Merced has been "RSN" since at least 1995. Now we hear that the Itanium is really just a beta chip to wet the markets, and the real good version is going to be the Itanium II.
They're having lots of trouble, but given a 32-bit chip that can scale up up up, they can hold on. Don't forget that their current share in the 64-bit big iron market is 0%, so Itanium represents a new market, and one that has much more to do with Sparc and Alpha than anything AMD puts out. (Sledgehammer will be a consumer chip.)
Except that the Pentium4 is on a new core, so the P6-core wasn't the last Intel ever made.
Looke is right -- Intel was already hyping "Merced" even back in 1995, and telling customer that the P6 would be the last 32-bit chip that Intel would ever make. (Hell, they were already making background noise about the 64-bit EPIC project with HP around the time the Pentium P5 shipped.)
Fact was, they've been working on the P7 core in the background for a long time, which turned out to be a smart investment considering the questionable state of Itanium.
AMD probably doesn't have any short term scalability problems. By the time those 1.4 GHz P4s hit the mainstream market, AMD will most likely have a 1.4GHz chip out that is faster for real work.
(On a side note, Intel is running a hellava lot of "Blue Man" ads for the Pentium !!! during the football. That says that they've got a huge backlog of older chips and the PIV is not ready for prime time.)
The real problem comes for AMD when Intel starts getting up in the 4+Ghz range. Intel says that the PIV won't have any problem, AMD might have to come back with a new design (or rely on it's x86-64 features to move product.)
Couldn't keep this flamewar on a technical level, eh?
One thing to note is that Sun's SunRay thing uses the exact same 'remote framebuffer' idea that WTS uses. Even though this requires more network bandwidth and more server oomph, it has an appeal because it's much simpler and easy to implement compared to X11's complex and kludgy 'remote drawing' protocol.
Of course, until MS gets Win2000 terminal services stable, the argument is mooot anyway. (#4 isn't a joke above in our office...)
NuBus worked fine for it's intended purpose -- workstations costing $5000+.
Don't forget that Mac's were pushing 21" monitors at 24-bit back when PC games had to ship in both a CGA and VGA versions. Wasn't cheap, wasn't supposed to be either.
Since Microsoft sells a product called Interix, which is logo-certified as UNIX(tm), I find that legend doubtful.
Maybe it was true at one time, but the time-limit expired, or it was all over when MS finally sold their SCO shares. Anyway, you can bet that their experience with XENIX and Unix Industry Politics was one of the things that influenced them away from a Unix-base for Windows NT and towards a VMS style.
So's reading comprehension apparently. I get the analogy and the tech details, just pointing out the fruitlessness of such an effort by Microsoft.
I also understand the 10,000 view of the relationship between the NT kernel and the Win32 subsystem.
WINE (if coded by MS) doesn't lose any performance running Win32 apps, NT doesn't lose any performance running Win32 apps, so what's your point?
My point is that you're wrong - there's important semantic differences between NT and Unix. The Wine project has proposed Linux kernel patches to work around this problem in a non-portable way. And, silly, Win32 was designed to run on the NT kernel, so of course that's what it was optimised for.
So here it is again, real slow like: Microsoft has made a huge investment in software which indirectly depends on features found in the NT kernel. It would be non-trivial to move that software base to a different infrastructure without losing speed and efficency. Stop pretending that the mountain is going to come to Mohammad, even if for Allah, all is possible.
What about Microsoft Interix? I assume it runs X11 on top of the Unix subsystem (not quite right on the 'executive', sure) and not Win32.
Re:It's even more ludicrous...
on
Crack for Sale
·
· Score: 2
Most people who read Slashdot are college kids who got admitted to the CompSci department on their Math SAT scores alone, overlooking their embarassingly mediocre English score.
They then have proceeded to waste some of their best drinking years of their life by hanging out in the computer lab or on IRC and generally avoiding any sort of liberal arts or humanities class that they absolutely did not have to take. This is necesssary because those classes have many girls in them and professors that used to smoke dope 20 years ago, both of which make the Average Slashdot Reader very nervous and anxious to get back on IRC in the computer lab.
So, consequentally, when the Average Slashdot Reader gets his turn at moderation points, he is puzzled at the strange word "Insightful" and not quite sure what it means. He could go and look it up, but that would be something that a tweed jacketed dope smoking liberal arts professor would do. So feeling that "Score 5:Insightful" has a nice ring to it, he doles the label out onto posts which have a blunt sort of humor that the Average Slashdot Reader can recognize as not quite funny, but generally recognizable as humor. Since he can't quite articulate quite what that sort of post is, "Insightful" is good enough and then he moves on with his day, back on IRC or off to the cold girlless labs.
