AMD Sempron without 64bit = big mistake (UPDATED)

slvr_phoenix · Jun 16, 2004

I think you did not understood my point. I don't care about the Intel 64bit launch. I said that if you have a x86-64bit CPU today (only AMD are selling them NOW), you will be able to easily upgrade to 64bit, it will only a mather of software installation.

I got your point, and if you had paid attention you'd have noticed that I never disagreed with it.

Yes, current AMD64 owners are going to be able to upgrade to 64-bits once the software becomes available (at least if they aren't already running Linux), but there are a lot of people holding off on hardware purchase until the software is there. I know a number of people that want 64-bit but are waiting for software before they buy the hardware. And I don't personally know anyone who bought the hardware with the intention of upgrading the software when if becomes available. (Why buy the same OS twice, especially when it's so expensive?) So from my personal point of view the majority of SOHO 64-bit hardware sales are only going to occur once a 64-bit WinXP version exists.

So a lot of what happens in the future will depend on if Intel has 64-bit hardware available at the same time as MS releases a 64-bit WinXP. It will be interesting to see how the 64-bit push plays out. It will be interesting to see how Intel takes to it. It will be interesting to see how far AMD goes to push it. And most of all it will be interesting to see if advertisements make the same kind of idiotic claims that 64-bit is twice as fast as 32-bit like the Macintosh marketspeak buffoons tried. 😱

I'm ready to pop my popcorn and sit down for a good show.

<pre>"Build a man a fire and he's warm for the rest of the evening.
Set a man on fire and he's warm for the rest of his life." - Steve Taylor</pre>

TheRod · Jun 16, 2004

So a lot of what happens in the future will depend on if Intel has 64-bit hardware available at the same time as MS releases a 64-bit WinXP. It will be interesting to see how the 64-bit push plays out. It will be interesting to see how Intel takes to it. It will be interesting to see how far AMD goes to push it. And most of all it will be interesting to see if advertisements make the same kind of idiotic claims that 64-bit is twice as fast as 32-bit like the Macintosh marketspeak buffoons tried.

I totally agree! This will be very interesting. And Intel 64bit vs AMD 64bit perfromance will be another interesting thing to analyse. From what Microsoft said, AMD seems to have an edge over Intel, not necessarly in overall performance, but in implementation at least.

--
It's tricky to use words like AMD or Intel in a signature some users could think your are biased.

slvr_phoenix · Jun 16, 2004

that is the obvious attacks agianst amd's company somtimes, even amd fans are doing it, first it was the s939 pricing, now its if the semprons wont have 64bit capabilities. this is not about a bais on my part, its just ive never seen any sorts of attacks on what intel has done, no lengthy threads on intel's direction.

No offense, but I think that you have a severe problem distinguishing between a critique and an attack.

There's absolutely nothing wrong with analyzing a decision and agreeing or disagreeing with it, nor with discussing and debating that critique for that matter. It is our ability to do just so that differentiates us from robots. It's called free will. We don't have to agree with everything that other people say and do. You know? And that is the whole point of a forum after all, to discuss and debate the virtues of a topic.

So there will always be plenty of critiques. That is what makes us human. That doesn't make them attacks. Analysis is not an attack.

this is not about a bais on my part, its just ive never seen any sorts of attacks on what intel has done, no lengthy threads on intel's direction.

I think that you must have extreme tunnelvision then. Do a search for Presshott some time. There are plenty of criticisms of Intel on these boards. (Sometimes I think more so than of AMD.) If you truly cannot find a single critique of Scotty's heat or performance, or the P4EE's price and performance for that matter, then you've got to be blind. And those are just the most blatantly obvious and recent Intel criticisms. The're also Celeron's performance, Intel's new socket, Itanic, etc., etc., ad infinitum. You may not be intentionally biased, but the simple fact that you do not balance the AMD criticism with the Intel criticism shows a heavy weighting on your part, conscious or unconscious.

<pre>"Build a man a fire and he's warm for the rest of the evening.
Set a man on fire and he's warm for the rest of his life." - Steve Taylor</pre>

P4Man · Jun 16, 2004

> And I don't personally know anyone who bought the hardware
>with the intention of upgrading the software when if
>becomes available. (Why buy the same OS twice, especially
>when it's so expensive?)

