Alliance Google + Intel for Android on Atom

[Fabrizio Giudici]

http://www.reuters.com/article/2011/09/13/us-intel-google-idUSTRE78C5FF20110913

--
Fabrizio Giudici - Java Architect, Project Manager
Tidalwave s.a.s. - "We make Java work. Everywhere."
fabrizio...@tidalwave.it
http://tidalwave.it - http://fabriziogiudici.it

[Carl Jokl]

I wonder whether I am the only one who feels like the old x86 refuses
to die and it is bad enough to be stuck with in on the desktop but now
if Intel has their way I will be using it on my mobile too. I don't
want an x86 powered phone!

On Sep 16, 12:30 pm, Fabrizio Giudici <fabrizio.giud...@tidalwave.it>
wrote:
> http://www.reuters.com/article/2011/09/13/us-intel-google-idUSTRE78C5...

>
> --
> Fabrizio Giudici - Java Architect, Project Manager
> Tidalwave s.a.s. - "We make Java work. Everywhere."

> fabrizio.giud...@tidalwave.ithttp://tidalwave.it-http://fabriziogiudici.it

[Fabrizio Giudici]

On 09/16/2011 01:36 PM, Carl Jokl wrote:
> I wonder whether I am the only one who feels like the old x86 refuses
> to die and it is bad enough to be stuck with in on the desktop but now
> if Intel has their way I will be using it on my mobile too. I don't
> want an x86 powered phone!

I'm puzzled too. The latest wave of mobile appliances is technologically
successful and is able to run on a "clean" architecture (ARM), without
all the legacy encumbered of x86.

I'm wondering whether Intel is really interested in mobile and tablets,
or whether they want, just in case, to be able to migrate "real"
desktops to Android. Clearly, today Android is light years away from
being able to do things of a real desktop (more because of the lack of
serious apps than for architecture deficiencies, I'd say) but the thing
could change.

--
Fabrizio Giudici - Java Architect, Project Manager
Tidalwave s.a.s. - "We make Java work. Everywhere."

[Fabrizio Giudici]

On 09/16/2011 01:49 PM, Fabrizio Giudici wrote:
>
> I'm wondering whether Intel is really interested in mobile and
> tablets, or whether they want, just in case, to be able to migrate
> "real" desktops to Android. Clearly, today Android is light years away
> from being able to do things of a real desktop (more because of the
> lack of serious apps than for architecture deficiencies, I'd say) but
> the thing could change.

Further explanation. Intel today lives on Windows and, even though
marginally, on Mac OS X. Microsoft did some big mistake in the past, and
was somewhat forgiven. Their share is still very high. But they could do
another mistake and competition push more and more. Mac OS X (and Linux)
are not much relevant for Intel, I think, and there's always the chance
that Apple in future moves again to another processor (ARM). So, keeping
the "Android desktop" door open could be a good B plan.

PS Of course, I'm thinking of client side. The story on the server side
is different.

[Casper Bang]

> I'm puzzled too. The latest wave of mobile appliances is technologically 
> successful and is able to run on a "clean" architecture (ARM), without 
> all the legacy encumbered of x86.

Just to be clear, developing for ARM isn't quite that rosy. For every ARM CPU I've ever played with (Android phone, Synology NAS, iGala picture frame etc.) it's quite a tricky process to setup the toolchain and makefiles. Remember the SoC's use a mishmash of instruction sets and extensions, this is both the beauty and the complexity of the architecture. ARM sells/licences designs, Intel sells actual silicon... the latter carries more legacy aspects, but it is still more homogeneous. Btw. Linus Torwalds lately expressed his concerns regarding developing for ARM: http://www.itworld.com/mobile-wireless/175829/arm-and-linux-major-construction-ahead

> I'm wondering whether Intel is really interested in mobile and
> tablets, or whether they want, just in case, to be able to migrate
> "real" desktops to Android. 

I think it's simply a matter of them spreading their portfolio, mobile computing is after all destined to take over quite a bit of the typical desktop space. And with so much cash in the bank, why not? It was, after all, Intel's small research team in Israel that came up with the Core (Pentium-M micro-architecture) that ended up saving Intel from loosing to AMD, back when Oregon engineers went pipeline crazy with the NetBurst (Pentium 4) micro-architecture. In Intel's defense, they also tried with Itanium but, at least in the past decade, there was little incentive to leave the proven and backwards-compatible x86 platform. Times have changed though.

[opinali]

I guess ARM and partners should make a better work in the tooling space; developers (at least when using C or other languages but not handcoded Assembly) should not have more work than choosing a few compiler options in the IDE or makefile. Wintel's power always owed a lot on superior tooling... by the way, if I ever write my own History Of Java, I'll assign at least 50% of its success to the fact that Java allowed better tools to be written.

A+
Osvaldo

[Karsten Silz]

On Friday, September 16, 2011 1:36:39 PM UTC+2, Carl Jokl wrote:

I don't want an x86 powered phone!

I don't care if my iPhone or iPad runs on ARM or x86. I do care about battery life and weight (I think with tablets, weight is a feature). That's where Intel chips have been at a disadvantage: The "power consumption to computing power" ratio is way behind ARM.

But just as with operating systems, I welcome competition and hope that ARM will make x86 better for mobile use and the the other way around.