| Contents | Index |
This table summarizes what's new in Version 3.3 (R2011a):
| New Features and Changes | Version Compatibility Considerations | Fixed Bugs and Known Problems |
|---|---|---|
| Yes Details below | No | Bug
Reports Includes fixes |
New features and changes introduced in this version are:
Two new CORDIC functions, cordicpol2cart and cordicrotate, have been added to the toolbox. cordicpol2cart converts polar coordinates to Cartesian coordinates. cordicrotate rotates a complex input. Each of these functions uses efficient CORDIC-based approximations to perform the conversion and rotation, respectively.
All CORDIC functions now allow the data type of theta to be either signed or unsigned. This change expands the usability of these functions.
The number of iterations (niters) input argument is now an optional input to all CORDIC functions. If you do not specify the number of iterations or if you specify a value that is too large for the word length, the maximum reasonable number of iterations is used automatically.
A new fiaccel function has been added to the toolbox. fiaccel generates a MEX function and accelerates fixed-point code. These associated functions have also been added: coder.ArrayType, coder.Constant, coder.EnumType, coder.FiType, coder.newtype, coder.PrimitiveType, coder.resize, coder.StructType, coder.Type, and coder.typeof.
The following new signedness shortcuts have been added to the numerictype object syntax.
[] = 'Signedness','Auto'
0 = 'Signedness','Unsigned'
1 = 'Signedness','Signed'
For example, numerictype([],24,20) is a numerictype object with auto signedness, a word length of 24, and a fraction length of 20.
A new DataTypeOverride property has been added to the numerictype object. The numerictype DataTypeOverride property controls how fipref data type settings are applied to fi objects. This property allows you to conveniently ignore a global fipref setting or to turn fixed-point settings on or off without having to change every fi object.
If you set the numerictype DataTypeOverride property to 'off', the fi object uses the numerictype data type settings and ignores the fipref settings.
If you set the numerictype DataTypeOverride property to Inherit, which is the default, the fi object uses the fipref DataTypeOverride setting. Valid settings are ForceOffScaledDoubles, TrueDoubles, and TrueSingles.
By default, this property is not visible until you create a numerictype object and then explicitly set the desired value. See example in the New DataTypeOverrideAppliesTo Property Added to fipref Object section.
When you set the fipref DataTypeOverride property to any value other than ForceOff, the new DataTypeOverrideAppliesTo displays. DataTypeOverrideAppliesTo controls the data types to which the fipref DataTypeOverride property applies. Valid values of DataTypeOverrideAppliesTo are AllNumericTypes, Fixed-point, and Floating-point.
By default, this property is not visible until you create a numerictype or fipref object and then explicitly set the desired values. For example,
T = numerictype P = fipref % Change DataTypeOverride from its default T.DataTypeOverride = 'Off'; % Set fipref DataTypeOverride and % DataTypeOverrideAppliesTo P.DataTypeOverride = 'TrueDoubles'; P.DataTypeOverrideAppliesTo = 'FixedPoint;
R2011a introduces the capability to improve the performance of the NumericTypeScope by reducing the frequency with which the display updates. You can now choose between this new enhanced performance mode and the old behavior. By default, the scope operates in the new enhanced performance mode.
![]() | Version 3.4 (R2011b) Fixed-Point Toolbox Software | Version 3.2 (R2010b) Fixed-Point Toolbox Software | ![]() |

Learn how to apply early verification to your development process through these technical resources.
How much time do you spend on testing to ensure implementation meets system-level requirements?
| © 1984-2012- The MathWorks, Inc. - Site Help - Patents - Trademarks - Privacy Policy - Preventing Piracy - RSS |