Hmm.. you mean people didn't line up in front of the stores to buy a retail copy when "32 bit" Windows 95 was launched, but they bought a new PC instead ? You have a point about paying twice though, it would be nice if it where a free upgrade/service pack like release, but knowing MS, its not that likely

What is worse though, is that your best bet might be to have both 32 and 64 bit windows versions for while, unless MS really improves their 64 bit version over the current beta.

> So from my personal point of view the majority of SOHO
>64-bit hardware sales are only going to occur once a 64-bit
>WinXP version exists.

That is most likely true, but not because of your argument. Simply because I expect Intel and MS to release their 64 bit products more less at the same time (late this year). Since intel enjoys a 85% or something marketshare, its obvious the majority of 64 bit hardware won't be sold until intel sells it

But if they had 64 bit P4's out today, I am not too sure XP64's release later this year would change that much. People would buy the 64 bit P4's today just as well, and hapily run XP32 on it.

>It will be interesting to see how the 64-bit push plays
>out.

I don't think it will be very different from the windows 3.11->windows 95 transitions or windows9x->Win2k/XP. Some people will switch the same day and promise you that new OS is the best thing since sliced bread, other will complain, migrate slowly, point to lackluster software/driver support, say it aint worth it, tell you its slower, a memory hog,stick to the old windows for 5 years or longer. Also, some people will buy the retail copy, come home and discover it doesnt install on their "nearly new" $3.000 P4 system and they will be pissed.

Software support will follow a similar pattern, next year you'll see some native apps ported, but mostly everything will still be made available for XP32, other apps might just hop on the marketing bandwagon, and put "XP-64 ready" stickers on the boxes, even if its just 32 bit software,.. but in spite of all naysayers, in a few years, just about every new computer will be 64 bit capable and ship with a 64 bit OS, and slowly, high end apps will stop supporting the 32 bit legacy. It may take 10 years or longer though (possibly forever) until virtually all software is 64 bit. but who cares anyway? I'm not waiting for a 64 bit version of Office or WinZIP, but for 64 bit encoders, games, and some olap tools.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

slvr_phoenix · Jun 16, 2004

I totally agree! This will be very interesting. And Intel 64bit vs AMD 64bit perfromance will be another interesting thing to analyse.

I don't know if the performance will be interesting to compare. I mean Intel is fugging themselves up so badly on Scotty, how could they possibly even equal AMD at this point? It'd take one heck of a design fix for Intel to manage that I think.

From what Microsoft said, AMD seems to have an edge over Intel, not necessarly in overall performance, but in implementation at least.

Well sure. I mean AMD came up with the x86-64 extension. Intel only just started using it, and it seems likely that they were being compatible to the original design document, not to AMD's finalized implementation at that. It's a significant disadvantage, just like AMD's disadvantage with SSE extensions through the years. Designing hardware from concept documents is only slightly less annoying than flat out reverse engineering.

I see Intel adopting x86-64 is just to avoid not having a product to meet an emerging demand. I still think that Intel is trying to merge Itanium and Pentium into a single new chip. (Not a dual core with one of each, but a true merging.) I still think that was what Yamhill was, to develop an acceptable EPIC emulation layer into a P4 design. That product could quite possibly be what Intel is waiting to name a Pentium 5. And no, I don't think that anyone agrees with me on this. It seems to be my hunch alone. If x86-64 grows too much it may not be worth it in the end, so it may not even come to pass. But that doesn't mean that it wasn't what Yamhill was trying to do.

<pre>"Build a man a fire and he's warm for the rest of the evening.
Set a man on fire and he's warm for the rest of his life." - Steve Taylor</pre>

slvr_phoenix · Jun 16, 2004

Hmm.. you mean people didn't line up in front of the stores to buy a retail copy when "32 bit" Windows 95 was launched, but they bought a new PC instead ?

That and this are completely different for some huge reasons:
1) There had been 32-bit capable systems sold for many years before Win95 was launched. In fact 32-bit hardware had been available for so many years that the vast majority of PC users were already capable of supporting 32-bit software, even if they didn't know it. Today however the percentage of people who can upgrade to a 64-bit OS without a new hardware purchase is very small because this push is running a heck of a lot faster.

