Win64 trial available

[somebody]

And they will. The only thing that will stop them from charging $500 a seat
just for the OS is an alternative. How much does an MS seat cost by the
time you include enough software to actually do something right now?

Since I've reached an age, where I realize only time is truly
precious, the MS seat comes cheaper, much cheaper. I first installed
Linux almost a decade ago. And much of it was done the hard way. I'm
done with having the OS as a hobby. There's better things for me to do
with a computer. And everything I do, is much better on Windows.
That's how it is.

If MS pricing continues to get ugly, and Linux continues to evolve,
I'm thinking that Linux might bury Windows one day. And it's just as
it should be. It's difficult to imagine a company like MS becoming
that stupid. But they are, you know. Seem to happen to all companies
becoming big corps eventually.

ancra

 

[KCB]

AMD?

Ah, now this explains things. Why we're having this funny discussion.
There never was any "slap in intel's face". The fact that you see it
that way, and your generally benevolent view of Intel, and your
positive picture of an all conquering alliance between Intel and your
beloved Linux, is maybe severely tinting everything I say, in your
eyes.
Neither is there a conspiracy, though you seem to view the fact that
there isn't one, as "slap in intel's face"?

But Intel certainly maneuvers to "squash AMD". And I would view both
the rumored desire for a different 64-bit '86, and any eventual
release of such, as an element of that plan. And so, I'm confident,
should any intelligent person.


Rather the ears and eyes of a true zealot. I said "ages ago". AMD
wasn't even around ages ago.
Look, - the PC, as in "IBM-PC-compatible", crushed all competition,
and became the world standard. And it was definitely, one of the
lowest performing personal computer systems on the market. And Intel's
cpus were the slowest, crudest, ugliest existing. So performance is a
very minor reason for a system's success. Only good thing about this,
is that we got a standard.
If Intel introduces a, non '86-64, 64-bit '86 ISA, we won't have the
standard PC anymore. Is it this, you fail to understand? We will have
two different computer systems in competition to become the new market
hegemony.
From Highlander: There can only be one. And there will be only one
64-bit ISA.

Isn't the Itanium exactly what you're talking about here? If Intel had been
successful with it, and AMD had no license to produce it (which they don't),
AMD would have been left out in the cold. That was the whole idea, but
application / programming support never really caught on, and the price
never came down to where IT departments felt it was a good bang for the
buck. They see the migration to AMD-64 as a performance boost now, costing
less than an Intel solution, backward compatible with all 32-bit
applications (because they run natively on AMD-64 chips with no emulation
needed), and forward compatible with Win64 if they see a need to go that
route in the next few years. Sorry if all this has been covered previously,
but you guys just seemed to be dancing around it.

 

[Stacey]

Ah, now this explains things. Why we're having this funny discussion.
There never was any "slap in intel's face". The fact that you see it
that way, and your generally benevolent view of Intel,

I wasn't the one who said there was any slap in their face and I'm typing
this from an AMD machine... Next?

But Intel certainly maneuvers to "squash AMD".

And I'm sure AMD has no desire to take any of Intels market share?

Rather the ears and eyes of a true zealot. I said "ages ago". AMD
wasn't even around ages ago.

Been around since the 386/486 days..

Look, - the PC, as in "IBM-PC-compatible", crushed all competition,
and became the world standard. And it was definitely, one of the
lowest performing personal computer systems on the market.

But it was affordable. 35mm camera have become the "standard" while a 120
film hasselblad is a much higher performing camera..

They're not blind deaf and stupid. So what? What has that got to do
with anything?

Shows Intel isn't afraid of MS like people claim.

It was the Linux zealot, you, that brought Linux into this. Reacting
like Pavlov's dog, on the mere mentioning of MS and Windows.

And you jump into any thread defending AMD or MS because?

 

[Stacey]

Ha, you're serious?
This minor configuration issue, will be solved within a couple of
years.

And they've had how long to sort this out already?

 

[Stacey]

Since I've reached an age, where I realize only time is truly
precious, the MS seat comes cheaper, much cheaper.

I spend much less time -working on- my machine running linux. Windows might
take less time on the initial install (and that's a maybe) but once that's
done, there is basically zero maintainance. With windows one is constantly
chasing security holes, updating virus definitions, buying new AV software
etc etc.

And if you're buying 1000 machines for an office, when the software starts
costing $500,000 just to get setup, things become a little different. Just
ask amazon, they never showed a profit till they got M$ off their back.

 

[Matt]

Ha, you're serious?
This minor configuration issue, will be solved within a couple of
years.

