欢迎访问ic37.com |
会员登录 免费注册
发布采购

EP1SGX40GF1020I6 参数 Datasheet PDF下载

EP1SGX40GF1020I6图片预览
型号: EP1SGX40GF1020I6
PDF下载: 下载PDF文件 查看货源
内容描述: [Field Programmable Gate Array, 4697 CLBs, 41250-Cell, CMOS, PBGA1020, 33 X 33 MM, 1 MM PITCH, FBGA-1020]
分类和应用: 可编程逻辑
文件页数/大小: 272 页 / 1348 K
品牌: ALTERA [ ALTERA CORPORATION ]
 浏览型号EP1SGX40GF1020I6的Datasheet PDF文件第117页浏览型号EP1SGX40GF1020I6的Datasheet PDF文件第118页浏览型号EP1SGX40GF1020I6的Datasheet PDF文件第119页浏览型号EP1SGX40GF1020I6的Datasheet PDF文件第120页浏览型号EP1SGX40GF1020I6的Datasheet PDF文件第122页浏览型号EP1SGX40GF1020I6的Datasheet PDF文件第123页浏览型号EP1SGX40GF1020I6的Datasheet PDF文件第124页浏览型号EP1SGX40GF1020I6的Datasheet PDF文件第125页  
Stratix GX Architecture  
Table 4–13 shows the summary of input register modes for the DSP block.  
Table 4–13. Input Register Modes  
Register Input Mode  
Parallel input  
9 × 9  
v
v
18 × 18  
v
v
36 × 36  
v
Shift register input  
Multiplier  
The multiplier supports 9 × 9-, 18 × 18-, or 36 × 36-bit multiplication. Each  
DSP block supports eight possible 9 × 9-bit or smaller multipliers. There  
are four multiplier blocks available for multipliers larger than 9 × 9 bits  
but smaller than 18 × 18 bits. There is one multiplier block available for  
multipliers larger than 18 × 18 bits but smaller than or equal to 36 × 36 bits.  
The ability to have several small multipliers is useful in applications such  
as video processing. Large multipliers greater than 18 × 18 bits are useful  
for applications such as the mantissa multiplication of a single-precision  
floating-point number.  
The multiplier operands can be signed or unsigned numbers, where the  
result is signed if either input is signed as shown in Table 4–14. The  
sign_aand sign_bsignals provide dynamic control of each operand’s  
representation: a logic 1 indicates the operand is a signed number, a logic  
0 indicates the operand is an unsigned number. These sign signals affect  
all multipliers and adders within a single DSP block and you can register  
them to match the data path pipeline. The multipliers are full precision  
(that is, 18 bits for the 18-bit multiply, 36-bits for the 36-bit multiply, and  
so on), regardless of whether sign_aor sign_bset the operands as  
signed or unsigned numbers.  
Table 4–14. Multiplier Signed Representation  
Data A  
Data B  
Result  
Unsigned  
Unsigned  
Signed  
Unsigned  
Signed  
Unsigned  
Signed  
Signed  
Signed  
Unsigned  
Signed  
Signed  
Altera Corporation  
February 2005  
4–55  
Stratix GX Device Handbook, Volume 1  
 
 
 复制成功!