Windows 8 on ARM procressors
Ginny Caughey
some .NET code most likely will. Will Vulcan runtime DLLs be recompiled for
ARM?
Is anybody else going to the BUILD conference in Anaheim in September to
find out about this stuff?
Ginny
D.J.W. van Kooten
<ginny....@hotmail.com> wrote:
Hello Ginny,
>Has anybody given any thought to this? Obviously VO apps won't run there but
>some .NET code most likely will. Will Vulcan runtime DLLs be recompiled for
>ARM?
If I understand this well, tablet's based on Windows can now run
VO/Vulcan code and those based on Windows 8 require html5/ JavaScript
and Internet Explorer 10 and not longer on traditional exe's. Do you
mean to say that?
I would say that goes for every program languages. Windows will lose
attention anyway. MS tries hard to let apps run on Windows 8 just like
Apple and Android. They sold their mobile Windows to Nokia who will no
doubt exchange their buggy and disastrous Symbian to another loser,
Windows Phone 7.
For iPhone and Android, the 2 leading mobile OS systems, you can't use
Vulcan or VO or any traditional language either I would say.
Dick
Ginny Caughey
to continue to run our old VO Win32 apps as well as our .NET Windows apps.
On ARM processor runtime DLLs must be built specifically for ARM. I don't
know if .NET apps would need to be changed, but I have run .NET library code
built for desktop Intel Windows fine on ARM processors running on Windows CE
as long as there were no platform-specific calls in the code. (Note that not
only was the chip architecture different but so was the OS. There is a
Portable Library Project currently at CTP that helps .NET developers build
library code that will work on multiple .NET platforms.) I really don't know
how much this will change for Windows 8.
Microsoft hasn't said a lot yet about what their ARM-based tablets will or
won't run other than that they will have a new Start menuing system based on
HTML5. (Win8 laptops/desktops will have that too, but they will also have
the classic Start menu as an option.) They have also said they will not
provide some sort of VM or interface layer on the new ARM Win8 tablets to
allow them to run legacy Win32 code. They haven't said yet how you might
build traditional EXEs that target ARM Win8 tablets, but they showed Office
running on a tablet, and surely Office wasn't rewritten all in HTML5 for the
demo.
Brian is on top of all this and will be attending a meeting at Microsoft in
a few weeks to get all the details he needs to plan for the next version of
Vulcan. (Thanks, Brian, for the email.)
Ginny
"D.J.W. van Kooten" wrote in message
news:oqgku69b20nl0m48m...@4ax.com...
Geoff Schaller
we talking two different things here: RISC processors and HTML5? I mean
RISC processors have been around since the late 80's now they make a lot
of money from mobile phones in particular. But they don't have much
impact for desktops and I suspect, laptops. Ok so ARM cores number in
the 10's of billions now but I think the new front will be whether
technology can shift the traditional core into genuine competition with
one that uses a reduced instruction set. If it can, ARMs days are
numbered.
...and what you can do on an ARM core is necessarily limited. Will that
matter in 3 years time? It might.
But Windows 8 will still support all other architectures, it's just that
MS has announced ARM support. Perhaps this is good - perhaps it is too
late. But I don't think HTML 5 has anything to do with this. Regular
applications will still run Windows 8. How the Start button works is
rather immaterial. Certainly what W8 offers is a new paradigm, from
touch technology to massive device variety support, I certainly see it
unifying our development environment even more.
But it also marginalises Vulcan even further. You just cannot keep up
with this degree of environmental change with one developer for a
compiler product. We've now made the commitment to WPF, although it is
going to take some time to move over from WinForms. But we need apps and
libraries we can use on regular desktops, tablets and mobile phones. Yes
we are limiting ourselves to Windows platforms but we tell our clients
this and if they have an iPhone, they go out and get a second Windows
phone, to use for business. They keep the iphone for personal use and to
impress their daughters. So I don't see an issue here limiting ourselves
to Windows platforms. The cost of a new phone is low now - no reason not
to have two or more...
Geoff
"Ginny Caughey" <ginny....@hotmail.com> wrote in message
news:4dea9176$0$19586$c3e8da3$c8b7...@news.astraweb.com:
Richard Wooters
Did some searches over the weekend, on Windows 8, and didn’t find
anything conclusive but I did find some tidbits that I think are
interesting.