. IF they switched to linux kernel based operating system, they could lay off half their Windows developement staff and dispatch a quarter of those remaining to keep track of kernel changes
And what advantage would that give them? You'll note that Microsoft is now shipping Win2000 on 64-CPU boxes, which is getting up there into Sun's territory. Meanwhile, Linus+crew are still optimising for the 4-8 CPU case, which is the sweet spot of the market sure, but not sweet enough for IBM, SGI, and HP. Oh yeah, I'm sure that those three would be more than happy to do Microsoft R+D for them...
BTW, they are a company, so cost is always an object. There's nothing so unfixable about NT/2000 that would make it cheaper to switch kernels. Hell, it would probably be cheaper to keep the NT kernel and rebuild the entire user space from scratch (say, making it more modular and using text base config files and scripting languages) than it would be to put a Unix kernel underneath the existing NT user space.
So, let me get this: If Microsoft made a OS with a kernel kinda like WinNT's and a Win32 server kinda like WinNT's, they would be making something exactly like MacOS X?... Just checking.
The real reason this discussion is silly is that Microsoft has already made their bed -- when they chose not to build a Unix-deriviatve in the early 90s and instead chose hire a bunch of VMS engineers to design WinNT.
Everyone here is babbling about MS Office, but the real issue is things like Exchange and SQL Server which are also considerable profit centers for Microsoft. Any attempt to get these to run on a Unix-dervative would either require a considerable rewrite, or a emulation layer - which is exactly what MS FUDs Oracle and IBM about for their WinNT software. Furthermore performance would go into the toliet, leaving open the question whether Unix is really "better" if you are an existing MS BackOffice customer. Would you dump millions of dollars and 8 years of development and tuning to switch from a VMS-clone kernel to a Unix-clone kernel? It's not like the Pointy Hairs would even notice.
No, things Gnome-on-Windows are fun hacks to try to make Windows that some people might use to solve specific problems. They certainly aren't going to be very well supported, have no reflection on the quality of the actual Unix software, and you'd be a huge fool to make them any part of your "migration path".
If you want duel-environment fall-back migration capabilities, you'd probably be better off just moving to Linux and testing the heck out of your Windows programs on Wine, or just cough up the money for VMWare.
Microsoft is not immune to customer whims. If they were, we'd all have loaded up OS/2 in 1988 and Windows NT in 1993.
Fact is, they put millions of dollars into DOS-compatibility engineering for Windows 95, and then spent another wad of money developing DirectX (primarily so that game developers would drop DOS programs, which they really wanted to do because writing/buying your own hardware drivers sucks).
So, they did not just "stop making" DOS as they had tried to do in the past -- they went way out of their way to provide a more than adequate replacement.
Actually, that probably is how Microsoft treats 90,000 seat customers. However if you are some pidding small fry with only 9,000 seats, forget about it.
I strongly suggest you do some research about the disasterous history of Nuke plant construction in California. (Fault lines, backwards plans, rate pass-thrus for plants that never opened.) And then come back and blame it on the "greens".
I've seen the problem on Windows and Mac, so it wouldn't shock me if Linux/Unix versions had it too. (The Linux version has other troubles, of course.)
The whole Not Available on The Same Day argument is full of shit.
First of all, people who waited in line to buy it right away probably wanted to play it right away, not go home and fuck with driver installation for a few hours first.
Second, Q3A sold well on the first day, but it is still to this day a big retail seller. Maybe you bought it right away, but the vast majority of purchasers waited long enough that they could have gotten the Linux version.
Except that when the first versions of DirectX shipped, the game industry pretty much told Microsoft where to stick it, and stuck with DOS (on Windows 95).
And, you are deluded if you thought there was any sort of guarantee that people would switch off DOS. Microsoft had totally failed in DOS-elimination up to that point with both OS/2 and WinNT.
Re:Ship linux games on bootable CD with source.
on
id On Linux: Bad News
·
· Score: 1
My problem is that model for running games sucks goatse asss. If that's the best you can do, just pack it in now and join the NetBSD users in the cellar.
It's worth remembering that a few things happened all about at the same time:
NVidia shipped their TNT2 and 6 months later, the GeForce, thus ruling over the high-end market. Better yet, they promised Open Source drivers. Except that there were no working Linux drivers available at all during this period, and real open source drivers never shipped. Lots of Windows and Linux users rush to NVidia.
Id shipped Quake III on Linux at the same time as Windows. Except that for 3D, Linux was a mess of beta-quality stuff, none of which was a standard part of any OS distribution, and most drivers were missing.
No shock that Quake III didn't sell well on Linux, given the circumstances.
Slashdot Mirror
Merced has been "RSN" since at least 1995. Now we hear that the Itanium is really just a beta chip to wet the markets, and the real good version is going to be the Itanium II.