2) There were 32-bit extensions to Windows before Win95. The appeak of Win95 was not 32-bitness, but the significant user interface improvements over proir versions of Windows. It was a massive OS improvement. However WinXP to WinXP-64 is a purely under the hood upgrade, much like the initial 32-bitness added to early versions of Windows. It won't be as cool. It will just allow you to fully utilize your 64-bit hardware.

So with even these two factors it makes judging the present based on the past rather unviable and imperfect.

it would be nice if it where a free upgrade/service pack like release, but knowing MS, its not that likely

I completely agree, on both counts. (That it would be nice and that it is highly unlikely.) I was highly pissed off when M$ forced people to 'upgrade' from Win95 to Win95b just to fix a number of bugs that were found as early on as the Win95 beta, and then did the same with Win98 and Win98SE. It's even arguable if Win98 was significantly different enough from Win95 that it warranted paying for a whole new version of Windows, but I'll grudgingly give M$ that one.

Frankly, I was surprised that they didn't find a way to force people to pay for WinXP SP1. I'm even more surprised that WinXP SP2 is likely to be free as well. It completely goes against M$'s SOHO track record. But then, considering that WinXP is more based on a workstation version of instead of a SOHO version of windows, maybe it shouldn't be as surprising as it is. I don't know. It's a tough call how to treat a product when you turn a workstation product into your SOHO product. :\

I would actually be impressed if M$ released the 64-bit WinXP upgrade as something like a free SP3 that turned WinXP into a hybrid 32/64 OS (much like Win3.11, but in a cleaner way). I also highly doubt that such will happen. M$ has been getting miffed at releasing so many significant improvements to XP for free.

What is worse though, is that your best bet might be to have both 32 and 64 bit windows versions for while, unless MS really improves their 64 bit version over the current beta.

I'd like to think that they might actually improve significantly over the beta. I won't be surprised if they don't, but I hope that they will. They've certainly had enough time. Other than playing Pong, what could they possibly be delaying it this long for then?

That is most likely true, but not because of your argument. Simply because I expect Intel and MS to release their 64 bit products more less at the same time (late this year). Since intel enjoys a 85% or something marketshare, its obvious the majority of 64 bit hardware won't be sold until intel sells it

Well, there is that too. :O

But if they had 64 bit P4's out today, I am not too sure XP64's release later this year would change that much. People would buy the 64 bit P4's today just as well, and hapily run XP32 on it.

Maybe, but I think that would depend a lot on how the 64-bit products were launched. If cheaper and equally fast products were still available (kind of like Northwood vs. Prescott or AXP vs. A64) then I don't think many would be buying 64-bit yet anyway. So long as 64-bit remains expensive you won't see too many people going for it, at least not until marketing folks at Dell and whatnot can say that it's actually a full 64-bit platform and maybe even lie like Apple did.

I don't think it will be very different from the windows 3.11->windows 95 transitions or windows9x->Win2k/XP. Some people will switch the same day and promise you that new OS is the best thing since sliced bread, other will complain, migrate slowly, point to lackluster software/driver support, say it aint worth it, tell you its slower, a memory hog,stick to the old windows for 5 years or longer. Also, some people will buy the retail copy, come home and discover it doesnt install on their "nearly new" $3.000 P4 system and they will be pissed.

Software support will follow a similar pattern, next year you'll see some native apps ported, but mostly everything will still be made available for XP32, other apps might just hop on the marketing bandwagon, and put "XP-64 ready" stickers on the boxes, even if its just 32 bit software,.. but in spite of all naysayers, in a few years, just about every new computer will be 64 bit capable and ship with a 64 bit OS, and slowly, high end apps will stop supporting the 32 bit legacy.

I think (hope?) that you're wrong here.

1) The shift from WinXP to WinXP-64 should be a hell of a lot smoother when it comes to drivers and software. If it's not, then M$ really failed to do their job and learn from past lessons.

2) WinXP to WinXP-64 will be a very minor OS functionality change. This is extremely different from Win3.11 to Win95, and just as different from Win9x to Win2K/XP. Those were extreme OS changes. This isn't nearly as extreme. It's really not much different than just having added the 32-bit extensions to Win3, except that this time it should be even smoother.

