It's based on, IIRC, the EV67 system bus design, which is part of the Alpha architecture - so it contains a lot of material under license. AMD and IBM may have collaborated on the additional work done, but the contribution of the bus design can't be underestimated.
While SPARC and MIPS64 and other architectures now have Intel CPUs giving them a run for their money CPU-performance-wise, those systems still kick pretty much any Intel- or AMD-based system's ass in I/O bandwidth. AMD systems have a definite advantage over Intel, however, due to their licensing of the HyperTransport bus design from Alpha Processor Inc., or whoever the current holders are of the Alpha processor design - it really does kick a lot of ass. However systems like the SPARCs were designed with the ability to handle high I/O load in mind, and commmodity systems still aren't quite so much so.
Go download Creative's DOS drivers for the SB PCI128. VMware 4.5 provides a virtual ES1371 chip, which is the chip that the SB PCI128 card is built around. That should get you going.
Funny, I run multiple virtual machines (Win2K, FreeBSD 4.x, NetBSD, FreeBSD 5.x, etc.) simultaneously on my Athlon 1.33 GHz dualhead box at work (it has 2 GB of RAM), and I can run it quite comfortably. If you were running it on top of XP - which is kind of a piggy anyway - with XP _inside_ the VM - why the heck are you asking "uh, why's it so slow?". C'mon, this is a total gimme.
I've also used it for serial console netboot testing when I didn't have any real machines to experiment with (yay VMware PXE booting + bridging mode!), which I thought was pretty awesome. Very nice for getting a custom FreeBSD kernel generated for doing serial console right (grumble grumble).
The article was talking about VirtualPC for Windows, not VirtualPC for Mac. Please read it before you comment - it's been well known for awhile that the new release of VPC was going to include a native virtualization release for Windows/x86.
Though Win64 is, as usual I'm sure, a completely retooled API. Whereas those dirty UNIX hippies, they've managed to keep the same API. Funny, isn't it?
Progress would be good. However, I don't think Itanium is the way to go - it's a sinking ship in a lot of ways, and instead of patching the hull or abandoning ship, Intel just keeps bailing water. VLIW/EPIC turned out to be far from all it was cracked up to be (for many reasons). The processors are still ridiculously expensive. There are all-around better 64-bit architectures out there than Itanium. And given the current climate, we could do worse than x86_64.
Windows NT wasn't originally designed for x86. Hell, initially it was developed for a CPU that didn't even exist - when it was first being developed, it was targeted at Intel's i960 RISC architecture. However, because the i960 RISC chip was plagued with delays, it was ported to another architecture (I believe the first one was MIPS32). Dave Cutler's clean OS design (one of the major designers of DEC's VMS operating system, hired away by Microsoft) made this possible relatively quickly.
And by the way, the original NT moniker was actually a reference to the CPU simulator - named N-Ten - that the first i960-native builds of what became Windows NT ran on.
While you're right, Windows definitely was available for a variety of architectures, unfortunately there was a _serious_ shortage of software for Windows NT for PPC, MIPS and Alpha/AXP. A few Microsoft packages, like BackOffice, Visual C++, and a few other things, were available; most third-party software, however, was not ever built for anything but x86. The only reason the Alpha/AXP version had a longer lifetime, and apparently more software, was due to the FX!32 dynamic translation software that Digital developers created to run x86 binaries on Alpha. There was no real technical limitation to speak of, just momentum of Windows on x86.
Well, Microsoft has been working on an AMD64 port for longer than Sun has. While Solaris is already 64-bit clean, they have to get the entire OS up and running on AMD64 fairly quickly. Obviously they've hit a big milestone, so hopefully they'll be able to make their target. Of course, as the Register story mentions, they'll have a lot of negative momentum and impressions to counter even once the product is ready.
I work for a service provider. As I've said, I've seen BIND cache records for days, even weeks - way after DNS has changed for a domain, even after changing the master servers at the root. Any non-broken DNS server would have _expired_ the records, as it's supposed to do. I've since replaced BIND with PowerDNS on my employer's primary DNS servers (not just for this reason - also because it affords some very nice functionaily as far as allowing customers to host their DNS with us, and still manage it on their own schedule). We still run it at the office because it supports dynamic DNS, but I've considered adding dynamic DNS functionality to PowerDNS just to have the ability to get rid of BIND once and for all. It's a mess, and it needs to be replaced.
As I said, I would not say this if it weren't for the fact that I had _first hand_ encounters with BIND's brokenness. Hence why I expected someone to say I'm full of crap. But in this case, I know I'm right about this.