1) It sounds like .NET apps will be considered native apps in Windows
8.
2) It looks like x86-compiled apps will be labeled LEGACY and will
have to be run in a VM XP mode on the desktop, no ARM support of
course.
3) It sounds like the Micro$oft / Intel alliance is on shaky ground,
see item 2 above.
I think Vulcan.NET is better poisoned to move forward, than VO, at
this point and to not get the LEGACY label. There are some references
to unmanaged functions in the Vulcan runtime that will have to be
replaced and the macro compiler will have moved to managed code but I
don’t see these as major obstacles.
Item 3 REF http://www.extremetech.com/article2/0,2845,2385563,00.asp
Richard Wooters
Ginny Caughey
Thanks,
Ginny
"Richard Wooters" wrote in message
news:9b44c7bb-1621-4b60...@h36g2000pro.googlegroups.com...
Geoff Schaller
Dot Net is "native" now. VO became legacy, if you really want to give it
that label, years ago but that doesn't lessen its continuing usefulness.
But under W8 there is no suggestion of having to run an XP Mode style
interface (not that such is a burden anyway) for Win32 API based
languages. Far from it. Win32 will continue to be supported 'natively'
for some time or there would be a general revolt. But MS doesn't have to
not support it and doesn't gain anything by not doing so.
I also doubt there is any dent in the Intel/MS relationship. Intel
motherboards are still supreme in the server and PC world and W8 is not
going to change that.
And perhaps your suggestion that Vulcan is "poisoned" is more than a
Freudian slip <g>. It really isn't positioned to go forward anywhere and
will fade slowly and sadly into oblivion with a lot of other good ideas
that were poorly implemented.
Geoff
"Richard Wooters" <Ric...@rbj.com> wrote in message
news:9b44c7bb-1621-4b60...@h36g2000pro.googlegroups.com:
Ginny Caughey
I don't think Vulcan is "poisoned" but your other points are well taken. The
problem with Vulcan in contrast to other .NET languages are the parts that
are Win32 DLLs which Richard acknowledges. Microsoft isn't going to migrate
those to ARM (as they probably will their own runtime DLLs) but Brian might,
and it might not be a huge deal to do so. He will find out if that's a
reasonable thing to do in July. I'm confident he'll do the right thing once
there is some actual info to go on.
Mainly I was curious where Richard found a reference to the XP mode VM stuff
for x86 code (including .NET?) since I haven't seen it anywhere. I had a
discussion with Mary Jo Foley today about old code moving forward, and she
didn't mention that at all either. (And she tends to know that sort of
thing.)
Ginny
"Geoff Schaller" wrote in message
news:4ded5162$0$12390$c30e...@exi-reader.telstra.net...
Ginny Caughey
This seems to imply that COM can once again be a bridge between the old and
new just as it can between VO and .NET today:
http://forums.mydigitallife.info/threads/26404-Windows-8-(7955)-Findings-in-M3-Leak?p=442463&viewfull=1#post442463
Ginny
"Richard Wooters" wrote in message
news:9b44c7bb-1621-4b60...@h36g2000pro.googlegroups.com...
Hi Ginny,
Ginny Caughey
http://getwired.com/2011/06/02/windows-8-crossing-the-chasm/
"Geoff Schaller" wrote in message
news:4ded5162$0$12390$c30e...@exi-reader.telstra.net...
Richard,
Richard Wooters
Sorry for the late post, had to take care of a software emergency.
> Thanks for your thoughts, Richard. Do you have a reference for 2) below?
I have to make a correction, “Windows 7 Mode” not “XP Mode”. I had
reposted from an angry developer without double checking. There seems
to be a lot of rumors flying around out there, that’s causing stress
among some developers, over what Intel VP Renee James had said at the
Intel's Investor Meeting 2011 and what Micro$oft is not saying.
Although Micro$oft is denying the accuracy of Intel's claims they are
not saying which parts are True so speculation grows and until the
smoke clears we really won’t know what we have.
