HD64F38024DV Renesas Electronics America, HD64F38024DV Datasheet - Page 321

• Manufacturer Part Number: HD64F38024DV
• Description: IC H8/SLP MCU FLASH 80QFP
• Manufacturer: Renesas Electronics America
• Series: H8® H8/300L SLPr

Specifications of HD64F38024DV

• Core Processor: H8/300L
• Core Size: 8-Bit
• Speed: 10MHz
• Connectivity: SCI
• Peripherals: LCD, PWM, WDT
• Number Of I /o: 51
• Program Memory Size: 32KB (32K x 8)
• Program Memory Type: FLASH
• Ram Size: 1K x 8
• Voltage - Supply (vcc/vdd): 1.8 V ~ 5.5 V
• Data Converters: A/D 8x10b
• Oscillator Type: Internal
• Operating Temperature: -40°C ~ 85°C
• Package / Case: 80-QFP
• Lead Free Status / RoHS Status: Lead free / RoHS Compliant
• Eeprom Size: -

(4) Floating-Point Operation Specifications
This section describes the specifications for arithmetic operations on floating-point numbers in
C/C++, and for conversion between the decimal representation of floating-point numbers and
their internal representation during compilation and in C library processing.
(a) Specifications for arithmetic operations
i. Rounding of results
ii. Processing of overflows, underflows, and illegal operations
When the result of arithmetic operations on floating-point numbers exceeds the number
of valid limit in the mantissa in internal representation, the result is rounded according
to the following rules:
a. The result is rounded toward the closer of the two internal representations of the
b. When the result is exactly between the two approximating floating-point numbers, it
The following is performed in the event of an overflow, underflow, or illegal operation.
a. In the case of an overflow, the result is a positive or negative infinity, depending on
b. In the case of an underflow, the result is a positive or negative zero, depending on
c. In the case of an illegal operation, in which infinity values of the opposite sign have
d.
Note: Operations are performed on constant expressions during compilation. If an
approximating floating-point numbers.
is rounded to the floating-point number of which the last digit of the mantissa is 0.
the sign of the result.
the sign of the result.
been added, in which an infinity has been subtracted from another infinity of the
same sign, in which zero has been multiplied by infinity, in which zero is divided by
zero, or in which infinity is divided by infinity, the result is a not-a-number.
For the cases above, error numbers are set to variable errno which indicates an
error. For details on error numbers, refer to section 12.3, C Library Error
Messages. Whether an error has occurred can be checked by the errno value.
overflow, underflow, or illegal operation occurs, a warning level error message
is output.
307