5秒后页面跳转
ADSP21020 PDF预览

ADSP21020

更新时间: 2024-01-07 11:23:16
品牌 Logo 应用领域
亚德诺 - ADI 微处理器
页数 文件大小 规格书
32页 422K
描述
32/40-Bit IEEE Floating-Point DSP Microprocessor

ADSP21020 数据手册

 浏览型号ADSP21020的Datasheet PDF文件第2页浏览型号ADSP21020的Datasheet PDF文件第3页浏览型号ADSP21020的Datasheet PDF文件第4页浏览型号ADSP21020的Datasheet PDF文件第5页浏览型号ADSP21020的Datasheet PDF文件第6页浏览型号ADSP21020的Datasheet PDF文件第7页 
32/40-Bit IEEE Floating-Point  
DSP Microprocessor  
a
ADSP-21020  
FUNCTIONAL BLOCK DIAGRAM  
FEATURES  
Superscalar IEEE Floating-Point Processor  
Off-Chip Harvard Architecture Maximizes Signal  
Processing Performance  
30 ns, 33.3 MIPS Instruction Rate, Single-Cycle  
Execution  
INSTRUCTION  
DATA ADDRESS  
CACHE  
JTAG TEST  
GENERATORS  
& EMULATION  
PROGRAM  
DAG 1  
DAG 2  
SEQUENCER  
100 MFLOPS Peak, 66 MFLOPS Sustained Performance  
1024-Point Complex FFT Benchmark: 0.58 ms  
Divide (y/x): 180 ns  
Inverse Square Root (1/x): 270 ns  
32-Bit Single-Precision and 40-Bit Extended-Precision  
IEEE Floating-Point Data Formats  
32-Bit Fixed-Point Formats, Integer and Fractional,  
with 80-Bit Accumulators  
PROGRAM MEMORY ADDRESS  
DATA MEMORY ADDRESS  
PROGRAM MEMORY DATA  
DATA MEMORY DATA  
EXTERNAL  
ADDRESS  
BUSES  
EXTERNAL  
DATA  
BUSES  
IEEE Exception Handling with Interrupt on Exception  
Three Independent Computation Units: Multiplier,  
ALU, and Barrel Shifter  
REGISTER FILE  
TIMER  
Dual Data Address Generators with Indirect, Immedi-  
ate, Modulo, and Bit Reverse Addressing Modes  
Two Off-Chip Memory Transfers in Parallel with  
Instruction Fetch and Single-Cycle Multiply & ALU  
Operations  
Multiply with Add & Subtract for FFT Butterfly  
Computation  
Efficient Program Sequencing with Zero-Overhead  
Looping: Single-Cycle Loop Setup  
Single-Cycle Register File Context Switch  
15 (or 25) ns External RAM Access Time for Zero-Wait-  
State, 30 (or 40) ns Instruction Execution  
IEEE JTAG Standard 1149.1 Test Access Port and  
On-Chip Emulation Circuitry  
ARITHMETIC UNITS  
MULTIPLIER  
SHIFTER  
ALU  
multiplier operations. These computation units support IEEE  
32-bit single-precision floating-point, extended precision  
40-bit floating-point, and 32-bit fixed-point data formats.  
Data Register File  
A general-purpose data register file is used for transferring  
data between the computation units and the data buses, and  
for storing intermediate results. This 10-port (16-register)  
register file, combined with the ADSP-21020’s Harvard  
architecture, allows unconstrained data flow between  
computation units and off-chip memory.  
223-Pin PGA Package (Ceramic)  
Single-Cycle Fetch of Instruction and Two Operands  
The ADSP-21020 uses a modified Harvard architecture in  
which data memory stores data and program memory stores  
both instructions and data. Because of its separate program  
and data memory buses and on-chip instruction cache, the  
processor can simultaneously fetch an operand from data  
memory, an operand from program memory, and an  
instruction from the cache, all in a single cycle.  
GENERAL DESCRIPTION  
The ADSP-21020 is the first member of Analog Devices’ family  
of single-chip IEEE floating-point processors optimized for  
digital signal processing applications. Its architecture is similar  
to that of Analog Devices’ ADSP-2100 family of fixed-point  
DSP processors.  
Fabricated in a high-speed, low-power CMOS process, the  
ADSP-21020 has a 30 ns instruction cycle time. With a high-  
performance on-chip instruction cache, the ADSP-21020 can  
execute every instruction in a single cycle.  
Memory Interface  
Addressing of external memory devices by the ADSP-21020 is  
facilitated by on-chip decoding of high-order address lines to  
generate memory bank select signals. Separate control lines  
are also generated for simplified addressing of page-mode  
DRAM.  
The ADSP-21020 features:  
Independent Parallel Computation Units  
The arithmetic/logic unit (ALU), multiplier and shifter  
perform single-cycle instructions. The units are architecturally  
arranged in parallel, maximizing computational throughput. A  
single multifunction instruction executes parallel ALU and  
The ADSP-21020 provides programmable memory wait  
states, and external memory acknowledge controls allow  
interfacing to peripheral devices with variable access times.  
REV. C  
Information furnished by Analog Devices is believed to be accurate and  
reliable. However, no responsibility is assumed by Analog Devices for its  
use, nor for any infringements of patents or other rights of third parties  
which may result from its use. No license is granted by implication or  
otherwise under any patent or patent rights of Analog Devices.  
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.  
Tel: 617/329-4700 Fax: 617/326-8703  

与ADSP21020相关器件

型号 品牌 获取价格 描述 数据表
ADSP-21020 ADI

获取价格

32/40-Bit IEEE Floating-Point DSP Microprocessor
ADSP-21020BG-100 ADI

获取价格

32/40-Bit IEEE Floating-Point DSP Microprocessor
ADSP-21020BG-120 ADI

获取价格

32/40-Bit IEEE Floating-Point DSP Microprocessor
ADSP-21020BG-80 ADI

获取价格

32/40-Bit IEEE Floating-Point DSP Microprocessor
ADSP-21020KG-100 ADI

获取价格

32/40-Bit IEEE Floating-Point DSP Microprocessor
ADSP-21020KG-133 ADI

获取价格

32/40-Bit IEEE Floating-Point DSP Microprocessor
ADSP21020KG60 ADI

获取价格

ADSP21020KG60
ADSP-21020KG-60 ETC

获取价格

32-Bit Digital Signal Processor
ADSP-21020KG-80 ADI

获取价格

32/40-Bit IEEE Floating-Point DSP Microprocessor
ADSP21020KU60 ADI

获取价格

ADSP21020KU60