Freigeben über


Verwenden von D3DXMath

D3DXMath ist eine Mathematische Hilfsbibliothek für Direct3D-Anwendungen. D3DXMath ist ein langjähriges Unternehmen, ist in D3DX 9 und D3DX 10 enthalten und geht auch auf ältere Versionen von DirectX zurück.

Hinweis

Die D3DX-Hilfsbibliothek (D3DX 9, D3DX 10 und D3DX 11) ist für Windows 8 veraltet. Daher wird dringend empfohlen, dass Sie zu DirectXMath migrieren und nicht D3DXMath verwenden.

 

DirectXMath hat einen Großteil der gleichen Funktionen in D3DXMath, und intern enthält D3DXMath eine Reihe von prozessorspezifischen Optimierungen. Der Hauptunterschied besteht darin, dass D3DXMath im D3DX9* gehostet wird. DLLs und D3DX10*. DLLs und nur sehr wenige Funktionen sind inlined. Die Aufrufkonvention der DirectXMath-Bibliothek ist explizit SIMD-freundlich, während D3DXMath Lade- und Speicherkonvertierungen durchführen muss, um die SIMD-Optimierung zu implementieren.

Mischen von DirectXMath und D3DXMath

D3DX11 enthält nicht D3DXMath, und im Allgemeinen wird stattdessen die Verwendung von DirectXMath empfohlen. Sie können jedoch weiterhin mit D3DX9 und/oder D3DX10 in Ihrer Anwendung verknüpfen. Daher können Sie D3DXMath weiterhin verwenden oder D3DXMath und DirectXMath gleichzeitig in Ihrer Anwendung verwenden.

Im Allgemeinen ist es sicher, einen XMVECTOR* in eine Funktion zu umwandeln, die D3DXVECTOR4* verwendet, oder eine XMMATRIX* in eine Funktion zu umwandeln, die D3DXMATRIX* akzeptiert. Die Inverse ist jedoch nicht allgemein sicher, da XMVECTOR und XMMATRIX 16 Byte ausgerichtet sein müssen, während D3DXVECTOR4 und D3DXMATRIX keine solche Anforderung haben. Wenn diese Anforderung nicht eingehalten wird, kann dies zur Laufzeit zu ungültigen Ausrichtungsausnahmen führen.

Es ist sicher, einen XMVECTOR* in eine Funktion zu umwandeln, die D3DXVECTOR2* oder D3DXVECTOR3* akzeptiert, aber nicht umgekehrt. Sowohl Ausrichtungsprobleme als auch die Tatsache, dass D3DXVECTOR2 und D3DXVECTOR3 kleinere Strukturen sind, machen dies zu einem unsicheren Vorgang.

Hinweis

D3DX (und damit D3DXMath) gilt als Legacy und ist für Windows Store-Apps, die unter Windows 8 ausgeführt werden, nicht verfügbar und nicht im Windows 8 SDK für Desktop-Apps enthalten.

 

Verwenden von DirectXMath mit Direct3D

Sowohl DirectXMath als auch D3DXMath sind optional, wenn Sie mit Direct3D arbeiten. Direct3D 9 hat D3DMATRIX und D3DCOLOR als Teil der Direct3D-API zur Unterstützung der (jetzt legacy) Fixed-Function-Pipeline definiert. D3DXMath in D3DX9 erweitert diese Direct3D 9-Typen um gängige mathematische Grafikvorgänge. Für Direct3D 10.x und Direct3D 11 verwendet die API nur die programmierbare Pipeline, sodass es keine API-spezifische Struktur für Matrizen oder Farbwerte gibt. Wenn die neueren APIs einen Farbwert benötigen, verwenden sie ein explizites Array von Floatwerten oder einen generischen Puffer von Konstantendaten, der vom HLSL-Shader interpretiert wird. HLSL selbst kann entweder Zeilen-Haupt- oder Spalten-Hauptmatrixformate unterstützen. Daher liegt das Layout ganz bei Ihnen (weitere Informationen finden Sie unter HLSL, Matrixreihenfolge. Wenn Sie Spalten-Hauptmatrixformate in Ihren Shadern verwenden, müssen Sie die DirectXMath-Matrixdaten transponieren, wenn Sie sie in Ihre konstanten Pufferstrukturen platzieren). Die Bibliotheken DirectXMath und D3DXMath sind zwar optional, bieten jedoch beide allgemeine grafikbezogene Funktionen und sind daher bei der Direct3D-Programmierung äußerst praktisch.

