Windows 10 On ARM: 7 Observations - InformationWeek

InformationWeek is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

IoT
IoT
Comments
Windows 10 On ARM: 7 Observations
Newest First  |  Oldest First  |  Threaded View
Page 1 / 2   >   >>
PhilBiker
50%
50%
PhilBiker,
User Rank: Apprentice
11/17/2014 | 4:14:26 PM
I agree: Being able to run on ARM is important - be hardware agnostic if possible.
As a very satisfied owner of a Surface RT tablet, two quotes stood out for me from the article:

"As mentioned, previous ARM-powered Windows tablets failed partly due to poor marketing and customer communication. "

and 

"Even if Intel chips serve Microsoft's needs for the present, continued ARM development is prudent over the long term."  (which followed a discussion of how ARM chips are getting to be as powerful as desktop chips, at least in the Apple hardware.

Windows RT is an amazing OS, and the RT based Surface and Nokia tablets are really, really great pieces.  Surface RT and WWinodws on ARM certainly did not fail due to any lack of technicalogical merit - they are first rate machines!  Windows 8.1 RT is really the best of both worlds - tablet and PC.  Having Office bundled makes it an incredible bargain.  It runs great.  It would be short sighted to abandon it, - instead, open it up to legacy Windows x86 code (if it is feasible) and continue to run with it.  The customer can only win, though it may cost Microsoft in the short/medium run.
hho927
50%
50%
hho927,
User Rank: Ninja
10/17/2014 | 3:05:49 PM
Re: If Windows has never succeeded on an ARM device...
I'm not into hardware but I think ARM instructions are very limited. You can keep the NT kernel but you have to re-architect it to fit ARM. It may be 1 instruction on ARM can take 10 instructions on x86 (and vice versa)

Modified windows to fit touch is not going to work well. They need a touch oriented GUI. If an app is not designed for touch then it does not make sense to run it on a touch device.

I use a buch of devices (with different OSes) and that is my conclusion.
rradina
50%
50%
rradina,
User Rank: Ninja
10/16/2014 | 10:35:01 PM
Re: If Windows has never succeeded on an ARM device...
Based on devices running Windows Phone, it has lower memory requirements than Android.  I have a Nokia 520 and it ran Windows Phone 8 just fine with 512MB RAM.  After I upgraded to WIndows 8.1, it struggles to remain "fast and fluid".  It would run better with 1GB of RAM.  However, Microsoft is still releasing it's entry-level model with 512MB RAM.

I also own a One Plus One Android 4.4.4 w/CyanogenMod.  Freshly booted it's pretty close to using about 1GB of RAM.  Almost identical to my Dell V8 Pro Windows tablet.

At least regarding memory and "bloatedness", I don't see much difference between Windows and Android.  Based on experience, I'd actually feel comfortable saying Windows Phone is more efficient from a processor perspective than Android.  My One Plus One has a Snapdragon 801 2.5Ghz quad core.  It's a very snappy performer but it doesn't seem any more fluid than my lowly Nokia 520 with a 1Ghz single core -- at least with Windows Phone 8.  (As I said, 8.1 needs more than 512MB RAM and is sluggish switching between apps and screens.

 
rradina
50%
50%
rradina,
User Rank: Ninja
10/16/2014 | 10:26:12 PM
Re: If Windows has never succeeded on an ARM device...
Why is running on DEC Alpha so different from ARM that porting is difficult or impossible?

Why is Android bad on x86?  It's just Linux, isn't it?  Is it the Dalvik VM that doesn't run well on x86?  Dalvik is open-source Apache and likely "C"/C++ code.  x86 processors have a rich history of excellent "C" compilers.  I find it difficult to believe Dalvik is at a severe disadvantage on the x86 architecture.  Linux runs great on x86 as well as PowerPC and lots of different processor architectures.  What makes Android so special on ARM that apparently only it and iOS have game?  What makes Windows fail?

Windows CE was definitely just an altered Windows and totally unsuitable on small form factor devices and barely supported touch.  It was awful.  However, Microsoft's new "modern" nee "Metro" world is a very touch-first centric approach.  It also runs applications in special sandboxes to offer greater protection than offered by native Windows apps.

Why does it matter that Android has a different user space/can't run most regular Linux apps?  You seemingly state this as a great advantage.  Why?
TerryB
50%
50%
TerryB,
User Rank: Ninja
10/15/2014 | 1:18:33 PM
Re: If Windows has never succeeded on an ARM device...
I'm with @rradina on this, be very interested to know exactly why you think MS is failing. Windows has always had "bloated" reputation, Linux much more slick and compact in code. Is it strictly a matter that Windows can't be (or hasn't been) stripped down enough to reduce the number of instructions the chip must execute? Or are the reasons much deeper and much more technical?

