|
sgdk
|
Mathematical methods. More...
Go to the source code of this file.
Classes | |
| struct | Vect2D_u16 |
| 2D Vector structure - u16 type. More... | |
| struct | Vect2D_s16 |
| 2D Vector structure - s16 type. More... | |
| struct | Vect2D_u32 |
| 2D Vector structure - u32 type. More... | |
| struct | Vect2D_s32 |
| 2D Vector structure - s32 type. More... | |
| struct | Vect2D_f16 |
| 2D Vector structure - f16 (fix16) type. More... | |
| struct | Vect2D_f32 |
| 2D Vector structure - f32 (fix32) type. More... | |
| struct | Mat2D_f16 |
| 2x2 Matrice structure - f16 (fix16) type. Internally uses 2 2D vectors. More... | |
| struct | Mat2D_f32 |
| 2x2 Matrice structure - f32 (fix32) type. Internally uses 2 2D vectors. More... | |
| struct | Vect3D_u16 |
| 3D Vector structure - u16 type. More... | |
| struct | Vect3D_s16 |
| 3D Vector structure - s16 type. More... | |
| struct | Vect3D_u32 |
| 3D Vector structure - u32 type. More... | |
| struct | Vect3D_s32 |
| 3D Vector structure - s32 type. More... | |
| struct | Vect3D_f16 |
| 3D Vector structure - f16 (fix16) type. More... | |
| struct | Vect3D_f32 |
| 3D Vector structure - f32 (fix32) type. More... | |
| struct | Mat3D_f16 |
| 3x3 Matrice structure - f16 (fix16) type. Internally uses 3 3D vectors. More... | |
| struct | Mat3D_f32 |
| 3x3 Matrice structure - f32 (fix32) type. Internally uses 3 3D vectors. More... | |
| struct | Vect4D_f16 |
| 4D Vector structure - f16 (fix16) type. More... | |
| struct | Vect4D_f32 |
| 4D Vector structure - f32 (fix32) type. More... | |
| struct | Mat4D_f16 |
| 4x4 Matrice structure - f16 (fix16) type. Internally uses 4 4D vectors. More... | |
| struct | Mat4D_f32 |
| 4x4 Matrice structure - f32 (fix32) type. Internally uses 4 4D vectors. More... | |
Defines | |
| #define | min(X, Y) (((X) < (Y))?(X):(Y)) |
| Returns the lowest value between X an Y. | |
| #define | max(X, Y) (((X) > (Y))?(X):(Y)) |
| Returns the highest value between X an Y. | |
| #define | clamp(X, L, H) (min(max((X), (L)), (H))) |
| Returns L if X is less than L, H if X is greater than H or X if in between L and H. | |
| #define | abs(X) (((X) < 0)?-(X):(X)) |
| Returns the absolute value of X. | |
| #define | PI 3.14159265358979323846 |
| PI number (3,1415..) | |
| #define | FIX16_INT_BITS 10 |
| #define | FIX16_FRAC_BITS (16 - FIX16_INT_BITS) |
| #define | FIX16_INT_MASK (((1 << FIX16_INT_BITS) - 1) << FIX16_FRAC_BITS) |
| #define | FIX16_FRAC_MASK ((1 << FIX16_FRAC_BITS) - 1) |
| #define | FIX32_INT_BITS 22 |
| #define | FIX32_FRAC_BITS (32 - FIX32_INT_BITS) |
| #define | FIX32_INT_MASK (((1 << FIX32_INT_BITS) - 1) << FIX32_FRAC_BITS) |
| #define | FIX32_FRAC_MASK ((1 << FIX32_FRAC_BITS) - 1) |
| #define | FASTFIX16_INT_BITS 8 |
| #define | FASTFIX16_FRAC_BITS (16 - FASTFIX16_INT_BITS) |
| #define | FASTFIX16_INT_MASK (((1 << FASTFIX16_INT_BITS) - 1) << FASTFIX16_FRAC_BITS) |
| #define | FASTFIX16_FRAC_MASK ((1 << FASTFIX16_FRAC_BITS) - 1) |
| #define | FASTFIX32_INT_BITS 16 |
| #define | FASTFIX32_FRAC_BITS (32 - FASTFIX32_INT_BITS) |
| #define | FASTFIX32_INT_MASK (((1 << FASTFIX32_INT_BITS) - 1) << FASTFIX32_FRAC_BITS) |
| #define | FASTFIX32_FRAC_MASK ((1 << FASTFIX32_FRAC_BITS) - 1) |
| #define | FIX16(value) ((fix16) ((value) * (1 << FIX16_FRAC_BITS))) |
| Convert specified value to fix16. | |
| #define | FIX32(value) ((fix32) ((value) * (1 << FIX32_FRAC_BITS))) |