Es ist sicher, einen XMVECTOR* in einen D3DVECTOR* oder XMMATRIX* in D3DMATRIX* zu umwandeln, da Direct3D 9 keine Ausrichtungsannahmen über die eingehende Datenstruktur trifft. Es ist auch sicher, XMCOLOR in D3DCOLOR zu umwandeln. Sie können eine 4-Float-Darstellung von Farbe über XMStoreColor() in XMCOLOR konvertieren, um das 8:8:8:8:8 32-Bit-DWORD abzurufen, das D3DCOLOR entspricht.

Wenn Sie mit Direct3D 10.x oder Direct3D 11 arbeiten, verwenden Sie in der Regel DirectXMath-Typen, um eine Struktur für jeden Ihrer Konstantenpuffer zu erstellen. In diesen Fällen hängt dies hauptsächlich von Ihrer Fähigkeit ab, die Ausrichtung zu steuern, um diese effizient zu gestalten, oder XMStore*()-Vorgänge zum Konvertieren von XMVECTOR- und XMMATRIX-Daten in die richtigen Datentypen. Beim Aufrufen von Direct3D 10.x- oder Direct3D 11-APIs, die ein float[4]-Array von Farbwerten erfordern, können Sie einen XMVECTOR* oder XMFLOAT4* mit den Farbdaten umwandeln.

Portieren von D3DXMath

D3DXMath-Typ DirectXMath-Entsprechung
D3DXFLOAT16 HÄLFTE
D3DXMATRIX XMFLOAT4X4
D3DXMATRIXA16 XMMATRIX oder XMFLOAT4X4A
D3DXQUATERNION
D3DXPLANE
D3DXCOLOR
XMVECTOR wird anstelle eindeutiger Typen verwendet. Daher müssen Sie wahrscheinlich einen XMFLOAT4-Hinweisverwenden:**D3DXQUATERNION::operator *** ruft die D3DXQuaternionMultiply-Funktion auf, die zwei Quaternionen multipliziert. Wenn Sie jedoch nicht explizit die XMQuaternionMultiply-Funktion verwenden, erhalten Sie eine falsche Antwort, wenn Sie **XMVECTOR::operator *** für ein Quaternion verwenden.
D3DXVECTOR2 XMFLOAT2
D3DXVECTOR2_16F XMHALF2
D3DXVECTOR3 XMFLOAT3
D3DXVECTOR4 XMFLOAT4 (oder, wenn Sie garantieren können, dass die Daten 16-Byte-ausgerichtet sind, XMVECTOR oder XMFLOAT4A )
D3DXVECTOR4_16F XMHALF4

 

Hinweis

Es gibt keine direkte Entsprechung zu D3DXVECTOR3_16F in XNAMath.

 

D3DXMath-Makro DirectXMath-Entsprechung
D3DX_PI XM_PI
D3DX_1BYPI XM_1DIVPI
D3DXToRadian XMConvertToRadians
D3DXToDegree XMConvertToDegrees

 