Sure, you'll still probably have the people having the problems that you mentioned, but this time around it should be a lot smoother than before. There shouldn't be nearly as many people having those problems, if anyone at all in some of those cases.

It may take 10 years or longer though (possibly forever) until virtually all software is 64 bit. but who cares anyway? I'm not waiting for a 64 bit version of Office or WinZIP, but for 64 bit encoders, games, and some olap tools.

That is another important factor to the transition. There were a lot of good reasons to go to 32-bit. There are a lot less good reasons to go to 64-bit. Oh sure, for some people there are still some good reasons, but the biggest really is just the memory handling, and that could have been fixed without a full 64-bit transition. So it will be interesting to see how this cycle actually goes. On the one hand the hardware saturation and the limitations are a lot less this time around so it should take much more time for majority acceptance as before. On the other hand the non-execution handling and the larger memory capacity may make the push into acceptance faster. Time will tell. But certainly a purist absolute 64-bitness is a long way away and is also highly unimportant to achieve. Hopefully if M$ pushes a purist 64-bit OS then it will be a lot better done than their 32-bit purist WinME was.

You know, it's strange, but I have this nagging feeling that the push into 64-bitness will come more from game developers than from any other direction, Intel and AMD, Dell and HP included. **ROFL**

<pre>"Build a man a fire and he's warm for the rest of the evening.
Set a man on fire and he's warm for the rest of his life." - Steve Taylor</pre>

Stimpy · Jun 16, 2004

64bit on the general office PC is not required. Hell most companies are starting look at very thin client PCs with zero management, be is Suns Java desktop, Xwindows, citrix, terminal server, etc, etc
It makes sence for AMD to churn out small die 32bit chips at a cheap price for these markets and the emerging ex-3rd world markets.

P4Man · Jun 16, 2004

>So with even these two factors it makes judging the present
>based on the past rather unviable and imperfect.

My point was just that people do buy retail OS's, and do not stick to whatever their PC came with forever. So your point "1" is irrelevant, of course less people will buy XP64 this christmass than people bought Win95 ten years ago. But a lot of people that have the hardware, will buy the OS.

You do have a valid point about windows 95 being a much more compelling upgrade though, no argument there. But still, windows95 was hyped not so much for its new GUI as for its "32 bitness" which "allowed you to finally multitask properly" (which was sort of true as well, it did finally support preemptive multitasking, not that it had anything to do with 32 bitness, but remember printing in win 3.1 ?)

> They've certainly had enough time. Other than playing
>Pong, what could they possibly be delaying it this long for
>then?

Oh but if it arrives this year, it won't even be a year late ! When was the last time MS released an OS that was not *at least* 2 years overdue ? How long did intel wait for the IPF version ? Wasnt windows 95 something like 4 years overdue ? are you still expecting longhorn this decade ?

>1) The shift from WinXP to WinXP-64 should be a hell of a
>lot smoother when it comes to drivers and software. If it's
>not, then M$ really failed to do their job and learn from
>past lessons.

I disagree. XP64 breaks driver compatibility, which may hurt initially, but is a good thing in the long run. I very much prefer such a disruptive approach over windows 95 compatibility model that ensured we felt 16 bit pains until nearly 15 years after 32 bit cpu's became mainstream. I'm all for backward compatibility on the application level, but I don't want ugly hacks in my kernel to allow some people to use 10 year old crap drivers. But this may well result in a rough ride initially. So be it. I felt those pains when I moved to NT as well, had to replace my scanner among other things, but all in all, I was quite a bit more happy with NT than wintendo95.

>2) WinXP to WinXP-64 will be a very minor OS functionality
>change

I disagree again. Its a complete rebuild and redesign of the kernel. AMD64 does not change things dramatically on the application level, but it requires very significant changes on the OS kernel level. In fact, I'm not sure you are aware, but XP64 is based on the Itanium kernel, not the x86 kernel. Its not a quick hack/recompile and it can't be.