They're having lots of trouble, but given a 32-bit chip that can scale up up up, they can hold on. Don't forget that their current share in the 64-bit big iron market is 0%, so Itanium represents a new market, and one that has much more to do with Sparc and Alpha than anything AMD puts out. (Sledgehammer will be a consumer chip.)
Except that the Pentium4 is on a new core, so the P6-core wasn't the last Intel ever made.
Looke is right -- Intel was already hyping "Merced" even back in 1995, and telling customer that the P6 would be the last 32-bit chip that Intel would ever make. (Hell, they were already making background noise about the 64-bit EPIC project with HP around the time the Pentium P5 shipped.)
Fact was, they've been working on the P7 core in the background for a long time, which turned out to be a smart investment considering the questionable state of Itanium.
AMD probably doesn't have any short term scalability problems. By the time those 1.4 GHz P4s hit the mainstream market, AMD will most likely have a 1.4GHz chip out that is faster for real work.
(On a side note, Intel is running a hellava lot of "Blue Man" ads for the Pentium !!! during the football. That says that they've got a huge backlog of older chips and the PIV is not ready for prime time.)
The real problem comes for AMD when Intel starts getting up in the 4+Ghz range. Intel says that the PIV won't have any problem, AMD might have to come back with a new design (or rely on it's x86-64 features to move product.)
Couldn't keep this flamewar on a technical level, eh?
One thing to note is that Sun's SunRay thing uses the exact same 'remote framebuffer' idea that WTS uses. Even though this requires more network bandwidth and more server oomph, it has an appeal because it's much simpler and easy to implement compared to X11's complex and kludgy 'remote drawing' protocol.
Of course, until MS gets Win2000 terminal services stable, the argument is mooot anyway. (#4 isn't a joke above in our office...)
Unlike, say, NetWare as a DOS app, which can return to DOS.
NuBus worked fine for it's intended purpose -- workstations costing $5000+.
Don't forget that Mac's were pushing 21" monitors at 24-bit back when PC games had to ship in both a CGA and VGA versions. Wasn't cheap, wasn't supposed to be either.
Since Microsoft sells a product called Interix, which is logo-certified as UNIX(tm), I find that legend doubtful.
Maybe it was true at one time, but the time-limit expired, or it was all over when MS finally sold their SCO shares. Anyway, you can bet that their experience with XENIX and Unix Industry Politics was one of the things that influenced them away from a Unix-base for Windows NT and towards a VMS style.
English's a bitch, aint it?
So's reading comprehension apparently. I get the analogy and the tech details, just pointing out the fruitlessness of such an effort by Microsoft.
I also understand the 10,000 view of the relationship between the NT kernel and the Win32 subsystem.
WINE (if coded by MS) doesn't lose any performance running Win32 apps, NT doesn't lose any performance running Win32 apps, so what's your point?
My point is that you're wrong - there's important semantic differences between NT and Unix. The Wine project has proposed Linux kernel patches to work around this problem in a non-portable way. And, silly, Win32 was designed to run on the NT kernel, so of course that's what it was optimised for.
So here it is again, real slow like: Microsoft has made a huge investment in software which indirectly depends on features found in the NT kernel. It would be non-trivial to move that software base to a different infrastructure without losing speed and efficency. Stop pretending that the mountain is going to come to Mohammad, even if for Allah, all is possible.
Yeah, it's expecially likely because Apple Thought Of It First(tm)!
What about Microsoft Interix? I assume it runs X11 on top of the Unix subsystem (not quite right on the 'executive', sure) and not Win32.
Most people who read Slashdot are college kids who got admitted to the CompSci department on their Math SAT scores alone, overlooking their embarassingly mediocre English score.
They then have proceeded to waste some of their best drinking years of their life by hanging out in the computer lab or on IRC and generally avoiding any sort of liberal arts or humanities class that they absolutely did not have to take. This is necesssary because those classes have many girls in them and professors that used to smoke dope 20 years ago, both of which make the Average Slashdot Reader very nervous and anxious to get back on IRC in the computer lab.
So, consequentally, when the Average Slashdot Reader gets his turn at moderation points, he is puzzled at the strange word "Insightful" and not quite sure what it means. He could go and look it up, but that would be something that a tweed jacketed dope smoking liberal arts professor would do. So feeling that "Score 5:Insightful" has a nice ring to it, he doles the label out onto posts which have a blunt sort of humor that the Average Slashdot Reader can recognize as not quite funny, but generally recognizable as humor. Since he can't quite articulate quite what that sort of post is, "Insightful" is good enough and then he moves on with his day, back on IRC or off to the cold girlless labs.