D3DXMath-Funktion DirectXMath-Entsprechung
D3DXBoxBoundProbe BoundingBox::Intersects(XMVECTOR, XMVECTOR, float&)
D3DXComputeBoundingBox BoundingBox::CreateFromPoints
D3DXComputeBoundingSphere BoundingSphere::CreateFromPoints
D3DXSphereBoundProbe BoundingSphere::Intersects(XMVECTOR, XMVECTOR, float&)
D3DXIntersectTriFunction TriangleTests::Intersects
D3DXFloat32To16Array XMConvertFloatToHalfStream
D3DXFloat16To32Array XMConvertHalfToFloatStream
D3DXVec2Length XMVector2Length oder XMVector2LengthEst
D3DXVec2LengthSq XMVector2LengthSq
D3DXVec2Dot XMVector2Dot
D3DXVec2CCW XMVector2Cross
D3DXVec2Add XMVectorAdd
D3DXVec2Subtract XMVectorSubtract
D3DXVec2Minimize XMVectorMin
D3DXVec2Maximize XMVectorMax
D3DXVec2Scale XMVectorScale
D3DXVec2Lerp XMVectorLerp oder XMVectorLerpV
D3DXVec2Normalize XMVector2Normalize oder XMVector2NormalizeEst
D3DXVec2Hermite XMVectorHermite oder XMVectorHermiteV
D3DXVec2CatmullRom XMVectorCatmullRom oder XMVectorCatmullRomV
D3DXVec2BaryCentric XMVectorBaryCentric oder XMVectorBaryCentricV
D3DXVec2Transform XMVector2Transform
D3DXVec2TransformCoord XMVector2TransformCoord
D3DXVec2TransformNormal XMVector2TransformNormal
D3DXVec2TransformArray XMVector2TransformStream
D3DXVec2TransformCoordArray XMVector2TransformCoordStream
D3DXVec2TransformNormalArray XMVector2TransformNormalStream
D3DXVec3Length XMVector3Length oder XMVector3LengthEst
D3DXVec3LengthSq XMVector3LengthSq
D3DXVec3Dot XMVector3Dot
D3DXVec3Cross XMVector3Cross
D3DXVec3Add XMVectorAdd
D3DXVec3Subtract XMVectorSubtract
D3DXVec3Minimize XMVectorMin
D3DXVec3Maximize XMVectorMax
D3DXVec3Scale XMVectorScale
D3DXVec3Lerp XMVectorLerp oder XMVectorLerpV
D3DXVec3Normalize XMVector3Normalize oder XMVector3NormalizeEst
D3DXVec3Hermite XMVectorHermite oder XMVectorHermiteV
D3DXVec3CatmullRom XMVectorCatmullRom oder XMVectorCatmullRomV
D3DXVec3BaryCentric XMVectorBaryCentric oder XMVectorBaryCentricV
D3DXVec3Transform XMVector3Transform
D3DXVec3TransformCoord XMVector3TransformCoord
D3DXVec3TransformNormal XMVector3TransformNormal
D3DXVec3TransformArray XMVector3TransformStream
D3DXVec3TransformCoordArray XMVector3TransformCoordStream
D3DXVec3TransformNormalArray XMVector3TransformNormalStream
D3DXVec3Project XMVector3Project
D3DXVec3Unproject XMVector3Unproject
D3DXVec3ProjectArray XMVector3ProjectStream
D3DXVec3UnprojectArray XMVector3UnprojectStream
D3DXVec4Length XMVector4Length oder XMVector4LengthEst
D3DXVec4LengthSq XMVector4LengthSq
D3DXVec4Dot XMVector4Dot
D3DXVec4Add XMVectorAdd
D3DXVec4Subtract XMVectorSubtract
D3DXVec4Minimize XMVectorMin
D3DXVec4Maximize XMVectorMax
D3DXVec4Scale XMVectorScale
D3DXVec4Lerp XMVectorLerp oder XMVectorLerpV
D3DXVec4Cross XMVector4Cross
D3DXVec4Normalize XMVector4Normalize oder XMVector4NormalizeEst
D3DXVec4Hermite XMVectorHermite oder XMVectorHermiteV
D3DXVec4CatmullRom XMVectorCatmullRom oder XMVectorCatmullRomV
D3DXVec4BaryCentric XMVectorBaryCentric oder XMVectorBaryCentricV
D3DXVec4Transform XMVector4Transform
D3DXVec4TransformArray XMVector4TransformStream
D3DXMatrixIdentity XMMatrixIdentity
D3DXMatrixDeterminant XMMatrixDeterminant
