinched the name for its browser).
I'm not sure that I like Ubuntu phone enough to consider that development interesting, though, it seems to me to be geared to much toward social media.
>Ubuntu phones have been available for a while now - don't they do this already?
But why? If to make this system work you need the bulk of a wireless keyboard and mouse, you might as well carry a stick-shaped Linux computer. That way, you can do your work on a big screen, but also take a telephone call. Or just grab your phone as you nip out to the pub for half and hour.
This 2-in-1 desktop/phone system seems like a lot of kerfuffle just to save on the cost of an SoC in a plastic case.
Ubuntu have been proposing this concept for a while. Microsoft have played with it. Meanwhile, many people just use device-independent services such as Gmail - where an email I start writing on my phone I can finish on my laptop - and it is in this direction that Apple have moved (Yeah, I know that it is in Apple's interests to sell you both a Mac and an iPhone).
Heck, I had a Sony phone with a real microHDMI socket on it. Grand. But it was nowhere near as convenient or flexible as using a Chromecast to display content on a big TV.
"But why? If to make this system work you need the bulk of a wireless keyboard and mouse, you might as well carry a stick-shaped Linux computer."
Different use cases I suppose:
1. Keep the big peripherals at home and use the phone as a phone elsewhere.
2. Carry a keyboard & mouse in luggage. Trade-off keyboard size vs convenience* to personal taste.
3. Hot desking - a keyboard, mouse & monitor will be available in a remote office & carrying a phone is more convenient than a laptop.
There are probably takers for each of these. You might not be one of them, it doesn't mean everyone has to follow you.
*I used to have a Nokia Communicator, a clamshell tending to the size & weight of a brick. The keyboard was quite tichy and so was the 80x24 screen but back then you could get away with hanging a modem off the back of a computer so I did remote admin with that with no real problems. Eventually I replaced it with the next generation wich was smaller - big mistake. But for some reason I can't really get along with on-screen keyboards, even on a tablet.
The belief that because of moores law, no care is taken regarding well optimised code as machines will just get quicker.
Horrible cycle that, and I say this as a linux user, Linux distros are starting to follow.
At least in the older days of more fixed hardware, Amigas and ealier, the machines where not as flexible, so things were done to make them as efficient as possible. Even now some of the Demos being produced are astounding.
I could be wrong, but I don't think I am. Make em rather flimsy with a life cycle that is laughable at 2-3 years so they can continue to feed the beast. Phones are no different. FPS rates on a phone that would shame a gaming PC from a few years ago? WHAT IS THE FREAKING POINT!?
*sobs over psion 5 carcass*
The belief that because of moores law, no care is taken regarding well optimised code as machines will just get quicker.
But they do generally get quicker :)
The problem with optimisation is that it often makes code harder to debug and maintain. That means longer development cycles and higher skill levels from the developers. The world already struggles to produce enough software developers so putting more demands on their ability and time could be a none-starter. For damn' sure it'll bump up the development costs.
It would also require cooperation from management. An understanding across almost the entire company that quality matters more than time and cost to market. Except that in reality much as we developers might whinge the truth is that for most companies time and cost to market are most important.
Windows has (so far) been the world's leading computer desktop software. Does anyone think it got where it is today by being high quality, well optimised code?
We have to be smart about it. Optimise where we need to. Take more care where we need to. Try and develop tools that help us generate better code.
> Windows has (so far) been the world's leading computer desktop software.
It may be the most numerous, but Microsoft has always been a follower and not a leader.
I doubt that there has been anything useful in Windows* that wasn't done previously in some other system. Of course they _claimed_ they were innovative, but was usually something they copied.
* the qualification 'useful' is to eliminate the dog and the paperclip.
The belief that because of moores law, no care is taken regarding well optimised code as machines will just get quicker.
That's no longer really the case: compilers are producing pretty efficient code. We do demand more from our software in ways we don't think matter so much but most of the time the code is doing nothing waiting for us to stimulate it.
Programs use more memory because there is more of it and more memory is usually the easiest way to speed things up. Try opening up a modern spreadsheet on an older machine. RAM, storage and bandwidth now seem limitless so why not use them?
That said, it still amazes me quite how big some software updates turn out to be. Each part of an MS Office for Mac update weighs in at around 1GB. I'm sure MS could distribute smaller binary diffs if they really put their minds to it.
