The limit is lifted under 64-bit Windows, so 32-bit programs that could benefit from addressing a full 4GB should show improvements. Critics say the need for a full 4GB of address space is currently still rare.
Increased complexity
For enterprises, the advent of 64-bit Windows and 64-bit extended chip architectures means, if nothing else, more choice and more complexity, says Gartner's Enck. At the simplest level, 64-bit-capable CPUs have become mainstream and many companies will begin to use them by default, if they are not already. The other options are to migrate to 64-bit Windows while sticking with 32-bit applications, or to use 64-bit Windows and some 64-bit applications.
Servers, workstations and PCs with Windows x64 pre-installed are becoming readily available, with systems already available or coming soon from vendors such as Acer, Alienware, Dell, FSC, Fujitsu, HP, Hitachi, IBM, NEC and Unisys. Some manufacturers, such as Dell and HP, also have upgrade programmes in place for customers with 64-bit capable hardware wanting to shift to 64-bit Windows without voiding their warranties.
The 64-bit operating system requires 64-bit drivers, even if all the applications are 32-bit. This could be a problem, at least for the near future; while many drivers are already out there, quite a few vendors haven't yet made drivers available.
Software with embedded device drivers will need to be recoded, and 16-bit applications don't work at all; these two types of software make up 20 percent of the applications currently running on Windows, according to Microsoft.
Microsoft is planning to maintain a directory of drivers and applications that have passed an x64 compatibility test, and promises to have the driver issue solved — for new hardware at least — by the time Longhorn appears. The biggest driver headache, though, could turn out to be caused by older devices — companies could well find that no 64-bit driver is forthcoming for, say, that three-year-old printer.
Applications
The current situation with applications is even more rarified — the applications simply aren't there. This will begin to change later this year with Microsoft and others gearing up application launches, but some key gaps remain.
For instance, most 32-bit desktop-level firewalls and antivirus products are incompatible with 64-bit Windows. Symantec, for one, says it doesn't currently support 64-bit Windows; the company says it is monitoring 64-bit adoption, and will eventually support the new operating system in its security products, but has given no timeline. McAfee offers limited enterprise support for 64-bit systems, but won't offer full support until sometime next year.
Some other applications present strange quirks. Microsoft decided to include both 32-bit and 64-bit versions of Internet Explorer in Windows x64, for the very good reason that 32-bit add-ons are incompatible with the 64-bit browser. That means the 64-bit IE can't handle Flash, Java or toolbars, and must launch PDF files in an external application.
The desktop
Server applications are likely to begin appearing before those aimed at the desktop, though workstations with 64-bit operating systems are expected to make up a significant proportion of shipments next year.
Outside of specialised workstations carrying out operations like video processing or financial simulations, there is currently little need for desktop applications to address large amounts of memory, say industry observers. While 1 GB is currently plenty for most workstations, requirements are likely to rise as desktops handle more digital content; 4GB DVD files are increasingly common on the desktop, for instance. Such requirements could mean a real need for the vast amounts of memory enabled by 64-bit software.
Microsoft says that once 64-bit hardware and operating systems become more widely used, developers will start to come up with ways of using all that power. Just think — searching could be 1,000 times faster if all your documents are stored in main memory instead of on a physical disk, Microsoft's Allchin told the Windows Hardware Engineering Conference.
This approach seems backward to some. "From a developer perspective, this is creating a market as opposed to responding to a market requirement," says RedMonk's Governor. "At the moment it's a technology in search of an application."
Talkback
In your article "The importance of being 64-bit"
So what is all the fuss about, exactly? Technically speaking, a 64-bit chip is one that has integer registers that are 64 bits wide, allowing them to process 64 bits at a time. CPUs store the address of locations in virtual memory in integer registers. This means that the total amount of data the CPU can keep in its working area is determined by how wide these integer registers are.
Which isn't the case for the CPUs you are talking about. The X86 64 bit CPUs still have 32 bit integers. They have 64 bits of address space but the default integer word size is still 32 bits -- that's why it only increased the die size of AMD's original chip by something like 7%. If it really had a 64 bit integer word size the chip would have nearly doubled in size.
It is very sad that the AMD-Linux x86-64 initiative is completely hijacked by WinTel in your article. It is thus IMO a biased and incomplete mis-information.
BTW, to the previous comment, making "default" integer size 64-bit won't actually increase the die-size a bit. The ALU and registers in AMD64 chips are already 64-bit capable. It's the OS / compiler / programming langauge that dictates the 32-bit integer size now (and that is not an unreasonable default).