>You know, it's strange, but I have this nagging feeling
>that the push into 64-bitness will come more from game
>developers than from any other direction, Intel and AMD,
>Dell and HP included. **ROFL**

For the desktop market, yes, I agree, and its not surprising at all. Games have been an extremely important factor in driving this business forward. If you exclude the corporate market, I wouldnt be surprised if over 50% of mid, to high end computers are mainly purchased for gaming.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

slvr_phoenix · Jun 16, 2004

My point was just that people do buy retail OS's, and do not stick to whatever their PC came with forever. So your point "1" is irrelevant

I don't see how it is irrelevant that most people who want the new OS won't have the hardware to run it so they'll have to buy a whole new PC. (Much unlike when Win95 came out.)

But yes, some people do buy retail OSs to upgrade their system. That's how I have a license of both Win95 and Win98SE with no home. The P133 that they used to run on has died and gone to PC heaven. It's funny to think that I was originally using DOS 6.22 and Windows 3.11 on that box. It'd almost be worth trying to revive just for laughs.

but remember printing in win 3.1 ?

Just about every form of multitasking in Win3.11 was a joke. I seem to remember a version of DOS (not M$) that did better multitasking than that.

Oh but if it arrives this year, it won't even be a year late ! When was the last time MS released an OS that was not *at least* 2 years overdue ? How long did intel wait for the IPF version ? Wasnt windows 95 something like 4 years overdue ? are you still expecting longhorn this decade ?

**ROFL** Too true. Hmm ... Longhorn ... that's a scary topic in and of itself. Luckily it probably won't show up nearly as frightening as it could just because they'll have to scrap a lot in order to release it sometime before Satan ice skates to work.

I disagree. XP64 breaks driver compatibility, which may hurt initially, but is a good thing in the long run. I very much prefer such a disruptive approach over windows 95 compatibility model that ensured we felt 16 bit pains until nearly 15 years after 32 bit cpu's became mainstream.

I'm sorry, but I've got to totally disagree. I never felt those 16-bit pains and thought that M$ couldn't have done anything dumber than what they did to make WinME. The only driver problems that I ever knew were not 16-bit problems, but idiots-who-couldn't-program-their-way-out-of-a-cardboard-box-writing-drivers problems. Done right there's no reason why the 32-bit drivers couldn't still be supported, and there would be a heck of a lot less pissed off people if it were done that way. There's nothing worse than being unable to upgrade your OS because some of your hardware won't have drivers if you do. And I'm not just talking about vendors who haven't gotten around to writing the new drivers yet, but vendors who simply don't support that product anymore at all.

I disagree again. Its a complete rebuild and redesign of the kernel. AMD64 does not change things dramatically on the application level, but it requires very significant changes on the OS kernel level. In fact, I'm not sure you are aware, but XP64 is based on the Itanium kernel, not the x86 kernel. Its not a quick hack/recompile and it can't be.

I knew that it was based on the XP64 kernel, but then Itanium (badly) supports 32-bit x86 instructions and so does an Athlon 64, so obviously the kernel must support both, and thus should be capable of running both 32-bit and 64-bit hardwares if M$ programmed it correctly. So functionally it should be pretty much the same, even if the kernel level changes significantly to support the differences in the hardwares.

For the desktop market, yes, I agree, and its not surprising at all. Games have been an extremely important factor in driving this business forward. If you exclude the corporate market, I wouldnt be surprised if over 50% of mid, to high end computers are mainly purchased for gaming.

I could be biased about that 50% number because I'm a scientific programmer and know a lot of people who do very technical work on their PCs and thus enjoy more power. Most of them (sadly) don't even game. Even most of my games are old. I only just started up a new character in Arcanum. :\ Strange, I know.

But then it seems to me that more and more households have enough disposable income to provide not just family PCs, but child-specific PCs. And since kids pretty much only want to play games, there you go. I can't say yes or no to 50%, but I definately see a shift to a more game-driven purchase cycle.

<pre>"Build a man a fire and he's warm for the rest of the evening.
Set a man on fire and he's warm for the rest of his life." - Steve Taylor</pre>

P4Man · Jun 17, 2004

>Just about every form of multitasking in Win3.11 was a
>joke. I seem to remember a version of DOS (not M$) that did
>better multitasking than that.

Neither Windows nor DrDOS (the dos version you are referring to) really supported multitasking. They both supported primitive taskswitching only.

