Floating-Point Arithmetic - Intel Agilex User Manual

Variable precision dsp blocks
Hide thumbs Also See for Agilex:
Table of Contents

Advertisement

®
1. Intel
Agilex
Variable Precision DSP Blocks Overview
UG-20213 | 2019.04.02

1.2.2. Floating-Point Arithmetic

Table 3.
Supported Combinations of Operational Modes and Features
Variable-Precision
DSP Block Resource
1 variable precision
FP32 multiplication mode
DSP block
FP32 addition or subtraction
mode
FP32 multiplication with
addition or subtraction mode
FP32 multiplication with
accumulation mode
FP32 vector one mode
FP32 vector two mode
Sum of two FP16 multiplication
mode
Sum of two FP16 multiplication
with FP32 addition mode
Sum of two FP16 multiplication
with accumulation mode
FP16 vector one mode
FP16 vector two mode
FP16 vector three
4 Variable precision
Floating-point complex
DSP blocks
multiplication
Table 4.
Supported Combinations of Operational Modes and Dynamic Control Features
Variable-Precision DSP Block
Resource
1 variable precision DSP block
4 Variable precision DSP blocks
Send Feedback
Operation Mode
Operation Instance
Operation Mode
FP32 multiplication mode
FP32 adder or subtract mode
FP32 multiplier adder or subtract mode
FP32 multiplier accumulate mode
FP32 vector one mode
FP32 vector two mode
Sum of two FP16 multiplication mode
Sum of two FP16 multiplication with FP32
addition mode
Sum of two FP16 multiplication with
accumulation mode
FP16 vector one mode
FP16 vector two mode
FP16 vector three
Floating-point complex multiplication
Supported
Chainin Support
1
No
1
No
1
Yes
1
No
1
Yes
1
Yes
1
No
1
Yes
1
No
1
Yes
1
Yes
1
No
1
No
®
Intel
Agilex
Variable Precision DSP Blocks User Guide
Chainout Support
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Dynamic ACCUMULATE
No
No
No
Yes
No
No
No
No
Yes
No
No
Yes
No
7

Advertisement

Table of Contents
loading

Table of Contents