| Convert specified value to fix32. | |
| #define | FASTFIX16(value) ((fastfix16) ((value) * (1 << FASTFIX16_FRAC_BITS))) |
| Convert specified value to "fast" fix16. | |
| #define | FASTFIX32(value) ((fastfix32) ((value) * (1 << FASTFIX32_FRAC_BITS))) |
| Convert specified value to "fast" fix32. | |
| #define | fix16Add(a, b) _Pragma("GCC error \"This method is deprecated, simply use '+' operator to add fix16 values together.\"") |
| #define | fix16Sub(a, b) _Pragma("GCC error \"This method is deprecated, simply use '-' operator to subtract fix16 values.\"") |
| #define | fix16Neg(a) _Pragma("GCC error \"This method is deprecated, simply use '0 - value' to get the negative fix16 value.\"") |
| #define | fix32Add(a, b) _Pragma("GCC error \"This method is deprecated, simply use '+' operator to add fix32 values together.\"") |
| #define | fix32Sub(a, b) _Pragma("GCC error \"This method is deprecated, simply use '-' operator to subtract fix32 values.\"") |
| #define | fix32Neg(a) _Pragma("GCC error \"This method is deprecated, simply use '0 - value' to get the negative fix32 value.\"") |
| #define | distance_approx(dx, dy) _Pragma("GCC error \"This method is deprecated, use getApproximatedDistance(..) instead.\"") |
Typedefs | |
| typedef Vect2D_u16 | V2u16 |
| alias for Vect2D_u16 | |
| typedef Vect2D_s16 | V2s16 |
| alias for Vect2D_s16 | |
| typedef Vect2D_u32 | V2u32 |
| alias for Vect2D_u32 | |
| typedef Vect2D_s32 | V2s32 |
| alias for Vect2D_s32 | |
| typedef Vect2D_f16 | V2f16 |
| alias for Vect2D_f16 | |
| typedef Vect2D_f32 | V2f32 |
| alias for Vect2D_f32 | |
| typedef Vect3D_u16 | V3u16 |
| alias for Vect3D_u16 | |
| typedef Vect3D_s16 | V3s16 |
| alias for Vect3D_s16 | |
| typedef Vect3D_u32 | V3u32 |
| alias for Vect3D_u32 | |
| typedef Vect3D_s32 | V3s32 |
| alias for Vect3D_s32 | |
| typedef Vect3D_f16 | V3f16 |
| alias for Vect3D_f16 | |
| typedef Vect3D_f32 | V3f32 |
| alias for Vect3D_f32 | |
| typedef Vect4D_f16 | V4f16 |
| alias for Vect4D_f16 | |
| typedef Vect4D_f32 | V4f32 |
| alias for Vect4D_f32 | |
| typedef Mat2D_f16 | M2f16 |
| alias for Mat2D_f16 | |
| typedef Mat2D_f32 | M2f32 |
| alias for Mat2D_f32 | |
| typedef Mat3D_f16 | M3f16 |
| alias for Mat3D_f16 | |
| typedef Mat3D_f32 | M3f32 |
| alias for Mat3D_f32 | |
| typedef Mat4D_f16 | M4f16 |
| alias for Mat4D_f16 | |
| typedef Mat4D_f32 | M4f32 |
| alias for Mat4D_f32 | |
Functions | |
| u32 | mulu (u16 op1, u16 op2) |
| 16x16=32 unsigned multiplication. Force GCC to use proper 68000 mulu instruction. | |
| s32 | muls (s16 op1, s16 op2) |
| 16x16=32 signed multiplication. Force GCC to use proper 68000 muls instruction. | |
| u16 | divu (u32 op1, u16 op2) |
| Direct divu instruction (unsigned 32/16=16:16) access using inline assembly to process op1/op2 operation. | |
| s16 | divs (s32 op1, s16 op2) |
| Direct divs instruction (signed 32/16=16:16) access using inline assembly to process op1/op2 operation. | |
| u16 | modu (u32 op1, u16 op2) |
| Direct divu instruction (unsigned 32/16=16:16) access using inline assembly. | |
| s16 | mods (s32 op1, s16 op2) |
| Direct divs instruction (signed 32/16=16:16) access using inline assembly. | |
| u32 | divmodu (u32 op1, u16 op2) |
| Direct divu instruction (unsigned 32/16=16:16) access using inline assembly to process op1/op2 operation and op1op2 at same time. | |