That's because BIND's caching mechanism is, and has for some time now been, painfully broken. BIND will cache records until way, way after their TTL is up, even though it's really not supposed to. I know some people will say "no, it's not, you're full of shit!", but I've experienced it firsthand, so I can tell you for sure - it's broken. Therefore I continue to suggest using other DNS server software - there are a lot of alternatives, so take advantage of them.
The whole idea of widescreen movies is that you see them the way they were filmed - which is in wide screen. Why chop off the sides of the picture to have more filled pixels, when really what you're seeing is less of the movie?
especially considering NT 4.0 was a 64bit OS on Alpha back in 1996
Wrong - Windows NT for Alpha/AXP was NOT 64-bit. It used the Alpha's 32-bit mode. This was a well-known issue at the time. (I was working for government environmental monitoring facility at the time, and we had some company come in and demo NT on an AlphaServer for us, so I learned a thing or two about it.)
Also, Microsoft may have internal builds of 64-bit Windows, but no shipping products for IA64 or x86_64 so far. That's right, not one. So yes, MS is very much behind the curve. Linux was 64-bit on Alpha some time ago, for example.
I don't necessarily think it's such a great idea for Gnome or other projects to go with Mono, but Apple? I mean, for all the talk, Microsoft is still "the bad guys" from where Apple sits. And yes, I know MS didn't develop Mono - Miguel de Icaza and other open-source developers have. However, Microsoft hasn't yet started playing hard-core intellectual-property games with C# and the.Net class libraries - and you know it's going to happen, you really can't deny it. It's just a matter of time. And I think it's a fight that Apple probably, if they're smart, wants to stay out of.
Well, you don't have to be _rich_ to be able to afford any of those things necessarily. You just need to get paid fairly. Unfortunately, things aren't fair, are they? But yeah, being rich would be nice...
Wow... Haven't we sung this song and danced this dance before with a previous ICH revision? I seem to remember Intel making a similar request/demand of board makers before because of MCH issues, with a similar offer of compensation. Hasn't Intel learned from past mistakes? C'mon guys... just one more reason to use non-Intel (AMD! PowerPC!) hardware - keeps you away from this sort of trouble.
You cannot write a java compiler in java, it is written (and funny, that) in this case in C++
That must be some mighty strong crack you be a-smokin' there. You do realize that the Sun Java compiler (aka javac) is written in Java, don't you? Please go download the Sun or IBM JDK and look for yourself - javac is just a wrapper that calls a class in the com.sun.tools.javac namespace to start the Java compiler. Now no more talking out of turn...
It's actually "The Spirits Within". And I saw it and enjoyed it (yeah, I know...), but one mistake they made that made the movie feel way too surreal was motion blur - way, way too much of it. To very unnatural levels. 3D FPS games and roller coasters don't make me motion sick, but watching that movie, that's what motion sickness has got to feel like.
Because MIT- and BSD- licensed code can be harvested at will for any good ideas/code it might have by commercial developers. And you know, unrestricted commercial use is every business' right, and you licensing your "open source/free software" under a license that doesn't allow good American companies to use it as they will is just wrong, and.. and... un-American.. or something...
I believe a famous research scientist (might've been Einstein?) once said that to make something "from scratch", that "first you must make the universe". Nothing ever occurs in a vacuum, and we get ideas from all sorts of places. Things we hear, things we see, people we talk to. Even things totally unrelated may help to spark an idea. So to say that anything anyone does is completely, totally, 100% unquestionably original is pretty hard to justify. Even Linus admits he was inspired by Unix - that's not a bad thing, even if this Ken Brown character wants to paint it that way.
Yeah, except for the fact that (a) it's then incredibly difficult to allow customers to manage DNS on their own - something that I've come to really appreciate (we have several customers who host their DNS with us, but want to manage their zone contents themselves), and (b) the way that software like cPanel does it is not a good solution (we have one customer who handles his own DNS on a box running cPanel, and I'm regularly having to fix that for him). Also, (c) the half-way solutions of making a database, and using a bunch of scripts to regenerate the zone files periodically is always a mess - if the scripts should break, updates don't get applied, but if they do, hand-editing the zone files isn't a viable option.
I use PowerDNS for our DNS servers at work, and I and our customers are very pleased with it. We have a frontend (that I wrote) that integrates with our billing system, so users can log in and make changes to their domains, and have them take effect immediately. They never have to worry about trailing dots, domain serial numbers, or getting the SOA format right, not to mention multiple CNAMEs assigned to a single name (which will cause BIND to throw the zone out) or other mistakes like that - our frontend prevents errors like that. It's made DNS provisioning and management so much easier - provisioning is error-free now. Why would anyone want to use BIND? Seriously?
Slashdot Mirror
It's based on, IIRC, the EV67 system bus design, which is part of the Alpha architecture - so it contains a lot of material under license. AMD and IBM may have collaborated on the additional work done, but the contribution of the bus design can't be underestimated.