I remember using Assembler back in college to write o/s. Mostly a lot of "move register" commands, not much you can tell those chips to do. So it seems like running the fewest instructions possible is about the only coding technique you can use for both performance and power use.
hho927
50%
50%
hho927,
User Rank: Ninja
10/15/2014 | 12:35:01 PM
Re: If Windows has never succeeded on an ARM device...
1) DEC Alpha is a different than ARM. You can port your os from DEC Alpha to ARM, but your os is not going to run efficiently. One would think android runs better on x86 but Android is not that good on x86 platform.

2) Android GUI was designed with touch interface in mind. MS GUI is just altered 'windows' to make it fit on a touch screen. Android has a diffrent 'user space'/can't run most regular linux apps on android.
rradina
50%
50%
rradina,
User Rank: Ninja
10/15/2014 | 10:37:08 AM
Re: If Windows has never succeeded on an ARM device...
You haven't provided any credible argument to support these assertions:

***********************

SAID BY YOU:

"No! The reason they fail because instead of comming up with a new architecture on Arm, they force windows down to arm. It was started by Gates. Force Windows on everything. Windows CE was terrible. Windows 8 on arm is not that great.

Except GUIs, nothing changed. All latest windows are still based on NT architecture (which was based on OS2 and DEC VMS)."

...

"Google has made architectural changes to Android (which is based on Linux) to fit ARM/phones/tablets."

***********************

What new architecture are they missing?  What does the term "fit" mean?  Is it RAM size?  Power management?

Whatever "fit" means, what did Google do for Linux on ARM that Microsoft has failed to do for NT on ARM?  Other than RISC vs. CISC, what's special about ARM vs. Intel?  ARM CPUs have power management.  Intel has power management.  If it's memory requirements, NT started on machines with 16MB of RAM.  What changes did Google make to Linux to make it fit phones/tablets that Microsoft has failed to do or simply cannot do with NT?

Remember, your original assertions are such that Google has done for Linux what Microsoft has failed to do for Windows and that is why it will continue to fail.  Can you support those assertion with a reasoned argument?
hho927
50%
50%
hho927,
User Rank: Ninja
10/14/2014 | 7:57:13 PM
Re: If Windows has never succeeded on an ARM device...
Google changed Binder, ashmem, pmem, logger, wakelocks (power management). I answered you. RISC is a technology. There are many RISC cpus.

 
rradina
50%
50%
rradina,
User Rank: Ninja
10/14/2014 | 7:38:47 PM
Re: If Windows has never succeeded on an ARM device...
You didn't answer my question. If you don't know, fine but you've made made a few significant statements regarding Microsoft's mistakes that need more context.
hho927
50%
50%
hho927,
User Rank: Ninja
10/14/2014 | 7:23:11 PM
Re: If Windows has never succeeded on an ARM device...
ARM is Acorn RISC Machine. RISC was first a concept from Berkeley. Acorn had limited resources so they tried to make it as simple as possible. The result was a processor that uses so little power. And that is a major advantage on phones,tablets.

Every vendor come up with their own RISC version (DEC, Alpha, SUN, IBM, Motorola, etc). Ironically Acorn went bankrupt after It spun off ARM.
Page 1 / 2   >   >>


2020 State of DevOps Report
2020 State of DevOps Report
Download this report today to learn more about the key tools and technologies being utilized, and how organizations deal with the cultural and process changes that DevOps brings. The report also examines the barriers organizations face, as well as the rewards from DevOps including faster application delivery, higher quality products, and quicker recovery from errors in production.
News
The State of Chatbots: Pandemic Edition
Jessica Davis, Senior Editor, Enterprise Apps,  9/10/2020
Commentary
Deloitte on Cloud, the Edge, and Enterprise Expectations
Joao-Pierre S. Ruth, Senior Writer,  9/14/2020
Slideshows
Data Science: How the Pandemic Has Affected 10 Popular Jobs
Cynthia Harvey, Freelance Journalist, InformationWeek,  9/9/2020
Register for InformationWeek Newsletters
Video
Current Issue
IT Automation Transforms Network Management
In this special report we will examine the layers of automation and orchestration in IT operations, and how they can provide high availability and greater scale for modern applications and business demands.
White Papers
Slideshows
Twitter Feed
Sponsored Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.
Sponsored Video
Flash Poll