| s32 | divmods (s32 op1, s16 op2) |
| Direct divs instruction (signed 32/16=16:16) access using inline assembly to process op1/op2 operation and op1op2 at same time. | |
| fix16 | intToFix16 (s16 value) |
| Convert integer to fix16. | |
| s16 | fix16ToInt (fix16 value) |
| Convert fix16 to integer. | |
| fix32 | fix16ToFix32 (fix16 value) |
| Convert specified fix16 value to fix32. | |
| fix16 | fix16Round (fix16 value) |
| Round the specified value to nearest integer (fix16). | |
| s16 | fix16ToRoundedInt (fix16 value) |
| Round and convert the specified fix16 value to integer. | |
| fix16 | fix16Frac (fix16 value) |
| Return fractional part of the specified value (fix16). | |
| fix16 | fix16Int (fix16 value) |
| Return integer part of the specified value (fix16). | |
| fix16 | fix16Mul (fix16 val1, fix16 val2) |
| Compute and return the result of the multiplication of val1 and val2 (fix16). | |
| fix16 | fix16Div (fix16 val1, fix16 val2) |
| Compute and return the result of the division of val1 by val2 (fix16). | |
| fix16 | fix16Avg (fix16 val1, fix16 val2) |
| Compute and return the result of the average of val1 by val2 (fix16). | |
| fix16 | fix16Log2 (fix16 value) |
| Compute and return the result of the Log2 of specified value (fix16). | |
| fix16 | fix16Log10 (fix16 value) |
| Compute and return the result of the Log10 of specified value (fix16). | |
| fix16 | fix16Sqrt (fix16 value) |
| Compute and return the result of the root square of specified value (fix16). | |
| fix16 | sinFix16 (u16 value) |
| Compute sinus of specified value and return it as fix16. The input value is an integer defined as [0..1024] range corresponding to radian [0..2PI] range. | |
| fix16 | cosFix16 (u16 value) |
| Compute cosinus of specified value and return it as fix16. The input value is an integer defined as [0..1024] range corresponding to radian [0..2PI] range. | |
| fix32 | intToFix32 (s32 value) |
| Convert integer to fix32. | |
| s32 | fix32ToInt (fix32 value) |
| Convert fix32 to integer. | |
| fix16 | fix32ToFix16 (fix32 value) |
| Convert specified fix32 value to fix16. | |
| fix32 | fix32Round (fix32 value) |
| Round the specified value to nearest integer (fix32). | |
| s32 | fix32ToRoundedInt (fix32 value) |
| Round and convert the specified fix32 value to integer. | |
| fix32 | fix32Frac (fix32 value) |
| Return fractional part of the specified value (fix32). | |
| fix32 | fix32Int (fix32 value) |
| Return integer part of the specified value (fix32). | |
| fix32 | fix32Mul (fix32 val1, fix32 val2) |
| Compute and return the result of the multiplication of val1 and val2 (fix32). WARNING: result can easily overflow so its recommended to stick with fix16 type for mul and div operations. | |
| fix32 | fix32Div (fix32 val1, fix32 val2) |
| Compute and return the result of the division of val1 by val2 (fix32). WARNING: result can easily overflow so its recommended to stick with fix16 type for mul and div operations. | |
| fix32 | fix32Avg (fix32 val1, fix32 val2) |
| Compute and return the result of the average of val1 by val2 (fix32). | |
| fix32 | sinFix32 (u16 value) |
| Compute sinus of specified value and return it as fix32. The input value is an integer defined as [0..1024] range corresponding to radian [0..2PI] range. | |
| fix32 | cosFix32 (u16 value) |
