Intel 80486DX2

From Wikipedia, the free encyclopedia

An Intel i486DX2-66 Microprocessor, top view
An Intel i486DX2-66 Microprocessor, top view
... and bottom view with gold plated pins.
... and bottom view with gold plated pins.

The Intel's i486DX2 is a CPU produced by Intel that was introduced in 1992 . The i486DX2 was identical to the i486DX but for the addition of " clock doubling" technology. It was the first chip to use clock doubling, whereby the processor performs two clock cycles per single cycle of the memory bus. Essentially, the processor's speed is set to double the speed of the system bus. Because of this, an i486DX2 is faster than an i486DX-based system at the same bus speed; a normal DX chip performs one processor clock cycle per system bus cycle, while a DX2 chip performs two processor clock cycles per bus cycle. The chip is based on the Intel x86 architecture.

For many players of video games during the early and mid 1990s, towards the end of the MS-DOS gaming era, the i486DX2-66 was a very popular processor. Often coupled with 8 - 16 MB RAM and a VLB video card, the CPU was capable of playing every title available for several years after its release, making it a "sweet spot" in CPU performance and longevity. The introduction of 3D graphics spelled the end of the 486's reign, because of its heavy use of floating point calculations and the need for faster cache and more memory bandwidth. Developers also began to target the Pentium almost exclusively with assembly optimizations (e.g. Quake). An i486DX2-50 version was also available, but as the bus speed was 25Mhz rather than 33Mhz this was a significantly less popular processor.

i486DX2 CPU core
i486DX2 CPU core

There are two major versions of the DX2. Identified by P24 and P24D, the latter has a faster L1 cache mode called "write-back" that improves performance. The original P24 version only offers the slower "write-through" cache mode. AMD and Cyrix both produced a 486-level competition for the Intel i486DX2.

The first i486SX chips were i486DX chips that had faulty floating-point units during manufacture which were then disabled. Later SX chips were intentionally manufactured as such.