Sign in with
Sign up | Sign in

Your 64-Bit Check List: Potential Issues You Might See

Your 64-Bit Check List: Potential Issues You Might See

Are You Running a 64-bit OS?

If you don’t know the answer, then you probably aren’t. First of all, 32-bit and 64-bit Windows versions do not behave differently; they look and feel the same. The easiest way to find out the exact version of your operating system is to look at the system properties under Control Panel -> System (right-click on your Computer and select “Properties”). The window that appears will tell you the details of your Windows version, your hardware, and the system type (see the screen shot above). In this case, it’s a “64-bit Operating System." But what’s the difference between 32 and 64 bits?

Differences Between 32-Bit and 64-Bit Systems

When we talk about 64-bit systems, we have to differentiate between 64-bit computing (processing), which is important for achieving high performance, and 64-bit addressing to support large amounts of memory. A 64-bit system has 64-bit wide registers, supporting 64-bit data types, and it addresses data in 64 bits internally (16 exabytes max, instead of 4 gigabytes). External addressing and buses may be different, though; an example is memory addressing, which is typically limited to 40 or 48 bits.

A 64-bit OS requires a 64-bit processor to operate. Most 64-bit systems can execute 32-bit software in a so-called “compatibility mode,” which is really important due to the fact that native 64-bit applications are still rather rare. The processor switches to the 32-bit mode for individual clock cycles when necessary. Running a 32-bit OS on a 64-bit CPU typically means that the processor runs in legacy mode all the time. While 64-bit software can perform better on 64-bit operating systems if it is well implemented, 32-bit applications run at roughly the same performance on 64-bit systems.

64-Bit Advantages

There are a few items that need to be mentioned, as 64-bit systems do in fact have several advantages that go beyond the issues of processing and addressing. First, a 32-bit version of Windows is limited to a maximum of 4 GB RAM, and so will not give you the full memory for your applications—operating system processes are mapped into the address space, resulting in an effective memory capacity of 3 GB, or sometimes a bit more. Effectively, the maximum memory capacity is limited to only 3+ GB. A 64-bit version of Windows will give you access to more available memory as you add it.

Secondly, the handling of large files is simpler on 64-bit systems with lots of memory. Think of a 5 GB file on a 32-bit Windows system with only 3 GB of available RAM: the system has to remap the required region of the file into the available memory, and might end up remapping all the time.

Finally, there are scientific applications that do not deliver sufficiently precise results unless there are a sufficient number of bits for floating point operations. In these cases, 64-bit applications on 64-bit operating systems are the only real choice.

64-Bit Disadvantages

The downsides to 64-bit computing are more complex memory handling, effectively leading to higher memory capacity requirements, and a lack of 64-bit software in general. On the one hand, not all applications benefit from being ported to 64 bits. On the other hand, low-level components, such as drivers, are still not available for all of the devices that you might need to operate. Drivers operate between the hardware and the operating system, which prevents almost all of them from being executed in 32-bit compatibility mode. In other words: if there are no 64-bit drivers for a scanner, printer, video editing card or other device, you will not be able to use it in a 64-bit environment.

Ask a Category Expert

Create a new thread in the UK Article comments forum about this subject

Example: Notebook, Android, SSD hard drive

Display all 8 comments.
This thread is closed for comments
  • 0 Hide
    mi1ez , 16 June 2009 16:41
    "The list of features introduced by the EFI includes DRM, [...]"

    Can you elaborate?
  • 0 Hide
    waxdart , 16 June 2009 16:48
    mi1ez"The list of features introduced by the EFI includes DRM, [...]"Can you elaborate?

    It's all about jamming DRM/Trusted Computing down everyone’s throat.

  • 0 Hide
    Tweedledum , 16 June 2009 17:38
    "operating system processes are mapped into the address space, resulting in an effective memory capacity of 3 GB". Thats not actualy correct, its that fact that the systems hardware, HDD, IDE Controler, Graphics Card, ect, all have memory addresses that have to come out of the same range as the physical memory. This is why 4GB often shows as ~3GB in the BIOS as well.
  • 0 Hide
    Anonymous , 17 June 2009 05:46
    The comment about "scientific applications that do not deliver sufficiently precise results" on 32 bit OSs is nonsense. Floating point processors since the 8087, which partnered the first generation of IBM PC processors, have offered hardware support for 64 bit floating point numbers. The main difference that 64 bit processing makes is to increase the address space, allowing bigger problems to be tackled.
  • 0 Hide
    Anonymous , 18 June 2009 10:58
    But why would you want to boot from a 2TB+ RAID array anyway?

    If I had a 2TB+ RAID array, I'd use it as secondary storage and boot the OS (be it WinXP, Vista, Linux) from a smaller HDD.
    Is it is now, I use a 120GB HDD for WinXP SP3 and 2 500GB HDDs in a RAID0 array for backup/storage/music/etc...
  • 0 Hide
    phcahill , 18 June 2009 18:49
    No 64bit version of office. No 64bit odbc drivers either. Not good when interfacing Sql Server to office data.
  • 0 Hide
    jwoollis , 22 June 2009 14:50
    Few if any people seem to have considered or mentioned the fact that if there are hardware compatability issues with 64bit Windows, it would be possible to run 32bit windows in a Virtual Environment such as VMware to support older applications. In respect of drivers, many USB devices can be used through a virtual machine by using a network based USB HUB such as Belkin F5L009 which bypasses the host connection issue altogether. It would be possible to connect many none USB device (serial, parallel, PCMCIA, ExpressCard etc) via a USB connection (albeit possibly at a reduced speed). This would leave support only for certain motherboard components and Expansion cards as the only real problem for supported drivers and limit choice to perhaps a generic Vista Driver or a replacement component.
  • 0 Hide
    jwoollis , 22 June 2009 14:56
    I am curious as to the potential to use any memory above the maximum limit (including memory remapped by using large memory devices such as video cards) for a 32bit OS where this limit varies between 3 and 4GB not counting remapped memory.

    Is it possible to reclaim some of this memory either by just using a RAMDRIVE or failing this, could it not be possible to use a Hypervisor to install multiple OS and split the physical memory between the two machines having either 2x2GB or 3GB+1GB or similar splits.