That largely happened on the Amiga after Commodore went bust, and thus no faster models were coming out...
It also happened because the hardware as fixed, so you could program it directly instead of having to go through multiple performance sapping abstraction layers.
"don't forget just how much RAM Firefox sucks up just being open..."
That's true of just about any browser when you point them to the same pages. I regularly switch between Firefox and Chromium, and after pointing them to the same content, I find their memory use is comparable. It's the media-heavy web pages themselves that are the problem. Sometimes, I wonder if the Web should never have been made dynamic; if they wanted interactive content, why not just use a graphical terminal or VNC instead?
We wanted more than 256 colours.
Yeah there is that, currently playing X com 2 and it looks lovely, on a speccy everyone would be built out of large square blocks with a flat 2d view, it would detract from the game. Let alone watching the PC trying to think, it would take a week.*
*Reminds me of rendering a test render on a 386/486 for 3DS max, we had time to build a LAN over a 3 floor house, set up doom, played all night. Bloody thing still hadn't finished.
What went wrong?
IMHO, it really started going downhill when XML got popular. XML is great for data interchange, but then folks had to start using it for general data storage too. Why go to all the trouble to write (fairly simple) code to pull data out of a delimited or fixed-field file, when you can load a big library and just call some standardized functions to pull the data out of all the XML cobwebs surrounding it? Your computer's lack of memory capacity is not the problem of the programmer...
I do remember Novell used XML quite a bit under Netware 6.5 and OES2 to store their configuration files. I always thought that was quite lazy on their part. What program, besides ConsoleOne, is really interested in parsing and understanding the ConsoleOne config files?
So (again IMVHO) acceptance of XML as a data storage standard seemed to coincide with a general idea that the users could/should just throw more memory in their machines and do all this whizzy new stuff.
And before I forget -Get Off My Lawn!!!
I can see that you may still need Android because of the plethora of small apps aimed directly at smartphone/tablet hardware. Few desktops have built in support of compass, accelerometer etc.
I can see the attraction of carrying your phone around then plugging it into a TV or monitor to get the full screen experience and full fat Linux programs for email, office automation, Usenet (remember that?). You still need to carry a keyboard but that can be quite light.
But why oh why oh why aren't they showing a tablet with the choice of two desktops? My Xperia Z has a high resolution screen and I can already attach a keyboard and mouse. Screen area at 10.1" is comparable to netbooks.
Is this just a marketing thing to differentiate from a very expensive Linux/Windows laptop with detachable keyboard which can also run Android? Scared of polluting a high priced market?
I could - I've got both X and Opie running on my Yoga :-)
(I don't know why, really, given the pain involved in tweaking the opie sources/install to fit a less old linux install. I mean, my Sharp Zaurus actually still works...)
"Raspberry Pi 4 core arm chip with 1GB RAM runs a Linux desktop quite well and they are on the cusp of having OpenGL available, that should make it just fine."
Indeed, and that's despite Pi being *intentionally and understandably* signficantly behind the current market leading ARM technology. Throw that restriction away and you'd get a somewhat faster ARM. And it would likely be obsolete within a few months - which may or may not be a problem for some markets.
Pity the article did not even mention Ubuntu. Windows Continuum, takes idea (maybe) from Ubuntu convergence, and Intel is "broadly similar to ...Continuum". My personal take is that this Mir vs Wayland debate all over again. Intel is a Wayland fan and Mir hater. If Mir can deliver convergence already then it is important for hem to prove that Wayland can do the same.
If I understood it correctly, while Windows Continuum lets you run applications with a desktop interface, it still does not let you run just any shrink-wrapped Windows 10 x86 binary application; the applications still have to be for the mobile version of Windows 10, to run on ARM. This, on the other hand, lets one run real Linux programs, not Android programs which are modified to accept the mouse buttons and expect a keyboard.
... started chasing this several years ago. As several commentards have already mentioned -- see the Ubuntu Edge (not to be confused with Microsoft's Edge Browser). I think the much-maligned Unity desktop was designed to be usable on both phones and PCs. Of course PC users hated Unity with strong hatred, just as PC users hated Win 8. Which Microsoft created in order to converge phones, tablets, and PCs in a single OS.
