80386
RapidCAD
Introduction date: | October 1985 | |
Category: | 32-bit microprocessor | |
Frequency: | 16-33MHz | |
Technology: | CHMOS III | |
Number of transistors: | 275.000 (1.5 Microns) | |
Addressable memory: | 4Gbyte | |
Number of instructions: | 129 with 32 registers (eight 32-bit) | |
History: |
|
|
Second source: | AMD, Chips, Cyrix and Texas Instruments |
Intel | |||||
A80386-16 | A80386DX-16 | A80386-20 | A80386DX-20 IV | A80386-25 | A80386DX-25 |
SX217 |
A80386DX-25 IV | A80386-33 IV | ||||
NG80386SX-16 | KU80386SLB1A-20 | NG80386SX-20 | NG80386SX-25 | KU80386EX-25 | |
RapidCAD |
|||||
RAPIDCAD-1 | RAPIDCAD-1 | ||||
Evaluation Sample USA Q612 |
USA SZ624 |
||||
RAPIDCAD-2 | RAPIDCAD-2 | ||||
Evaluation Sample USA Q613 |
USA SZ625 |
The Intel RapidCAD is, in rough terms, an Intel 486DX with the 8 KB on-chip
cache removed and a 386 pinout. To software, it appears to be a 386DX with a
math coprocessor, as the 486-specific instructions have been removed from its
instruction set. It is marketed by Intel as "the ultimate coprocessor" and, as
the name implies, its main purpose is to replace the 386DX in existing systems
and thereby boost the performance of floating-point intensive applications such
as CAD software, spreadsheets, and math packages.
RapidCAD is delivered as a set of two chips. RapidCAD-1 is a 132-pin PGA (pin
grid array) chip that goes into the CPU socket and replaces the 386DX. It
contains the CPU and the FPU (floating-point unit). RapidCAD-2 is a 68-pin PGA
chip that goes into the 387 coprocessor (or EMC) socket; it contains a simple
PLA whose only purpose it to generate the FERR signal for the motherboard logic:
this provides full PC-compatible handling of floating-point exceptions. Many CPU
instructions execute in one clock cycle in the RapidCAD, just as on the Intel
486. The RapidCAD is constrained, however, by use of the standard 386DX bus
interface, since every bus cycle takes two CPU clock cycles. This means that
instructions may be executed by the RapidCAD faster than they can be fetched
from memory. But as most FPU instructions take longer to execute than CPU
instructions, they are not greatly slowed down by this slow bus interface, and
most of them execute in the same time as in the Intel 486DX. Therefore, the
Intel RapidCAD provides higher overall floating-point performance than any
386/387 combination. Results from the SPEC benchmarks, a common workstation UNIX
benchmark suite, show that the RapidCAD provides 85% more floating-point and 15%
more integer performance than an Intel 386DX/387DX combination at the same clock
frequency.