DNZ - Revision history

TArntsen: Make link (48-bit floating point)

2013-04-04T14:32:54Z

Make link (48-bit floating point)

← Older revision		Revision as of 14:32, 4 April 2013
Line 31:		Line 31:
	{{main\|32-bit floating point}}		{{main\|32-bit floating point}}

	The ND-100/110 can have a 48-bit floating point CPU (standard configuration) or a 32-bit floating point CPU (customer option).		The ND-100/110 can have a [[48-bit floating point]] CPU (standard configuration) or a 32-bit floating point CPU (customer option).
	For 32-bit floating point operations DNZ works as described above, except that the T register is not used or affected.		For 32-bit floating point operations DNZ works as described above, except that the T register is not used or affected.
	On 32-bit floating point hardware only scaling factors of -16<sub><small>10</small></sub> should be used.		On 32-bit floating point hardware only scaling factors of -16<sub><small>10</small></sub> should be used.

TArntsen: Register notation uniformity fix

2013-03-21T08:34:25Z

← Older revision		Revision as of 08:34, 21 March 2013
Line 4:		Line 4:
	\|Format=DNZ ''<scaling>''		\|Format=DNZ ''<scaling>''
	\|Code=152 000		\|Code=152 000
	\|Affected=(T), (A), (D)		\|Affected=T, A, D
	\|Architecture=[[ND-100]], [[ND-110 CPU\|ND-110]]		\|Architecture=[[ND-100]], [[ND-110 CPU\|ND-110]]
	}}		}}

TArntsen: Typo: floting->floating

2010-07-01T10:07:55Z

Typo: floting->floating

← Older revision		Revision as of 10:07, 1 July 2010
Line 31:		Line 31:
	{{main\|32-bit floating point}}		{{main\|32-bit floating point}}

	The ND-100/110 can have a 48-bit floating point CPU (standard configuration) or a 32-bit ~~floting~~ point CPU (customer option).		The ND-100/110 can have a 48-bit floating point CPU (standard configuration) or a 32-bit floating point CPU (customer option).
	For 32-bit floating point operations DNZ works as described above, except that the T register is not used or affected.		For 32-bit floating point operations DNZ works as described above, except that the T register is not used or affected.
	On 32-bit floating point hardware only scaling factors of -16<sub><small>10</small></sub> should be used.		On 32-bit floating point hardware only scaling factors of -16<sub><small>10</small></sub> should be used.

TArntsen: Page ref. number update

2010-06-29T11:33:30Z

Page ref. number update

← Older revision		Revision as of 11:33, 29 June 2010
Line 36:		Line 36:

	== References ==		== References ==
	*{{ND-doc\|06.014.02\|pages 120, ~~144~~, 146 and 240}}		*{{ND-doc\|06.014.02\|pages 120, 145, 146 and 240}}
	*{{ND-doc\|06.029.01\|~~page 70~~ and 72}}		*{{ND-doc\|06.029.01\|pages 71 and 72}}


	[[Category:ND-100 instructions]]		[[Category:ND-100 instructions]]

TArntsen: DNZ instruction

2010-06-29T11:31:18Z

DNZ instruction

New page

{{InstructionBox
|Mnemonic=DNZ
|Description=Denormalize (floating to integer)
|Format=DNZ ''<scaling>''
|Code=152 000
|Affected=(T), (A), (D)
|Architecture=[[ND-100]], [[ND-110 CPU|ND-110]]
}}

'''DNZ''' is an assembly instruction. It converts the number in the [[floating point accumulator]] to a single precision fixed point number in the [[A register]], using the ''scaling'' of the DNZ instruction as a scaling factor.

The ''scaling'' is given to the conversion of -128 to 127 (approximately 10-39 to 1039).
-1
When converting to integers, a scaling factor of -1610 will give a fixed point number with the same value as the integer part of the floating point number. A larger scaling factor will result in a higher fixed point number on CPUs with 48-bit floating point units. On 32-bit floating point units a scaling factor of -1610 should ''always'' be used. Other values will not cause a different result but will affect the overflow test.

The [[T register|T]] and [[D register|D registers]] (part of the 48-bit floating point accumulator) will be cleared after this instruction.

If the instruction causes underflow the A register will also be cleared.

If the resulting integer has an absolute value larger than 32767 overflow occurs and the error flag (Z) is set to one.

Internally the original floating point value will be truncated, and negative numbers will be converted to positive numbers
before conversion, the result converted back to a negative number.

== Example ==
*DNZ-20 (''code 152360'') Convert from floating point, scaling factor -1610.
T-A-D value of 3.14 will result in A = 3

==== DNZ on 32-bit floating point CPUs ====
{{main|32-bit floating point}}

The ND-100/110 can have a 48-bit floating point CPU (standard configuration) or a 32-bit floting point CPU (customer option).
For 32-bit floating point operations DNZ works as described above, except that the T register is not used or affected.
On 32-bit floating point hardware only scaling factors of -1610 should be used.

== References ==
*{{ND-doc|06.014.02|pages 120, 144, 146 and 240}}
*{{ND-doc|06.029.01|page 70 and 72}}

[[Category:ND-100 instructions]]