Ref “http://www.theregister.co.uk/2011/05/19/
microsoft_contradicts_renee_james_on_windows_8/”
Ref “http://news.softpedia.com/news/Windows-8-to-Feature-Windows-7-
Mode-201156.shtml”
Ref “http://www.channelregister.co.uk/2011/06/06/
windows_tablets_without_silverlight_dot_net/”
On Jun 6, 2:05 pm, "Ginny Caughey" <ginny.onl...@hotmail.com> wrote:
> Thanks for your thoughts, Richard. Do you have a reference for 2) below?
>
> Thanks,
> Ginny
>
> "Richard Wooters" wrote in message
>
> news:9b44c7bb-1621-4b60...@h36g2000pro.googlegroups.com...
>
> Hi Ginny,
>
> Did some searches over the weekend, on Windows 8, and didn’t find
> anything conclusive but I did find some tidbits that I think are
> interesting.
>
> 1) It sounds like .NET apps will be considered native apps in Windows
> 8.
>
> 2) It looks like x86-compiled apps will be labeled LEGACY and will
> have to be run in a VM XP mode on the desktop, no ARM support of
> course.
>
> 3) It sounds like the Micro$oft / Intel alliance is on shaky ground,
> see item 2 above.
>
> I think Vulcan.NET is better poisoned to move forward, than VO, at
> this point and to not get the LEGACY label. There are some references
> to unmanaged functions in the Vulcan runtime that will have to be
> replaced and the macro compiler will have moved to managed code but I
> don’t see these as major obstacles.
>
> Item 3 REFhttp://www.extremetech.com/article2/0,2845,2385563,00.asp
>
> Richard Wooters
>
> On Jun 2, 10:43 am, "Ginny Caughey" <ginny.onl...@hotmail.com> wrote:
>
>
>
> > Has anybody given any thought to this? Obviously VO apps won't run there
> > but
> > some .NET code most likely will. Will Vulcan runtime DLLs be recompiled
> > for
> > ARM?
>
> > Is anybody else going to the BUILD conference in Anaheim in September to
> > find out about this stuff?
>
> > Ginny- Hide quoted text -
>
> - Show quoted text -
Richard Wooters
> This seems to imply that COM can once again be a bridge between the old and
> new just as it can between VO and .NET today:http://forums.mydigitallife.info/threads/26404-Windows-8-(7955)-Findi...
It would seem COM has never really gone away, just taken on a new
form.
Richard Wooters
On Jun 7, 4:42 am, "Ginny Caughey" <ginny.onl...@hotmail.com> wrote:
> Richard,
>
> This seems to imply that COM can once again be a bridge between the old and
> new just as it can between VO and .NET today:http://forums.mydigitallife.info/threads/26404-Windows-8-(7955)-Findi...
>
> Ginny
>
> "Richard Wooters" wrote in message
>
> news:9b44c7bb-1621-4b60...@h36g2000pro.googlegroups.com...
>
> Hi Ginny,
>
> Did some searches over the weekend, on Windows 8, and didn t find
> anything conclusive but I did find some tidbits that I think are
> interesting.
>
> 1) It sounds like .NET apps will be considered native apps in Windows
> 8.
>
> 2) It looks like x86-compiled apps will be labeled LEGACY and will
> have to be run in a VM XP mode on the desktop, no ARM support of
> course.
>
> 3) It sounds like the Micro$oft / Intel alliance is on shaky ground,
> see item 2 above.
>
> I think Vulcan.NET is better poisoned to move forward, than VO, at
> this point and to not get the LEGACY label. There are some references
> to unmanaged functions in the Vulcan runtime that will have to be
> replaced and the macro compiler will have moved to managed code but I
> don t see these as major obstacles.
>
> Item 3 REFhttp://www.extremetech.com/article2/0,2845,2385563,00.asp
>
> Richard Wooters
>
> On Jun 2, 10:43 am, "Ginny Caughey" <ginny.onl...@hotmail.com> wrote:
>
>
>
> > Has anybody given any thought to this? Obviously VO apps won't run there
> > but
> > some .NET code most likely will. Will Vulcan runtime DLLs be recompiled
> > for
> > ARM?
>
> > Is anybody else going to the BUILD conference in Anaheim in September to
> > find out about this stuff?
>