 The exposed die of an Intel 80486DX2 microprocessor. |
|
| Produced | From 1989 to 2007 |
|---|---|
| Common manufacturer(s) | Intel IBM |
| Max CPU clock | 16 MHz to 100 MHz |
| FSB speeds | 16 MHz to 50 MHz |
| Min feature size | 1, 0.8, 0.6 µm |
| Instruction set | x86 |
| Cores | 1 |
| Socket(s) | Socket 1 Socket 3 |
The Intel i486, otherwise known as the 80486, was the first tightly pipelined x86 design. Introduced in 1989, it was also the first x86 chip to use more than a million transistors, due to a large on-chip cache and an integrated floating point unit. It represents a fourth generation of binary compatible CPUs since the original 8086 of 1978, and it was the second 32-bit x86 design after the 80386.
A 50 MHz 80486 was reportedly able to perform 41 million instructions per second [1] and was able to reach 50 MIPS peak (see below).
(The i486 was so named, without the usual 80-prefix, because of a court ruling that prohibited trademarking numbers like 80486. Later, with the Pentium, Intel dropped number-based naming altogether.)
Contents |
[edit] Improvements
The instruction set of the i486 is very similar to its predecessor, the Intel 80386, with the addition of only a few extra instructions, such as CMPXCHG which executes the Compare-and-swap atomic operation and the XADD which executes the Fetch-and-add atomic operation. Though many atomic test-and-set instructions have existed since the 8086/8088, they did not correspond to the atomic instructions implemented in certain RISC processors, which made it harder to port some applications from these processors.
From a performance point of view, the architecture of the i486 is a vast improvement over the 80386. It has an on-chip unified instruction and data cache, an on-chip floating-point unit (FPU), and an enhanced bus interface unit. In addition, simple instructions (such as ALU reg,reg) execute in one clock cycle. These improvements yield a rough doubling in performance over the 386 at the same clock rate. A 386 (or 286) chip therefore has to reach 50 MHz to be comparable with low end parts in the 486 series[citation needed].
[edit] Differences between the 386 and 486
- An 8 KB on-chip SRAM cache stores the most commonly used instructions and data (16 KB and/or write-back on some later models). The 386 had no such internal cache but supported a slower off-chip cache.
- Tightly coupled pipelining allows the 486 to complete a simple instruction like ALU reg,reg or ALU reg,im every clock cycle. The 386 needed two clock cycles for this.
- Integrated FPU (disabled or absent in SX models) with a dedicated local bus gives faster floating point calculations compared to the i386+i387 combination.
- Improved MMU performance.
The 486 has a 32-bit data bus and a 32-bit address bus. This required either four matched 30-pin (8-bit) SIMMs or one 72-pin (32-bit) SIMM on a typical PC motherboard. The 32-bit address bus means that 4 GB of memory can be directly addressed.
The Intel project manager for the 80486 was Pat Gelsinger.
In May 2006 Intel announced that production of the 80486 would cease at the end of September 2007.[2] Although the chip had long been obsolete for personal computer applications, Intel had continued production for use in embedded systems.
[edit] Models
There are several suffixes and variants including:
- i486DX — The original chip (without any clock doubling).
- i486DX-S — SL Enhanced 486DX
- i486DXL — A 486DX with SMM (System Management Mode), stop clock, and power saving features.
- i486SX — an i486DX with the FPU part disabled or missing. Early variants were parts with disabled (defective[citation needed]) FPUs, later versions had the FPU removed from the die to reduce area and hence cost.
- i486SX-S — SL Enhanced 486SX
- i486SXL — A 486SX with SMM (System Management Mode), stop clock, and power saving features.
- i486DX2 — the internal processor clock runs at twice the clock rate of the external bus clock.
- i486SX2 — i486DX2 with the FPU disabled.
- i486SL — low power version of the i486DX, reduced VCore, power conservation circuitry - mainly for use in portable computers.
- i486SL-NM — i486SL based on i486SX
- i487SX — i486DX with a slightly different pinout sold as an FPU upgrade to i486SX systems; it was widely documented that an i487SX when installed completely disabled the existing i486SX on the motherboard, replacing it.
- i486 OverDrive — i486SX, i486SX2, i486DX2 or i486DX4. Marked as upgrade processors, some models had different pinouts or voltage handling abilities from 'standard' chips of the same speed stepping. Fitted to a coprocessor or "OverDrive" socket on the motherboard, worked the same as the i487SX.