While SPARC and MIPS64 and other architectures now have Intel CPUs giving them a run for their money CPU-performance-wise, those systems still kick pretty much any Intel- or AMD-based system's ass in I/O bandwidth. AMD systems have a definite advantage over Intel, however, due to their licensing of the HyperTransport bus design from Alpha Processor Inc., or whoever the current holders are of the Alpha processor design - it really does kick a lot of ass. However systems like the SPARCs were designed with the ability to handle high I/O load in mind, and commmodity systems still aren't quite so much so.
Go download Creative's DOS drivers for the SB PCI128. VMware 4.5 provides a virtual ES1371 chip, which is the chip that the SB PCI128 card is built around. That should get you going.
Did you read the article? This is specifically covering VirtualPC for Windows, not VirtualPC for MacOS.
Funny, I run multiple virtual machines (Win2K, FreeBSD 4.x, NetBSD, FreeBSD 5.x, etc.) simultaneously on my Athlon 1.33 GHz dualhead box at work (it has 2 GB of RAM), and I can run it quite comfortably. If you were running it on top of XP - which is kind of a piggy anyway - with XP _inside_ the VM - why the heck are you asking "uh, why's it so slow?". C'mon, this is a total gimme.
I've also used it for serial console netboot testing when I didn't have any real machines to experiment with (yay VMware PXE booting + bridging mode!), which I thought was pretty awesome. Very nice for getting a custom FreeBSD kernel generated for doing serial console right (grumble grumble).
The article was talking about VirtualPC for Windows, not VirtualPC for Mac. Please read it before you comment - it's been well known for awhile that the new release of VPC was going to include a native virtualization release for Windows/x86.
Though Win64 is, as usual I'm sure, a completely retooled API. Whereas those dirty UNIX hippies, they've managed to keep the same API. Funny, isn't it?
Progress would be good. However, I don't think Itanium is the way to go - it's a sinking ship in a lot of ways, and instead of patching the hull or abandoning ship, Intel just keeps bailing water. VLIW/EPIC turned out to be far from all it was cracked up to be (for many reasons). The processors are still ridiculously expensive. There are all-around better 64-bit architectures out there than Itanium. And given the current climate, we could do worse than x86_64.
Windows NT wasn't originally designed for x86. Hell, initially it was developed for a CPU that didn't even exist - when it was first being developed, it was targeted at Intel's i960 RISC architecture. However, because the i960 RISC chip was plagued with delays, it was ported to another architecture (I believe the first one was MIPS32). Dave Cutler's clean OS design (one of the major designers of DEC's VMS operating system, hired away by Microsoft) made this possible relatively quickly.
And by the way, the original NT moniker was actually a reference to the CPU simulator - named N-Ten - that the first i960-native builds of what became Windows NT ran on.
Yes, as soon as Tiger is released, you'll be correct. The current OS X release, however, is not 64-bit native.
While you're right, Windows definitely was available for a variety of architectures, unfortunately there was a _serious_ shortage of software for Windows NT for PPC, MIPS and Alpha/AXP. A few Microsoft packages, like BackOffice, Visual C++, and a few other things, were available; most third-party software, however, was not ever built for anything but x86. The only reason the Alpha/AXP version had a longer lifetime, and apparently more software, was due to the FX!32 dynamic translation software that Digital developers created to run x86 binaries on Alpha. There was no real technical limitation to speak of, just momentum of Windows on x86.
Well, Microsoft has been working on an AMD64 port for longer than Sun has. While Solaris is already 64-bit clean, they have to get the entire OS up and running on AMD64 fairly quickly. Obviously they've hit a big milestone, so hopefully they'll be able to make their target. Of course, as the Register story mentions, they'll have a lot of negative momentum and impressions to counter even once the product is ready.
Yo. Check this out, Jack.
I work for a service provider. As I've said, I've seen BIND cache records for days, even weeks - way after DNS has changed for a domain, even after changing the master servers at the root. Any non-broken DNS server would have _expired_ the records, as it's supposed to do. I've since replaced BIND with PowerDNS on my employer's primary DNS servers (not just for this reason - also because it affords some very nice functionaily as far as allowing customers to host their DNS with us, and still manage it on their own schedule). We still run it at the office because it supports dynamic DNS, but I've considered adding dynamic DNS functionality to PowerDNS just to have the ability to get rid of BIND once and for all. It's a mess, and it needs to be replaced.
As I said, I would not say this if it weren't for the fact that I had _first hand_ encounters with BIND's brokenness. Hence why I expected someone to say I'm full of crap. But in this case, I know I'm right about this.