. IF they switched to linux kernel based operating system, they could lay off half their Windows developement staff and dispatch a quarter of those remaining to keep track of kernel changes
And what advantage would that give them? You'll note that Microsoft is now shipping Win2000 on 64-CPU boxes, which is getting up there into Sun's territory. Meanwhile, Linus+crew are still optimising for the 4-8 CPU case, which is the sweet spot of the market sure, but not sweet enough for IBM, SGI, and HP. Oh yeah, I'm sure that those three would be more than happy to do Microsoft R+D for them...
BTW, they are a company, so cost is always an object. There's nothing so unfixable about NT/2000 that would make it cheaper to switch kernels. Hell, it would probably be cheaper to keep the NT kernel and rebuild the entire user space from scratch (say, making it more modular and using text base config files and scripting languages) than it would be to put a Unix kernel underneath the existing NT user space.
So, let me get this: If Microsoft made a OS with a kernel kinda like WinNT's and a Win32 server kinda like WinNT's, they would be making something exactly like MacOS X? ... Just checking.
The real reason this discussion is silly is that Microsoft has already made their bed -- when they chose not to build a Unix-deriviatve in the early 90s and instead chose hire a bunch of VMS engineers to design WinNT.
Everyone here is babbling about MS Office, but the real issue is things like Exchange and SQL Server which are also considerable profit centers for Microsoft. Any attempt to get these to run on a Unix-dervative would either require a considerable rewrite, or a emulation layer - which is exactly what MS FUDs Oracle and IBM about for their WinNT software. Furthermore performance would go into the toliet, leaving open the question whether Unix is really "better" if you are an existing MS BackOffice customer. Would you dump millions of dollars and 8 years of development and tuning to switch from a VMS-clone kernel to a Unix-clone kernel? It's not like the Pointy Hairs would even notice.
Forgot to mention that IHBT.
No, things Gnome-on-Windows are fun hacks to try to make Windows that some people might use to solve specific problems. They certainly aren't going to be very well supported, have no reflection on the quality of the actual Unix software, and you'd be a huge fool to make them any part of your "migration path".
If you want duel-environment fall-back migration capabilities, you'd probably be better off just moving to Linux and testing the heck out of your Windows programs on Wine, or just cough up the money for VMWare.
Microsoft is not immune to customer whims. If they were, we'd all have loaded up OS/2 in 1988 and Windows NT in 1993.
Fact is, they put millions of dollars into DOS-compatibility engineering for Windows 95, and then spent another wad of money developing DirectX (primarily so that game developers would drop DOS programs, which they really wanted to do because writing/buying your own hardware drivers sucks).
So, they did not just "stop making" DOS as they had tried to do in the past -- they went way out of their way to provide a more than adequate replacement.
Actually, that probably is how Microsoft treats 90,000 seat customers. However if you are some pidding small fry with only 9,000 seats, forget about it.
"IBM recommends Windows 2000 Professional for business" -- Every IBM PC hardware advert.
Only if you are running something like lame "winme", wimme. NT's shell line is in the registry.
Say in HKLM\Software\Microsoft\Windows NT\Current version\WinLogin\Shell. Strangely, this is a per-machine, not a per-user setting.
I strongly suggest you do some research about the disasterous history of Nuke plant construction in California. (Fault lines, backwards plans, rate pass-thrus for plants that never opened.) And then come back and blame it on the "greens".
I've seen the problem on Windows and Mac, so it wouldn't shock me if Linux/Unix versions had it too. (The Linux version has other troubles, of course.)
The whole Not Available on The Same Day argument is full of shit.
First of all, people who waited in line to buy it right away probably wanted to play it right away, not go home and fuck with driver installation for a few hours first.
Second, Q3A sold well on the first day, but it is still to this day a big retail seller. Maybe you bought it right away, but the vast majority of purchasers waited long enough that they could have gotten the Linux version.
Except that when the first versions of DirectX shipped, the game industry pretty much told Microsoft where to stick it, and stuck with DOS (on Windows 95).
And, you are deluded if you thought there was any sort of guarantee that people would switch off DOS. Microsoft had totally failed in DOS-elimination up to that point with both OS/2 and WinNT.
My problem is that model for running games sucks goatse asss. If that's the best you can do, just pack it in now and join the NetBSD users in the cellar.
It's worth remembering that a few things happened all about at the same time:
NVidia shipped their TNT2 and 6 months later, the GeForce, thus ruling over the high-end market. Better yet, they promised Open Source drivers. Except that there were no working Linux drivers available at all during this period, and real open source drivers never shipped. Lots of Windows and Linux users rush to NVidia.
Id shipped Quake III on Linux at the same time as Windows. Except that for 3D, Linux was a mess of beta-quality stuff, none of which was a standard part of any OS distribution, and most drivers were missing.
No shock that Quake III didn't sell well on Linux, given the circumstances.