All trademarks mentioned in this document

are trademarks of their respective owners.

Copyright 1999-2007 DCD – Digital Core Design. All Rights Reserved.

http://www.DigitalCoreDesign.com

http://www.dcd.pl

Information about shifted-out bits are stored

for rounding process.

Control Coprocessor –

manages execution

of all instructions and internal operation re-

quired to execute particular function.

Interface –

makes interface between exter-

nal device and DFPAU-DP internal 32-bit

modules. It contains data, control and status

registers. It can be configured to work with 8-,

16- and 32-bit processors.

PERFORMANCE

The following table gives a survey about

the Core area and performance in the AL-

TERA® devices after Place & Route (all key

features have been included):

Device

Speed

grade

Logic Cells

CYCLONE

-6

3660

79 MHz

CYCLONE-II

-6

3630

71 MHz

STRATIX

-5

3660

84 MHz

STRATIX-II

-3

2800

110 MHz

Core performance in ALTERA® devices

DFPAU-DP floating point instructions per-

formance has been compared to standard C

library functions delivered with every com-

mercial C compiler. Each program was exe-

cuted in the same system environments.

Number of clock periods were measured be-

tween input data loading into work registers

and output result storing after operation.

The results are placed in table below. Im-

provement has been computed as number of:

(NIOS-II CLK) divided by (NIOS-II+DFPAU-

DP CLK), required to execute particular in-

struction.

IEEE-754 FP Instruction

Improvement

Addition

12.0

Subtraction

11.7

Multiplication

10.6

Division

15.0

Square Root

21.5

Sine

11.8

Cosine

10.3

Tangent

10.1

Arcs Tangent

14.7

Average speed improvement:

13.1

More details are available in core docu-

mentation.

The following table gives a survey about

the 32-bit NIOS-II+DFPAU-DP performance

compared to 32-bit NIOS-II.

Device

Improvement

NIOS-II

1.0

NIOS-II+DFPAU (arithmetic)

14.1

NIOS-II+DFPAU (trigonometric)

11.7

NIOS-II+DFPAU (overall)

13.1