No need. The problems mentioned in the article were solved thirty years
ago.

MS builds software as if none of them have ever been to school.

Two years will be about enough time for Linux market share to reach
levels that hardware makers can't ignore. Then you will see stampedes
of users and software developers.

 

[Ruel Smith (Big Daddy)]

Two years will be about enough time for Linux market share to reach
levels that hardware makers can't ignore. Then you will see stampedes
of users and software developers.

I just read somewhere that they're predicting that Linux will have 6% share
on the desktop by 2007 and will pass the Macintosh. That's enough to get
more 3rd parties involved, but I think the number is a little pessimistic.
Currently, due to those that dual boot, only to use Windows on the internet
more, and Linux's ability to change browser identification, I think that
more people are using Linux than they give credit for. I'm willing to bet
that 6% marketshare will unofficially be obtained next year. There's just
no way of determining how much marketshare Linux has with any accuracy.
Windows comes pre-installed and they must count that if they're going by
sales. Some people dump the preinstalled Windows for Linux after they buy.
Still more download free versions like Mandrake and Fedora.

It's funny that it seems the Linux community's goal is the undoing of
Microslop, but it'll be the undoing of the Macintosh first.



--
Big Daddy Ruel Smith

My SuSE Linux machine uptime:
5:06pm up 8 days 17:38, 2 users, load average: 0.27, 0.21, 0.30

My Windows XP machine uptime:
Something less...

 

[Stacey]

I just read somewhere that they're predicting that Linux will have 6%
share on the desktop by 2007 and will pass the Macintosh. That's enough to
get more 3rd parties involved, but I think the number is a little
pessimistic. Currently, due to those that dual boot, only to use Windows
on the internet more, and Linux's ability to change browser
identification, I think that more people are using Linux than they give
credit for.

Exactly, I have my browser set as "IE5.0" because of all the sites that
demand I 'install the newest IE' if it's set to anything else. Of course
Konquerer works fine but unless it's set as IE those sites won't even allow
me to access them.

I'm willing to bet that 6% marketshare will unofficially be
obtained next year.

Wait till Longhorn and the next level of the "secure computing initiative"
and watch how many people jump ship!

 

[somebody]

No need. The problems mentioned in the article were solved thirty years
ago.

- Ha, got you! Not possible, since MS _invented_ the problem only some
years ago. ;-)
Jokes aside, this shows how different your and my outlook is. You
think that just because some technology existed 35, 25 or ten years
ago, it should be relevant.
That is not how MS goes about things. MS always goes from where they
are. And they have no problems at all with reinventing the wheel.
Rather, it's considered advantageous.

MS builds software as if none of them have ever been to school.

An opinion that shows how much you know about it.
The phrase itself, could perhaps cause a bright person to stop dead in
his track, and reconsider his basic presumptions.

Two years will be about enough time for Linux market share to reach
levels that hardware makers can't ignore. Then you will see stampedes
of users and software developers.

Hardware makers are not ignoring Linux. But I disagree. Two years are
not enough. Takes more, to get MS out of their trenches.
Sofar Linux been successful stopping MS from taking markets they had
their eyes on. Like web servers, TV & hollywood special effects,
special systems, like gambling, shopping etc.
But the real chip in the wall, is government organizations going
Linux.
Me, I think the first step in any actual retreat, will be small
business in the far east. But it won't happen in two years.

ancra

 

[somebody]

I wasn't the one who said there was any slap in their face

Nobody did,

and I'm typing this from an AMD machine... Next?

....and has what to do with what, and why should I care?

And I'm sure AMD has no desire to take any of Intels market share?

- Oh? Then let me assure you, that they are very busy trying to take
market share away from intel. And very intent and focused on it.
But again, what has that to do with - that a non compatible ISA from
Intel represents a bad scenario for users?

Shows Intel isn't afraid of MS like people claim.

Never heard of a such claim. In any way, what's the point? Intel
cannot get the desktop PC cpu market without MS. Has very little to do
with Linux.

And you jump into any thread defending AMD or MS because?

First of all, I'm very wary of defending MS. I'm just sick and tired
of all Linux bigots who can't stand me telling them I prefer Windows,
without going all to pieces at such unheard of blasphemy.
I'm also pretty sick of the 'politically correct' outrageous
vilification of MS and BG, that, frankly, is 93% sectarian bullshit
propaganda.
Secondly, you're not "defending" Linux here. You're being Linux
argumentative in a context where it has no role.

ancra

 

[Matt]

That is not how MS goes about things. MS always goes from where they
are. And they have no problems at all with reinventing the wheel.
Rather, it's considered advantageous.