- i486DX4 — designed to run at triple clock rate (not quadruple as often believed; the DX3, which was meant to run at 2.5x the clock speed, was never released). DX4 models that featured write-back cache were identified by an "&EW" laser etched into their top surface, while the write-through models were identified by "&E".
| Model | Specified max clock | Voltage | L1-Cache | Introduced | |
|---|---|---|---|---|---|
 |
i486DX (P4) | 20,25,33 MHz; 50 MHz | 5V | 8 KB WT | April 1989; April 1989; May 1990; June 1991 |
 |
i486SL | 20,25,33 MHz | 5V or 3.3V | 8 KB WT | Nov 1992 |
| i486DXL | ? | ? | 8 KB WT | ? | |
 |
i486SX (P23) | 16,20,25 MHz (33 MHz) | 5V | 8 KB WT | September 1991 (September 1992) |
 |
i486DX2 (P24) | 40/20, 50/25 MHz (66/33 MHz) | 5V | 8 KB WT | March 1992 (August 1992) |
| i486DX-S (P4S) | 33 MHz; 50 MHz | 5V or 3.3V | 8 KB WT | June 1993 | |
| i486DX2-S (P24S) | 40/20, 50/25 MHz (66/33 MHz) | 5V or 3.3V | 8 KB WT | June 1993 | |
| i486SX-S (P23S) | 25,33 MHz | 5V or 3.3V | 8 KB WT | June 1993 | |
| i486SXL | ? | ? | 8 KB WT | ? | |
 |
i486SX2 | 50/25, 66/33 MHz | 5V | 8 KB WT | March 1994 |
| IntelDX4 (P24C) | 75/25, 100/33 MHz | 3.3V | 16 KB WT | March 1994 | |
 |
IntelDX4WB | 100/33 MHz | 3.3V | 16 KB WB | October 1994 |
| i486DX2WB (P24D) | 50/25, 66/33 MHz | 5V | 8 KB WB | October 1994 | |
| i486DX2 (P24LM) | 30/90 MHz; 33/100 MHz | 2.5-2.9V | 8 KB WT | 1994 |
WT = Write-Through cache strategy, WB = Write-Back cache strategy
The specified maximum internal clock frequency (on Intels versions) ranged from 16 to 100 MHz. The 16 MHz i486SX model was used by Dell Computers but sometimes ridiculed for the fact that it was handily beaten by many 386 systems. One of the few 486 models specified for a 50 MHz bus (486DX-50) initially had overheating problems and was moved to the 0.8 micrometre fabrication process. However, problems continued when installed in local bus systems due to the high bus speed, making it rather unpopular with mainstream consumers as local bus video was considered a requirement at the time. It was soon eclipsed by the clock-doubled i486DX2 which instead ran the CPU logic at twice the external bus speed. However, the 486DX-50 remained popular with users of EISA systems. More powerful 486 iterations such as the OverDrive and DX4 were less popular (the latter available as an OEM part only), as they came out after Intel had released the next generation Pentium. Certain steppings of the DX4 also officially supported 50 MHz bus operation but was a seldom used feature.
[edit] Gaming
The 486DX2 66 MHz was the most widespread high-end 486 version.[citation needed]? For many players of video games during the early and mid 1990s, towards the end of the MS-DOS gaming era, it was often coupled with 8 - 16 MB RAM and a VLB video card. This configuration was capable of running every title[citation needed] available for several years after its release, making it a "sweet spot" in CPU performance and longevity. The introduction of 3D computer graphics spelled the end of the 486's reign, because 3D graphics make heavy use of floating point calculations and the need for faster CPU cache and more memory bandwidth. Developers also began to target the Pentium almost exclusively with x86 assembly language optimizations (e.g. Quake); many of these games required the speed of the Pentium's double-pipelined architecture anyway, so even if the code had been optimized for the 486 instead, it still would not have given satisfactory performance there.
[edit] Competitive alternatives
486 compatible processors have been produced by other companies such as IBM, Texas Instruments, AMD, Cyrix, UMC, and SGS Thompson. Some are near duplicates in terms of specifications and performance, some are not. (IBM had a requirement that parts in its machines be available from multiple sources; this may explain the IBM versions.) The 486 was, however, covered by many of Intel's patents relating to the 80386 as well as some of its own. Intel and IBM have broad cross-licenses of these patents, and AMD was granted rights to the relevant patents in the 1995 settlement of a lawsuit between the companies.[3]
AMD produced several models of the 486 using a 40 MHz bus (486DX-40, 486DX/2-80 & 486DX/4-120) which had no equivalent available from Intel, as well as a 90 MHz part using a proprietary 30 MHz external clock sold only to OEMs[citation needed]This sentence was somewhat mangled, not sure if the present interpretation about the 30 MHz is correct.. The fastest running 486 CPU, the Am5x86, ran at 133 MHz was released by AMD in 1995. 150 MHz and 160 MHz parts were planned but never officially released.