That's because BIND's caching mechanism is, and has for some time now been, painfully broken. BIND will cache records until way, way after their TTL is up, even though it's really not supposed to. I know some people will say "no, it's not, you're full of shit!", but I've experienced it firsthand, so I can tell you for sure - it's broken. Therefore I continue to suggest using other DNS server software - there are a lot of alternatives, so take advantage of them.
The whole idea of widescreen movies is that you see them the way they were filmed - which is in wide screen. Why chop off the sides of the picture to have more filled pixels, when really what you're seeing is less of the movie?
especially considering NT 4.0 was a 64bit OS on Alpha back in 1996
Wrong - Windows NT for Alpha/AXP was NOT 64-bit. It used the Alpha's 32-bit mode. This was a well-known issue at the time. (I was working for government environmental monitoring facility at the time, and we had some company come in and demo NT on an AlphaServer for us, so I learned a thing or two about it.)
Also, Microsoft may have internal builds of 64-bit Windows, but no shipping products for IA64 or x86_64 so far. That's right, not one. So yes, MS is very much behind the curve. Linux was 64-bit on Alpha some time ago, for example.
I don't necessarily think it's such a great idea for Gnome or other projects to go with Mono, but Apple? I mean, for all the talk, Microsoft is still "the bad guys" from where Apple sits. And yes, I know MS didn't develop Mono - Miguel de Icaza and other open-source developers have. However, Microsoft hasn't yet started playing hard-core intellectual-property games with C# and the .Net class libraries - and you know it's going to happen, you really can't deny it. It's just a matter of time. And I think it's a fight that Apple probably, if they're smart, wants to stay out of.
Well, you don't have to be _rich_ to be able to afford any of those things necessarily. You just need to get paid fairly. Unfortunately, things aren't fair, are they? But yeah, being rich would be nice...
Wow... Haven't we sung this song and danced this dance before with a previous ICH revision? I seem to remember Intel making a similar request/demand of board makers before because of MCH issues, with a similar offer of compensation. Hasn't Intel learned from past mistakes? C'mon guys... just one more reason to use non-Intel (AMD! PowerPC!) hardware - keeps you away from this sort of trouble.
You cannot write a java compiler in java, it is written (and funny, that) in this case in C++
That must be some mighty strong crack you be a-smokin' there. You do realize that the Sun Java compiler (aka javac) is written in Java, don't you? Please go download the Sun or IBM JDK and look for yourself - javac is just a wrapper that calls a class in the com.sun.tools.javac namespace to start the Java compiler. Now no more talking out of turn...
It's actually "The Spirits Within". And I saw it and enjoyed it (yeah, I know...), but one mistake they made that made the movie feel way too surreal was motion blur - way, way too much of it. To very unnatural levels. 3D FPS games and roller coasters don't make me motion sick, but watching that movie, that's what motion sickness has got to feel like.
Because MIT- and BSD- licensed code can be harvested at will for any good ideas/code it might have by commercial developers. And you know, unrestricted commercial use is every business' right, and you licensing your "open source/free software" under a license that doesn't allow good American companies to use it as they will is just wrong, and.. and... un-American.. or something...
</sarcasm>
I believe a famous research scientist (might've been Einstein?) once said that to make something "from scratch", that "first you must make the universe". Nothing ever occurs in a vacuum, and we get ideas from all sorts of places. Things we hear, things we see, people we talk to. Even things totally unrelated may help to spark an idea. So to say that anything anyone does is completely, totally, 100% unquestionably original is pretty hard to justify. Even Linus admits he was inspired by Unix - that's not a bad thing, even if this Ken Brown character wants to paint it that way.
Yeah, except for the fact that (a) it's then incredibly difficult to allow customers to manage DNS on their own - something that I've come to really appreciate (we have several customers who host their DNS with us, but want to manage their zone contents themselves), and (b) the way that software like cPanel does it is not a good solution (we have one customer who handles his own DNS on a box running cPanel, and I'm regularly having to fix that for him). Also, (c) the half-way solutions of making a database, and using a bunch of scripts to regenerate the zone files periodically is always a mess - if the scripts should break, updates don't get applied, but if they do, hand-editing the zone files isn't a viable option.
I use PowerDNS for our DNS servers at work, and I and our customers are very pleased with it. We have a frontend (that I wrote) that integrates with our billing system, so users can log in and make changes to their domains, and have them take effect immediately. They never have to worry about trailing dots, domain serial numbers, or getting the SOA format right, not to mention multiple CNAMEs assigned to a single name (which will cause BIND to throw the zone out) or other mistakes like that - our frontend prevents errors like that. It's made DNS provisioning and management so much easier - provisioning is error-free now. Why would anyone want to use BIND? Seriously?