Some friends may have heard of 64-bit systems and 32-bit systems. We generally use 32-bit systems. Currently, the software on the market is dominated by 32-bit systems. 64-bit is used very rarely before XP. The reason is a 64-bit system, which requires software hard disk drive and other aspects of support. Now many friends are using Win7 system, the new system is very supportive of 64-bit, including the manufacturer's driver also has 64-bit, talked for a long time everyone May not understand the difference between them, the following Yu Fei will briefly introduce the difference between the two:
First, the original intention of the design is different:
64-bit operating system is designed to meet the mechanical design and Analysis, 3D animation, video editing and authoring, as well as customer needs in areas such as scientific computing and high-performance computing applications that require large amounts of memory and floating point performance. In short, the words are: they are the operating platform for high-tech personnel to use the special software of this industry. The 32-bit operating system is designed for the average user.
Second, the configuration is different:
64-bit operating system can only be installed on a 64-bit computer (CPU must be 64-bit). At the same time, 64-bit commonly used software needs to be installed to achieve the best performance of 64-bit (x64). The 32-bit operating system can be installed on a 32-bit (32-bit CPU) or 64-bit (64-bit CPU) computer. Of course, the 32-bit operating system is installed on a 64-bit computer, and its hardware is just like a "big horse car": 64-bit performance will be greatly reduced.
Third, the operation speed is different:
64-bit CPU GPRs (General-Purpose Registers, general-purpose registers) data width is 64 bits, 64-bit instruction set can run 64-bit data instructions, that is, the processor once Can extract 64-bit data (as long as two instructions, extract 8 bytes of data at a time), double the 32-bit (requires four instructions, extract 4 bytes of data at a time), theoretically the performance will be improved accordingly 1 times.
Fourth, the addressing capabilities are different:
The advantages of 64-bit processors are also reflected in the system's control of memory. Since the address uses a special integer, an ALU (Arithmetic Logic Operator) and registers can handle larger integers, which are larger addresses. For example, Windows Vista x64 Edition supports up to 128 GB of memory and up to 16 TB of virtual memory, while 32-bit CPUs and operating systems can only support up to 4 GB of memory.
Fifth, software popularity is different:
Currently, 64-bit commonly used software is much less than 32-bit commonly used software. The reason is simple: there are relatively few users using a 64-bit operating system. Therefore, software developers must consider the “input-output ratio” and invest limited funds into the software of more users. This is also an important reason why 64-bit software is relatively expensive (to spread costs into fewer offerings).
Pros and Cons A common misconception is that 64-bit architectures are no better than 32-bit architectures unless the computer has more than 4 GB of memory installed. This is not entirely true: Some operating systems reserve a portion of the process address space for the operating system to use, resulting in a reduction in the address space available to the user program for mapping memory. For example, Windows XP DLLs and userland OS components map to the address space of each process, leaving only 2 to 3.8 GB (depending on their settings) of available address space, even if the computer is loaded with 4 GB of memory. This limitation does not appear in 64-bit Windows. The memory mapping of files is no longer useful for 32-bit architectures, especially the introduction of relatively inexpensive DVD burning technology. Files larger than 4 GB are no longer rare, and such large files cannot be easily mapped to 32-bit architecture memory; only a portion of the file can be mapped to the address space, and files are accessed in memory map, when needed, These ranges must be mapped into or mapped out of the address space. This is a problem because ample memory mapping is still the most efficient way to access from disk to memory, if the operating system is properly implemented. The main disadvantage of 64-bit architectures is that, with respect to 32-bit architectures, consuming the same amount of data can eliminate more memory space (due to swollen pointers, as well as other types and alignment fillers). This increases the memory requirements of the process and can affect the use of the high-performance processor cache. Maintaining a portion of the 32-bit model is a processing method and is generally reasonable and effective. In fact, the high-performance z/OS operating system takes this approach, requiring the program code to be stored in any number in the 32-bit address space, and the data objects can be (optionally) stored in a 64-bit region. Currently the main commercial software is built on 32-bit code, not 64-bit code, so you can't get the advantage of a larger 64-bit address space on a 64-bit processor, or a wider 64-bit register and data path. However, users of free or free software operating systems can already use proprietary 64-bit computing environments. Not all applications require a large amount of address space or operate on 64-bit data items, so these programs do not enjoy the benefits of larger address spaces or wider registers and data paths; primarily benefit from 64-bit versions of applications. And will not enjoy the use of x86 version, there will be more registers available. Software availability 64-bit systems sometimes lack corresponding software, which is written to a 32-bit architecture. The most serious problem is the incompatible driver. Although most software is executable in 32-bit compatibility mode (also known as emulation mode, or Microsoft WoW64 technology), it is usually not possible to execute drivers (or similar software) because programs are typically executed between the operating system and the hardware. Direct simulation is not available. Many open source code software packages can be easily compiled from source code to executable on 64-bit operating systems such as Linux. The required condition is the compiler (usually gcc) that is supplied to the 64-bit machine. Most 64-bit versions of the driver are currently not available, and the lack of available drivers can be frustrating with 64-bit operating systems. Because the device's drivers are usually executed inside the operating system kernel (kernel), it is possible to execute the kernel as a 32-bit process while supporting 64-bit consumer processes. At the expense of additional consumption in the kernel, this provides users with the benefit of 64-bit memory and performance without compromising the binary compatibility of existing 32-bit drivers. This mechanism stems from the fact that Mac OS X enables 64-bit processes and supports 32-bit drivers. ps: This explanation applies to almost all current systems.