欢迎访问ic37.com |
会员登录 免费注册
发布采购

P1750AES-20QLMB 参数 Datasheet PDF下载

P1750AES-20QLMB图片预览
型号: P1750AES-20QLMB
PDF下载: 下载PDF文件 查看货源
内容描述: [Microprocessor, 16-Bit, 20MHz, CMOS, QFP-68]
分类和应用: 时钟外围集成电路
文件页数/大小: 22 页 / 220 K
品牌: PYRAMID [ PYRAMID SEMICONDUCTOR CORPORATION ]
 浏览型号P1750AES-20QLMB的Datasheet PDF文件第1页浏览型号P1750AES-20QLMB的Datasheet PDF文件第2页浏览型号P1750AES-20QLMB的Datasheet PDF文件第4页浏览型号P1750AES-20QLMB的Datasheet PDF文件第5页浏览型号P1750AES-20QLMB的Datasheet PDF文件第6页浏览型号P1750AES-20QLMB的Datasheet PDF文件第7页浏览型号P1750AES-20QLMB的Datasheet PDF文件第8页浏览型号P1750AES-20QLMB的Datasheet PDF文件第9页  
P1750AE/SOS  
DIFFERENCES BETWEEN THE P1750A AND P1750AE  
The P1750AE achieves a 40% boost in performance (in clock cycles) over the P1750A. This reduction in clocks per  
instruction is because of three architectural enhancements:  
1) The inclusion of a 24 x 24 Multiply Accumulate (MAC) array.  
2) A reduction in non-bus cycles to 2 clocks (bus cycles remain at 4 clocks to maintain full compatibility with the  
CPU'speripheralchips).  
3) Branch calculation logic.  
Table 1. P1750A vs. P1750AE  
# of "Clocks" Required to Execute Selected Instructions  
P1750A  
# of Clocks  
23  
P1750AE  
# of Clocks  
4
Instruction  
Integer Multiply  
Throughput Increase(1)  
5.75  
Compare Between Limits  
Flt. Point Add/Subtract  
Flt. Point Multiply  
Flt. Point Compare  
Convert Flt. Point To Integer  
Shift Logical Left/Right  
Exchange  
24  
28  
43  
6
20  
18  
9
1.2  
1.56  
4.78  
1.5  
4
22  
9
16  
6
1.38  
1.5  
6
4
1.5  
Branch  
12  
8
1.5  
Note:  
1. Number of P1750A Clocks divided by number of P1750AE Clocks.  
Table 2. P1750AE BUILT-IN FUNCTIONS  
A core set of additional instructions has been included in the PACE1750AE. These instructions utilize the Built-In  
Function (BIF) opcode space. The objective of these new opcodes is to enhance the performance of the 1750AE in  
critical application areas such as navigation, DSP, transcendentals, and other LINPAK and matrix instructions.  
Below is a list of the BIFs and their execution times (N = the number of elements in the vector being processed).  
Address  
Mode  
Number of  
Instruction  
Mnemonic  
Clocks  
10+8N  
Notes  
Memory Parametric Dot Product - Single  
VDPS  
4F3(RA)  
Interruptable  
10+16N  
Memory Parametric Dot Product - Double  
3 x 3 Register Dot Product  
Double Precision Multiply Accumulate  
Polynomial  
VDPD  
R3DP  
MACD  
POLY  
CLAC  
STA  
4F1(RA)  
4F03  
4F02  
4F06  
4F00  
4F08  
4F04  
4F05  
4F07  
4F0F  
4F0D  
4F0E  
Interruptable  
6
8
7N-2  
4
Clear Accumulator  
Store Accumulator (32-Bit)  
Store Accumulator (48-Bit)  
Load Accumulator (32-Bit)  
Load Accumulator Long (48-Bit)  
Move MMU Page Block  
7
STAL  
LAC  
11  
9
LACL  
MMPG  
LTAR  
LTBR  
9
16+8N  
Privileged  
Load Timer A Reset Register  
Load Timer B Reset Register  
4
4
Do c um e nt # MICRO-7 REV B  
Pa g e 3 o f 22