To view parent comment, click here.
To read all comments associated with this story, please click here.
An x86 port of Android would be a red headed stepchild...
Best case it would be no worse than the more common ARM devices, it would run the same platform neutral or open source apps, with equivalent performance and battery life.
Worst case...
It won't run a large number of existing Android apps which are closed source and only compiled for ARM.
It will use more power than competing ARM designs, and therefore have inferior battery life.
An x86 windows tablet on the other hand, does offer benefits over an arm based windows tablet (availability of existing programs, albeit program unsuitable for use on a tablet ui)... The windows tablets will be different enough to the android tablets to mask any efficiency differences, and the x86 tablets will be sufficiently better than the crippled arm version of windows that many people will be happy to sacrifice some battery life for a massively superior device.
Whilst I agree with the main point of your post, I do question this assumption.
Android apps are written in "Dalvik", which is modelled after Java (it is not derived from Java, since Dalvik has no Java code). As I understood it, a Java app, and hence a Dalvik app, contains bytecode and not compiled binary.
http://en.wikipedia.org/wiki/Java_bytecode
http://en.wikipedia.org/wiki/Dalvik_%28software%29
"Dalvik is the process virtual machine (VM) in Google's Android operating system. It is the software that runs the apps on Android devices. Dalvik is thus an integral part of Android, which is typically used on mobile devices such as mobile phones and tablet computers as well as more recently on embedded devices such as smart TVs and media streamers. Programs are commonly written in Java and compiled to bytecode. They are then converted from Java Virtual Machine-compatible .class files to Dalvik-compatible .dex (Dalvik Executable) files before installation on a device."
Hence I would assume that Dalvik-compatible .dex (Dalvik Executable) files are architecture independent.
http://source.android.com/tech/dalvik/dalvik-bytecode.html
An x86 Android device could therefore execute the exact same Dalvik executable .dex files as an ARM device, without any re-compilation required.
Edited 2012-09-19 02:30 UTC




Member since:
2007-04-18
Let's be clear, this is pure unsupported conjecture. (N.B. not agreeing with you doesn't mean I agree with Thom.)
How about at least putting out a binary blob driver, like lots of vendors do? (nVidia, AMD, Broadcom, etc.) This argument has been shown to be wrong before.
You do realize that the non-Apple tablet market (for which this chip is almost certainly targeted) is about 100% Android (and thus Linux)? Windows 8 has 0% of the market. Thus, from a pure business perspective (even accepting your above arguments about IGPs), it makes next to no sense at all to target a non-existent market with a new product, unless there are other motives behind it as well (such as an exclusivity deal with Microsoft). Just to drive this point home a bit better, let's quote the linked article on the The Inquirer:
Is this absolute 100% proof that Intel and Microsoft have an exclusivity deal? No. But it makes it extremely likely, to the point of being beyond a reasonable doubt.
Edited 2012-09-15 10:22 UTC