| Compute cosinus of specified value and return it as fix32. The input value is an integer defined as [0..1024] range corresponding to radian [0..2PI] range. | |
| fastfix16 | intToFastFix16 (s16 value) |
| Convert integer to fastfix16. | |
| s16 | fastFix16ToInt (fastfix16 value) |
| Convert fastfix16 to integer. | |
| fastfix16 | fastFix16Round (fastfix16 value) |
| Round the specified value to nearest integer (fastfix16). | |
| s16 | fastFix16ToRoundedInt (fastfix16 value) |
| Round and convert the specified fastfix16 value to integer (fastfix16). | |
| fastfix16 | fastFix16Frac (fastfix16 value) |
| Return fractional part of the specified value (fastfix16). | |
| fastfix16 | fastFix16Int (fastfix16 value) |
| Return integer part of the specified value (fastfix16). | |
| fastfix16 | fastFix16Mul (fastfix16 val1, fastfix16 val2) |
| Compute and return the result of the multiplication of val1 and val2 (fastfix16). | |
| fastfix16 | fastFix16Div (fastfix16 val1, fastfix16 val2) |
| Compute and return the result of the division of val1 by val2 (fastfix16). | |
| fastfix32 | intToFastFix32 (s16 value) |
| Convert integer to fastfix32. | |
| s16 | fastFix32ToInt (fastfix32 value) |
| Convert fastfix32 to integer. | |
| fastfix32 | fastFix32Round (fastfix32 value) |
| Round the specified value to nearest integer (fastfix32). | |
| s32 | fastFix32ToRoundedInt (fastfix32 value) |
| Round and convert the specified fastfix32 value to integer. | |
| fastfix32 | fastFix32Frac (fastfix32 value) |
| Return fractional part of the specified value (fastfix32). | |
| fastfix32 | fastFix32Int (fastfix32 value) |
| Return integer part of the specified value (fastfix32). | |
| fastfix32 | fastFix32Mul (fastfix32 val1, fastfix32 val2) |
| Compute and return the result of the multiplication of val1 and val2 (fastfix32). WARNING: result can easily overflow so its recommended to stick with fix16 type for mul and div operations. | |
| fastfix32 | fastFix32Div (fastfix32 val1, fastfix32 val2) |
| Compute and return the result of the division of val1 by val2 (fastfix32). WARNING: result can easily overflow so its recommended to stick with fix16 type for mul and div operations. | |
| u32 | intToBCD (u32 value) |
| Binary to Decimal conversion. | |
| u32 | getApproximatedDistance (s32 dx, s32 dy) |
| Return euclidean distance approximation for specified vector. The returned distance is not 100% perfect but calculation is fast. | |
| u32 | getApproximatedDistanceV (V2s32 *v) |
| Return euclidean distance approximation for specified vector. The returned distance is not 100% perfect but calculation is fast. | |
| s32 | getApproximatedLog2 (s32 value) |
| Return 16.16 fixed point *approximation* of log2 of the specified 16.16 fixed point value. Ex: getLog2(1 << 16) = 0 getLog2(12345 << 16) = ~9.5 (real value = ~13.6) | |
| u16 | getLog2Int (u32 value) |
| Return integer log2 of specified 32 bits unsigned value. Ex: getLog2Int(1024) = 10 getLog2Int(12345) = 13 | |
| u32 | getNextPow2 (u32 value) |
| Return next pow2 value which is greater than specified 32 bits unsigned value. Ex: getNextPow2(700) = 1024 getNextPow2(18) = 32 | |
Variables | |
| const fix32 | sintab_f32 [1024] |
| const fix16 | sintab_f16 [1024] |
| const fix16 | log2tab_f16 [0x10000] |
| const fix16 | log10tab_f16 [0x10000] |
| const fix16 | sqrttab_f16 [0x10000] |
Mathematical methods.