>Longhorn ... that's a scary topic in and of itself.

Longhorn doesn't scare me, Palladium does.. food for another topic.

>I'm sorry, but I've got to totally disagree. I never felt
>those 16-bit pains <..> The only driver problems that I
>ever knew were not 16-bit problems, but
>idiots-who-couldn't-program-their-way-out-of-a-cardboard-bo
>x-writing-drivers problems.

No, the problem was maybe not the drivers as such. the problem was the kernel allowed 16 bit drivers in the first place, and therefore needed switched the cpu back and forth between 16 bit virtual-86 mode and 32 bit protected mode, which allowed misbehaving apps (or drivers) to crash the kernel and bluescreen windows. THAT was the problem. Its IMHO the main reason just about any app could (and would) crash the OS under Win9x, while that is pretty hard to do with an NT kernel. That and the fact quite a bit of the OS itselve was still 16 bit (Virtual 8086) legacy stuff.

> Done right there's no reason why the 32-bit drivers
>couldn't still be supported

Yes there is. XP64 doesn't support kernel-mode 32-bit code, and for good reason: kernel mode code gives you access to the virtual memory manager, and for instance, 32 bit drivers may not cope with the fact the VMM maps their memory beyond 2 or 4 GB. To support this, MS would have to screw the flat memory model, create memory holes or whatever. I'm honestly glad they didn't bother. Anything that requires kernel mapped memory (like hardware devicedrivers) ought to be aware of the fact they are running in a 64 bit address space. BTW, AFAIK, this isn't any different under Linux or any other OS. Bashing MS is popular, but at least it seems they are getting some of their priorities right with SP2, and XP64. Stability and security over backwards/legacy compatibility, I can only applaud them for that.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

P4Man · Jun 17, 2004

To get this thread back on track.. :
>"Bad news! AMD Sempron WILL NOT support 64bit "

It seems this either applies only to K7 based Semprons, or only temporarly for K8 based semprons:

"Microsoft expects significant adoption of 64-bit systems, and by the end of 2005, all of the processors sold by AMD and the majority of processors sold by Intel Corp. will support 64-bit computing, Gates predicted

Tim Wright, director of strategic marketing at Sunnyvale, Calif.-based AMD, agreed with Gates' prediction. "By the end of 2005, all we will be shipping is AMD 64," he said. "

<A HREF="http://www.computerworld.com/hardwaretopics/hardware/story/0,10801,92926p2,00.html" target="_new"> computerworld </A>

= The views stated herein are my personal views, and not necessarily the views of my wife. =

trooper11 · Jun 17, 2004

ill say this yet agian, i wasnt particularly pointing at your comments, but of others ive heard that are attacks, not crticisms, and yes i know the difference lol.

an attack is when those that make the comments like to make it sound like amd is the only company making bad decisions or making bad choices, while standing behind intel's. its just as bad when those do it to intel while sitcking behind amd. i just hate the whole fanboy stuff that can go on.

i never said you couldnt criticise, thats the only way aynthing can be improved apon, i didnt you shoudlnt do that. you were doing fine, but some try to blow it out of proportion and make it baseless, loosing all credibility of just trying 'attack'

just know im not trying to pick a fight with you lol, your fine .

yes i know about all the prescott threads,countless people have tlaked alot about such things, im not blind to that. like i said, it could just be since amd has alot of releases recently that its getting more attention. im just saying that the prescott discusion is not quite as high as it was and amd is more dominating the headlines anyway at the moment. like i i said, its probably just becuase of that.

as far as the sempron goes, it sounds liek amd will roll out the line first without 64bit options. i have a feeling they will try it out on socket A and s754 without 64bit capabilites, but then add those when they migrate to s939, kind of an added bonus for upgrading to them.

i also would think that by the time sempron does hit s939, that will spell the end of socket A

TheRod · Jun 17, 2004

What's so funny about this, is the fact that some analysts thinks that the Microsoft move to AMD64 is to help reduce the influence of Intel in the industry... Is there any "tech-conspiracy" web page out there?

--
It's tricky to use words like AMD or Intel in a signature some users could think your are biased.