That's been obvious for quite a few years now. But we are supposed to
be considering what is advantageous to computer users, not what is
advantageous to MS. It's easy for a zealot to forget that I guess.

 

[somebody]

That's been obvious for quite a few years now. But we are supposed to
be considering what is advantageous to computer users, not what is
advantageous to MS.

- Now you're thinking! Good for you. - Exactly! - And where is the
computer user?

It's easy for a zealot to forget that I guess.

Yes, at least it certainly seems so to me. I'm congratulating you on
your insight.

ancra

 

[somebody]

Isn't the Itanium exactly what you're talking about here? If Intel had been
successful with it, and AMD had no license to produce it (which they don't),
AMD would have been left out in the cold. That was the whole idea, but
application / programming support never really caught on, and the price
never came down to where IT departments felt it was a good bang for the
buck. They see the migration to AMD-64 as a performance boost now, costing
less than an Intel solution, backward compatible with all 32-bit
applications (because they run natively on AMD-64 chips with no emulation
needed), and forward compatible with Win64 if they see a need to go that
route in the next few years. Sorry if all this has been covered previously,
but you guys just seemed to be dancing around it.

I've just got the news from Intel's developer forum. It's now
official: It's the long rumored 'Yamhill' sort of '86-64 compatible
ISA. It will first appear on the "Nocona" core cpu, which will be
marketed as a 32/64 bit "Xeon" cpu. The "Potomac" multicore cpu, to
appear sometime in 2005 is also a 32/64 bit cpu, and will also be
marketed under the "Xeon" label, towards the server market.
Intel calls this 32-64 bit ISA "Compatibility Technology", or CT.
Intel is cautious about performance, and claim there will be no
performance enhancement from CT. So it mainly seem to be about
providing the address space. ...Or they want to direct performance
customers to Itanium.

It doesn't appear as if Intel will have anything solid for the desktop
until 2006-07. There will be a CT-chip aimed for workstations though.
Apparently, it is basically an enhanced 'Prescott' core *inside* a
new memory manager. I don't know what is meant by "aimed at
workstations" but it suggests price will be steep.

Intel denies AMD'86-64 compatibility. But "MS Windows for 64-bit
extended systems" as it is officially known, will run on both
AMD'86-64 and Intel CT-cpus. And Intel CT-cpus will also run some
AMD'86-64 binaries ("those currently being developed"). The questions
are: Will AMD'86-64 run Intel CT-binaries?
And will CT run all AMD'86-64 binaries, also those using AMD's extra
registers, or have Intel persuaded developers to cripple AMD's ISA?

My thoughts: Somewhat cautious relief, since there seem to be some
convergence, rather than split of the PC.
It also seem to be somewhat ad hoc. This really seem to be a 32-bit
cpu with 64-bit extended addressing, rather than 64-bit cpu that is
32-bit compatible. Which means Intel have reconsidered fairly
recently. An impression that is strengthened by the lack of pace of
Intel's 64-bit program. Cause for worry is that MS seem to have, at
least to some extent, remained in Intel's bed. It's now clear that the
delay of "Windows for 64-bit extended systems" is to allow Intel time
get their "Nocona" ready.

Jim Allchin, the prize arsehole at Microsoft, put it like this:
"Microsoft's and Intel's leadership continues to deliver powerful,
cost-effective, 64-bit computing to the broad IT market"

ancra

 

[Matt]

Isn't the Itanium exactly what you're talking about here? If Intel had been
successful with it, and AMD had no license to produce it (which they don't),
AMD would have been left out in the cold.

I could be wrong, but I think the antitrust laws would require Intel to
publish the specs for any new instructions, and that anybody would be
free to implement the instructions in silicon. Then AMD wouldn't
exactly be left out in the cold, but they would be left playing catchup,
as usual.

 

[somebody]

I could be wrong, but I think the antitrust laws would require Intel to
publish the specs for any new instructions, and that anybody would be
free to implement the instructions in silicon. Then AMD wouldn't
exactly be left out in the cold, but they would be left playing catchup,
as usual.

There are patents, as usual. So you're wrong. There is also a
settlement, so you're right.

AMD acquired x86 licensing in the early nineties, which they pay
royalties to Intel for. As they got competitive, Intel weren't happy
anymore and disputed the license.
The law wrangling eventually came to a settlement. Pretty onesided, as
you can imagine, since AMD didn't have financial resources for legal
battle. Intel has full access to all AMD patents and the right to
clone and ripoff any AMD technology. The settlement however also gives
AMD access to Intel patents, though they would have to pay royalties.
It may be access is limited to '86 derived technology, though. Which
wouldn't help AMD with Itanium. Nor do AMD have any "cloning" rights
like Intel. They have to implement anything themselves.