This unit provides basic maths methods.
You can find a tutorial about how use maths with SGDK here.
| #define distance_approx | ( | dx, | |
| dy | |||
| ) | _Pragma("GCC error \"This method is deprecated, use getApproximatedDistance(..) instead.\"") |
| #define FASTFIX16 | ( | value | ) | ((fastfix16) ((value) * (1 << FASTFIX16_FRAC_BITS))) |
Convert specified value to "fast" fix16.
EX:
ff16 v = FASTFIX16(-27.12);
| #define FASTFIX32 | ( | value | ) | ((fastfix32) ((value) * (1 << FASTFIX32_FRAC_BITS))) |
Convert specified value to "fast" fix32.
EX:
ff32 v = FASTFIX32(34.567);
| #define FIX16 | ( | value | ) | ((fix16) ((value) * (1 << FIX16_FRAC_BITS))) |
Convert specified value to fix16.
EX:
f16 v = FIX16(-27.12);
| #define FIX32 | ( | value | ) | ((fix32) ((value) * (1 << FIX32_FRAC_BITS))) |
Convert specified value to fix32.
EX:
f32 v = FIX32(34.567);
Direct divs instruction (signed 32/16=16:16) access using inline assembly to process op1/op2 operation and op1op2 at same time.
| op1 | first operand - dividende (32 bit) |
| op2 | second operand - divisor (16 bit) |
Direct divu instruction (unsigned 32/16=16:16) access using inline assembly to process op1/op2 operation and op1op2 at same time.
| op1 | first operand - dividende (32 bit) |
| op2 | second operand - divisor (16 bit) |
Direct divs instruction (signed 32/16=16:16) access using inline assembly to process op1/op2 operation.
| op1 | first operand (32 bit) |
| op2 | second operand (16 bit) |
Direct divu instruction (unsigned 32/16=16:16) access using inline assembly to process op1/op2 operation.
| op1 | first operand - dividende (32 bit) |
| op2 | second operand - divisor (16 bit) |
Return euclidean distance approximation for specified vector.
The returned distance is not 100% perfect but calculation is fast.
| dx | delta X. |
| dy | delta Y. |
Return euclidean distance approximation for specified vector.
The returned distance is not 100% perfect but calculation is fast.
| v | 2D vector. |
Return 16.16 fixed point *approximation* of log2 of the specified 16.16 fixed point value. Ex:
getLog2(1 << 16) = 0
getLog2(12345 << 16) = ~9.5 (real value = ~13.6)
| value | 16.16 fixed point value to return log2 of |
Return integer log2 of specified 32 bits unsigned value. Ex:
getLog2Int(1024) = 10
getLog2Int(12345) = 13
| value | value to return log2 of |
Return next pow2 value which is greater than specified 32 bits unsigned value. Ex:
getNextPow2(700) = 1024
getNextPow2(18) = 32
| value | value to return next pow2 |
Direct divs instruction (signed 32/16=16:16) access using inline assembly.
| op1 | first operand (32 bit) |
| op2 | second operand (16 bit) |
Direct divu instruction (unsigned 32/16=16:16) access using inline assembly.
| op1 | first operand (32 bit) |
| op2 | second operand (16 bit) |
16x16=32 signed multiplication. Force GCC to use proper 68000 muls instruction.
| op1 | first operand |
| op2 | second operand |
1.7.4 