P4Man · Jun 17, 2004

>What's so funny about this, is the fact that some analysts
>thinks that the Microsoft move to AMD64 is to help reduce
>the influence of Intel in the industry...

I'll tell you this: every move MS makes is to help increase its own influence over the industry.

MS benefits from a healthy competition in the cpu market, as that helps sell more (newer, faster, cheaper) machines, which in turn leads to more copies of windows being sold. So its fair to assume MS will not help intel increase its monopoly power if there isn't some major gain in it for MS itselve.

Itanium was a way for MS to get into the enterprise market they so very much want to crack, iAMD64 is a way to move x86 (and therefore windows) further up the foodchain as well. Both platforms make perfect sense for MS, no need for conspiracy theories in this regard.

However, using the same logic, its also reasonable to assume MS would not have supported an incompatible 64 bit x86 ISA from Intel. It supported AMD64 because it fit perfectly into their windows strategy (not because they love AMD so much), and because intel wasnt about to release their own 64 bit extentions to protect Itanium. MS knew Intel would have to follow them regardless, so adopted AMD64 and maybe, if you love conspiracies, made a gentlemens agreement with intel not to release the OS until intels reworked chips where ready.

= The views stated herein are my personal views, and not necessarily the views of my wife. =

TheRod · Jun 17, 2004

if you love conspiracies, made a gentlemens agreement with intel not to release the OS until intels reworked chips where ready.

I like this conspiracy theory, and it's quite possible... This can be called a near-reality conspiracy theory.

--
It's tricky to use words like AMD or Intel in a signature some users could think your are biased.

TheRod · Jun 17, 2004

Bashing MS is popular, but at least it seems they are getting some of their priorities right with SP2, and XP64. Stability and security over backwards/legacy compatibility, I can only applaud them for that.

I totally agree that Microsoft makes good move. Now they have nearly 100% of the market share. They can concentrate on security/stability and specific market needs.

The only bad move they made (my opinion) in the last months was to extends Win9x support.

I can't get rid of those people who thinks that Win9x is better than WinXP... These people are often complaining about WinXP not able to run "Cereal Box software". Damn! They don't understand that it's not WinXP fault... Or they still install and play 1995 games that were running in DOS. I don't want to hear anymore about those issues!!!

--
It's tricky to use words like AMD or Intel in a signature some users could think your are biased.

P4Man · Jun 17, 2004

> like this conspiracy theory, and it's quite possible...
>This can be called a near-reality conspiracy theory

Well, to tell you the truth, i think MS, if it made such a deal, really didn't sacrifice a lot, since they aren't ready with the OS anyway..

= The views stated herein are my personal views, and not necessarily the views of my wife. =

P4Man · Jun 17, 2004

>ese people are often complaining about WinXP not able to
>run "Cereal Box software"

LOL !! Indeed !

>They don't understand that it's not WinXP fault

They don't need to understand what causes it. Point is, they are right ! and if running cereal box software is so important to them, it may make sense to either hang on to Win9x, or at least have a dual boot. *I* gladly give up some compatibility for increased security, stability or performance, but I can't expect others to feel the same way.

Of course, the price for being able to run cerealbox soft is that they have to go elsewhere for support, cause I aint supporting it no longer, no way. Let them reformat and reinstall their own machines, I've had my share of that

= The views stated herein are my personal views, and not necessarily the views of my wife. =

TheRod · Jun 17, 2004

Of course, the price for being able to run cerealbox soft is that they have to go elsewhere for support, cause I aint supporting it no longer, no way. Let them reformat and reinstall their own machines, I've had my share of that

I do really understand you. I don't build PC everyday but when I build a PC for a friend or family I'm clear on one point : I know how it works and I will "control" the software that will be installed on the PC. And I'm clear that if they uninstall or decativate the anti-virus or the firewall. I will not supports them anymore! And it surprise me on how well it works. I take some time to explain them that it's important to keep firewall/AV/OS up-to-date to avois problems in the future. Education is the KEY to a better world!

--
It's tricky to use words like AMD or Intel in a signature some users could think your are biased.

AMD Sempron without 64bit = big mistake (UPDATED)

Splendid

Distinguished

Splendid

Distinguished

Splendid

Splendid

Distinguished

Distinguished

Splendid

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Distinguished

Share this page