According to MS, Intel's CT-ISA, which stands for "Clackamas
Technology" rather than "Compatibility" as I previously thought, is
"broadly" compatible with AMD'86-64.
According to Intel, CT is not compatible with AMD'86-64, but will run
AMD binaries, and will run on the same OSs.
According to AMD finally, CT is compatible with AMD'86-64, in fact,
_IS_ AMD'86-64. So it could be just Intel pride denying it.

But apparently there are some differences. Could be Intel are making
an effort to shut out AMD, but it could also be due to the technology
having come about in a different manner. As a belated slap on to a
32-bit design. I'll see if I can get hold of some technical documents
from Intel.

ancra

 

[Ruel Smith (Big Daddy)]

AMD acquired x86 licensing in the early nineties, which they pay
royalties to Intel for. As they got competitive, Intel weren't happy
anymore and disputed the license.
The law wrangling eventually came to a settlement. Pretty onesided, as
you can imagine, since AMD didn't have financial resources for legal
battle. Intel has full access to all AMD patents and the right to
clone and ripoff any AMD technology. The settlement however also gives
AMD access to Intel patents, though they would have to pay royalties.
It may be access is limited to '86 derived technology, though. Which
wouldn't help AMD with Itanium. Nor do AMD have any "cloning" rights
like Intel. They have to implement anything themselves.

Correct me if I'm wrong, but didn't AMD avoid an x86 license by getting sued
by Intel and winning in court. I believe the court said that their
processors were enough different. Maybe I'm confusing something...



--
Big Daddy Ruel Smith

My SuSE Linux machine uptime:
4:33pm up 1 day 21:19, 2 users, load average: 0.68, 0.29, 0.21

My Windows XP machine uptime:
Something less...

 

[somebody]

Correct me if I'm wrong, but didn't AMD avoid an x86 license by getting sued
by Intel and winning in court. I believe the court said that their
processors were enough different. Maybe I'm confusing something...

I wouldn't know what that was about. Could be this "cloning" thing.

But IBM originally strongarmed a license out of Intel for AMD.
The evolved conflict over this license ended in a settlement currently
in effect, giving Intel and AMD access to each others patents and
technologies, though I believe money and cloning rights is a strictly
oneway affair.

Though I don't know that for a fact, it seems logical to assume this
cross access only concerns x86 derived technology.

In that case AMD can't use it for Itanium. Don't think they need to
though. Intel have published Software developers guides for their new
64-bit extension. I've had a check, and it looks like AMD'86-64 to me.

ancra

 

[Matt]

There are patents, as usual. So you're wrong. There is also a
settlement, so you're right.

For clarity, let me distinguish between an instruction set and the
implementation of the instruction set (microcode and circuitry).

Are you saying that anybody who wants to implement a given instruction
set needs permission from whoever invented/designed/defined the
instruction set?

I would find it surprising if that were true. Didn't Lotus lose a suit
to MS when MS made a Lotus workalike spreadsheet? Something like: the
Lotus makers claimed infringement because Excel used the same keystrokes
to do the same things as Lotus-1-2-3. Seems like about the same
principle involved with instruction sets.

Also Apple lost the suit to MS wherein Apple claimed that MS stole their
Mac look and feel. The resulting legal principle is not as clear there,
because MS argued that window systems were invented long before, at
Xerox PARC and that Xerox had given away the rights or so.

Similarly I don't believe MS can charge somebody to use the programming
interface to its OSes. I mean for instance anybody can make and sell a
program that makes calls to XP functions. And that is not by the grace
of MS but by some antitrust case law, unless I am wrong.

Maybe by cloning, you mean outright copying of microcode and circuitry?

I recall a case some years back wherein somebody got caught copying
somebody else's microcode, but the details are dim.

 

[somebody]

For clarity, let me distinguish between an instruction set and the
implementation of the instruction set (microcode and circuitry).

Are you saying that anybody who wants to implement a given instruction
set needs permission from whoever invented/designed/defined the
instruction set?

No, I was primarily thinking on patents, simply. I would gather an
attempt to use an ISA without a license, would result in legal
proceedings, but I'd rather nor speculate in what would be the general
outcome of that.
In this particular case, AMD'86-64, there's no need, Intel _have_ a
license!

But, while I haven't checked this and don't know for certain, I'm
rather convinced Epic, for instance, is protected by a fair collection
of patents.

ancra