[edit] Motherboards and Buses
Early 486 machines were equipped with several ISA slots (using an emulated PC/AT-bus) and sometimes one or two 8-bit-only slots (compatible with the PC/XT-bus). [4] Many motherboards enabled overclocking of these up from the default 6 or 8 MHz to perhaps 16.5 or 20 MHz (half the i486 bus clock) in a number of steps, often from within the BIOS setup. Especially older peripheral cards normally worked well at such speeds as they often used standard MSI chips instead of slower (at the time) custom VLSI designs. This could give significant performance gains (such as for old video cards moved from a 386 or 286 computer, for example). However, operation beyond 8 or 10 MHz could sometimes lead to stability problems, at least in systems equipped with SCSI and/or sound cards.
Some motherboards came equipped with a 32-bit bus called EISA that was backward compatible with the ISA-standard. EISA offered a number of attractive features such as increased bandwidth, extended addressing, IRQ sharing, and card configuration through software (rather than through jumpers, DIP switches, etc) However, EISA cards were expensive and therefore mostly employed in servers and workstations. Consumer desktops often used the simpler but faster VESA Local Bus, unfortunately somewhat prone to electrical and timing-based instability; typical consumer desktops had ISA slots combined with a single VLB slot for a video card. VLB was gradually replaced by PCI during the final years of the 80486 period. Few Pentium motherboards had VLB support as VLB was based directly on the i486 bus; it was no trivial matter adapting it to the quite different Pentium-bus. ISA persisted through the original Pentium generation and was not completely displaced by PCI until the Pentium II era.
Late 486 boards were normally equipped with both PCI- and ISA-slots, and sometimes a single VLB slot as well. In this configuration VLB or PCI thoughput sometimes suffered greatly depending on how the buses were bridged. The VLB slot in these systems was usually only fully compatible with video cards (quite fitting as "VESA" stands for Video Electronics Standards Association); VLB-IDE, multi I/O, or SCSI cards could have problems on motherboards with PCI slots. The VL-Bus operated at the same clock speed as the i486-bus (basically being a local 486-bus) while the PCI bus also usually depended on the i486 clock but sometimes had a divider setting available via the BIOS. This could be set to 1/1 or 1/2, sometimes even 2/3 (for 50 MHz CPU clocks). Some motherboards limited the PCI clock to the specified maximum of 33 MHz and certain network cards depended on this frequency for correct bit-rates. The ISA clock was typically generated by a divider of the CPU/VLB/PCI clock (as implied above).
One of the earliest complete systems to use the 80486 chip was the Apricot VX FT, produced by United Kingdom hardware manufacturer Apricot Computers. Even overseas in the United States it drew attention as "The World's First 486" in a popular September 1989 issue of Byte magazine (shown right).
Later 486 boards also supported Plug-And-Play, the Microsoft technology designed to make component installation easier for consumers that began as a part of Windows 95.
[edit] Obsolescence
Windows 95 signaled the end of the 486 era due to its high memory requirements (16 MB to perform as well as Windows 3.x with just 8 MB). Many 486 users at that time were running eight 1 MB 30-pin SIMMs on a motherboard with as many SIMM sockets, leaving no available room for expansion (without replacing existing memory with larger SIMMs, which was expensive and not always supported by the motherboard.) As 4-Meg 30-pin SIMMs were still very expensive at that time, it usually made more sense to buy a Pentium than to spend a premium on upgrading a system that was nearing the end of its service life. In the general purpose desktop computer role, the 486s were used as budget machines for people who could not afford the latest computers, until around 2001, when Windows 95 support ended and Windows 98, ME, NT 4.0, 2000, NT 4.0 and XP required more powerful computers. A small proportion of 486s stayed in service much longer in dedicated roles off the desktop as servers, network hosts, routers, terminal emulators, process control systems, etc., running various operating systems other than Microsoft Windows 98 and later. Some people also kept 486-based PCs for playing classic games under MS-DOS, as to those people having the additional old computer taking up some physical space alongside their newer system was worth the benefit of being able to run their favorite old games perfectly and not having to accept the limitations (to fidelity and authenticity) of running those games in an emulator or the hassles of getting each game to work with the emulator.
[edit] Notes and references
- ^ geocities.com
- ^ reghardware.co.uk
- ^ amd.com
- ^ In general, 8-bit ISA slots in these systems were implemented just by leaving off the shorter "C"/"D" connector of the slot, though the copper traces for a 16-bit slot were still there on the motherboard; the computer could tell no difference between an 8-bit ISA adapter in such a slot and the same adapter in a 16-bit slot, and there were still enough 8-bit adapters in circulation that vendors figured they could save money on a few connectors this way. IBM was the first to do this in the IBM AT.
This article was originally based on material from the Free On-line Dictionary of Computing, which is licensed under the GFDL.
[edit] See also
- List of Intel microprocessors
- Motorola 68040, although not compatible, often positioned as the Motorola equivalent to the Intel 80486 in terms of performance and features.
- "80486 System Architecture" published by Mindshare (pdf)
[edit] External links
|
||||||||||||||||||||||||||||||||||||