There are only so many Big Ideas, and phone-tablet-desktop convergence is one of them. For me, the take-away from the article to hand is that the tech is close enough now that more players are starting to jump the bandwagon. Nice. I like Linux, it couldn't happen to a nicer OS. But I would wager a small amount that in a few years all OSes will run on phones + tablets + desktops. (And then on implantable chips spliced into your nervous system.)
Indeed.
I've already got an MHL adapter for my phone: big screen done even with some CEC support. I'm sure a USB keyboard could be made to work with the mix but a wireless/Bluetooth one would almost certainly make more sense.
Might be nice to see some performance numbers with any of the newer phones. My Raspi2 now runs my CI tests in about twice the time my 2010 Intel MacBook does. I reckon any post 2014 phone would probably run them faster.
No no no no no.
This isn't what people have in mind when it comes to a Linux desktop on a smart phone.
What we have here is another attempt at what Asus did with its netbooks.
I cant even remember the name of the distro...it sucked that hard.
Linux is a niche market, the vast majority of people that will be interested in carrying a "Linux desktop" around on a smart phone will want some key features...or at least I do...
1. A terminal with root access (this will clearly be out)
2. A package manager aligned with their favoured distro or some alternative thereof. E.g. apt, yum, pacman etc
2.1. If no package manager then a compiler of some sorts.
3. The ability to customise the filesystem as they see fit.
The year of Linux may never happen but engineers will always favour it.
They are the reason for most of the market share...cater to them.
Don't try and foist Linux on granddad and grandma because its cheaper.
PLEASE!
Ill pay a premium for a "pure" Linux phablet, I really will...just make it worth it.
Im sure a lot of geeks balls would explode if that ever happened.
This is why we wanted Sailfish to do well and missing features like the ones above are what basically killed it.
Dont try and yank us in with a product that has a pretty face only to reveal that what you actually have is a T girl with a manly set of cock and balls and hairy armpits*
*If thats your thing im sorry I offended, just swap the two examples round.
So what is the security model in this combo? Android's apps usually demand permission to do everything imaginable, including HW control, whether or not their primary function is related, and the security model is all or nothing: either agree or don't install, no granularity. Linux is a multiuser system with a relatively simple but robust and stable user/group/other model of permissions that is familiar, well understood, and works well in practice.
If the big idea behind the combo is having Android apps in the same system (including filesystem) as a desktop Linux, how will these two models co-exist? Will Android stuff run under a special euid? Will it be isolated from the rest of the system (Linux)? How?
If Intel "can go to production tomorrow" I assume they have the answers...
They will just take it away again, and give whatever special benefits it had to Microsoft. This is them pleading to let them stay relevant but as soon as they are free of the fear of imminent death it will be right back to the old Windows-only hardware shenanigans.
> Will Microsoft allow this, or will they just charge everyone who produces the Android/Linux continuum phone to pay a license fee?
You appear to have the strange idea that Microsoft invented the phone/desktop mix. They, as always, have just copied what other have done (eg Ubuntu) and stuck a different name on it.
My old Motorola Droid X did this years ago, though admittedly probably not quite as well. There was even a dock released for it that no one bought. When you plugged in an external display, you were presented with 3 choices if I recall correctly: "Photo mode", "Mirror display", or "Palm Desktop" The Palm Desktop mode was multitasking if I remember right, and worked fairly well, given the phone's rather meager processor compared to today's mobes. With a Bluetooth keyboard, it was comparable to a low-end netbook at least, and the whole touchscreen of the phone was a mousing surface. I played around with it once or twice, but mostly just used the phone to play Netflix or similar hooked up to a hotel TV when I was travelling. I did hook it up in a conf room once to display a Powerpoint, which worked very well, though I mostly did it just to annoy the people clutching less versatile iPhones.
Google could have taken this ball & ran away with it;
the developer of andromium os (http://www.andromiumos.com/) used to work for google,
& more than likely wanted this in android years ago,
for some reason, the idea didn't stick with google, & it seems he parted ways to push this out himself.
I have the app installed on my nexus 6p, coming from a nexus 5,
but have yet to see if it'll work with an hdmi dongle (though, it's said that the usb type c's hdmi is dormant in the 6p), google could have made it live, seeing it's in the usb -c specs to support hdmi out..
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
