TMP320C50KGD, TMP320LC50KGD
DIGITAL SIGNAL PROCESSOR
KNOWN GOOD DIE
SGZS008C -- JULY 1996 -- REVISED JUNE 2000
D
25-ns, 35-ns, and 50-ns Single-Cycle
Instruction Execution Time for 5-V
Operation
D
D
D
D
D
D
D
D
16-Bit Parallel Logic Unit (PLU)
16 × 16-Bit Multiplier, 32-Bit Product
Eleven Context Switch Registers
D
D
D
50-ns Single-Cycle Instruction Execution
Time for 3.3-V Operation
Two Buffers for Circular Addressing
Full-Duplex Synchronous Serial Port
Time-Division Multiplexed Serial Port (TDM)
Timer With Control and Counter Registers
Source-Code Compatible With All ’C1x and
’C2x Devices
RAM-Based Operation
-- 9 K - W o r d × 16-Bit Single-Access On-Chip
Program/Data RAM
Sixteen Software-Programmable Wait-State
Generators
-- 1056-Word × 16-Bit Dual-Access On-Chip
Data RAM
D
D
D
D
Divide-By-1 Clock Option
IEEE Standard 1149.1† Test Access Port
D
D
2K-Word × 16-Bit On-Chip Boot ROM
Operations are Fully Static
224K-Word × 16-Bit Maximum Addressable
External Memory Space (64K-Word
Program, 64K-Word Data, 64K-Word I/O,
and 32K-Word Global)
Fabricated Using the Texas Instruments (TI)
Enhanced Performance Implanted CMOS
(EPIC™) 0.64-μm Technology
D
32-Bit Arithmetic Logic Unit (ALU)
-- 32-Bit Accumulator (ACC)
-- 32-Bit Accumulator Buffer (ACCB)
description
The TMP320C50KGD digital signal processor (DSP) is a high-performance, 16-bit, fixed-point processor
manufactured in 0.64-μm double-level metal CMOS technology. The TMP320LC50KGD has the same
functionality as the ’C50KGD except for operation at 3.3 V instead of 5 V.
Texas Instruments Military Products currently employs three primary processes for the development of a known
good die (KGD), one of which is applied to the TMP320C50 and TMP320LC50 devices. This process, known
as hot-chuck-probe, uses a standard probed product that is tested again, this time at full data sheet
specifications, in wafer form at speed and elevated temperature (85°C). Each individual die then is sawed,
inspected, and packed for shipment. This flow produces a bare die that has been temperature-tested at speed
and is known to be good, without having to use a temporary package.
A number of enhancements to the basic ’C2x architecture give the ’C5x a minimum 2x performance over the
previous generation. A four-deep instruction pipeline, incorporating delayed branching, delayed call to
subroutine, and delayed return from subroutine, allows the ’C5x to perform instructions in fewer cycles. The
addition of a PLU gives the ’C5x a method of manipulating bits in data memory without using the ACC and ALU.
The ’C5x has additional shifting and scaling capability for proper alignment of multiplicands or storage of values
to data memory.
Please be aware that an important notice concerning availability, standard warranty, and use in critical applications of
Texas Instruments semiconductor products and disclaimers thereto appears at the end of this data sheet.
†
IEEE Standard 1149.1--1990, IEEE Standard Test-Access Port and Boundary-Scan Architecture
EPIC is a trademark of Texas Instruments.
PRODUCTION DATA information is current as of publication date.
Products conform to specifications per the terms of Texas Instruments
standard warranty. Production processing does not necessarily include
testing of all parameters.
Copyright © 2000, Texas Instruments Incorporated
On products compliant to MIL-PRF-38535, all parameters are tested
unless otherwise noted. On all other products, production
processing does not necessarily include testing of all parameters.
1
POST OFFICE BOX 1443 • HOUSTON, TEXAS 77251--1443