D3DXMatrixDecompose XMMatrixDecompose
D3DXMatrixTranspose XMMatrixTranspose
D3DXMatrixMultiply XMMatrixMultiply
D3DXMatrixMultiplyTranspose XMMatrixMultiplyTranspose
D3DXMatrixInverse XMMatrixInverse
D3DXMatrixScaling XMMatrixScaling
D3DXMatrixTranslation XMMatrixTranslation
D3DXMatrixRotationX XMMatrixRotationX
D3DXMatrixRotationY XMMatrixRotationY
D3DXMatrixRotationZ XMMatrixRotationZ
D3DXMatrixRotationAxis XMMatrixRotationAxis
D3DXMatrixRotationQuaternion XMMatrixRotationQuaternion
D3DXMatrixRotationYawPitchRoll XMMatrixRotationRollPitchYaw (Beachten Sie, dass die Reihenfolge der Parameter unterschiedlich ist: D3DXMatrixRotationYawPitchRoll takes yaw, pitch, roll, XMMatrixRotationRollPitchYaw takes pitch, yaw, roll)
D3DXMatrixTransformation XMMatrixTransformation
D3DXMatrixTransformation2D XMMatrixTransformation2D
D3DXMatrixAffineTransformation XMMatrixAffineTransformation
D3DXMatrixAffineTransformation2D XMMatrixAffineTransformation2D
D3DXMatrixLookAtRH XMMatrixLookAtRH
D3DXMatrixLookAtLH XMMatrixLookAtLH
D3DXMatrixPerspectiveRH XMMatrixPerspectiveRH
D3DXMatrixPerspectiveLH XMMatrixPerspectiveLH
D3DXMatrixPerspectiveFovRH XMMatrixPerspectiveFovRH
D3DXMatrixPerspectiveFovLH XMMatrixPerspectiveFovLH
D3DXMatrixPerspectiveOffCenterRH XMMatrixPerspectiveOffCenterRH
D3DXMatrixPerspectiveOffCenterLH XMMatrixPerspectiveOffCenterLH
D3DXMatrixOrthoRH XMMatrixOrthographicRH
D3DXMatrixOrthoLH XMMatrixOrthographicLH
D3DXMatrixOrthoOffCenterRH XMMatrixOrthographicOffCenterRH
D3DXMatrixOrthoOffCenterLH XMMatrixOrthographicOffCenterLH
D3DXMatrixShadow XMMatrixShadow
D3DXMatrixReflect XMMatrixReflect
D3DXQuaternionLength XMQuaternionLength
D3DXQuaternionLengthSq XMQuaternionLengthSq
D3DXQuaternionDot XMQuaternionDot
D3DXQuaternionIdentity XMQuaternionIdentity
D3DXQuaternionIsIdentity XMQuaternionIsIdentity
D3DXQuaternionConjugate XMQuaternionConjugate
D3DXQuaternionToAxisAngle XMQuaternionToAxisAngle
D3DXQuaternionRotationMatrix XMQuaternionRotationMatrix
D3DXQuaternionRotationAxis XMQuaternionRotationAxis
D3DXQuaternionRotationYawPitchRoll XMQuaternionRotationRollPitchYaw (Beachten Sie, dass die Reihenfolge der Parameter unterschiedlich ist: D3DXQuaternionRotationYawPitchRoll takes yaw, pitch, roll, XMQuaternionRotationRollPitchYaw takes pitch, yaw, roll)
D3DXQuaternionMultiply XMQuaternionMultiply
D3DXQuaternionNormalize XMQuaternionNormalize oder XMQuaternionNormalizeEst
D3DXQuaternionInverse XMQuaternionInverse
D3DXQuaternionLn XMQuaternionLn
D3DXQuaternionExp XMQuaternionExp
D3DXQuaternionSlerp XMQuaternionSlerp oder XMQuaternionSlerpV
D3DXQuaternionSquad XMQuaternionSquad oder XMQuaternionSquadV
D3DXQuaternionSquadSetup XMQuaternionSquadSetup
D3DXQuaternionBaryCentric XMQuaternionBaryCentric oder XMQuaternionBaryCentricV
D3DXPlaneDot XMPlaneDot
D3DXPlaneDotCoord XMPlaneDotCoord
D3DXPlaneDotNormal XMPlaneDotNormal
D3DXPlaneScale XMVectorScale
D3DXPlaneNormalize XMPlaneNormalize oder XMPlaneNormalizeEst
D3DXPlaneIntersectLine XMPlaneIntersectLine
D3DXPlaneFromPointNormal XMPlaneFromPointNormal
D3DXPlaneFromPoints XMPlaneFromPoints
D3DXPlaneTransform XMPlaneTransform
D3DXPlaneTransformArray XMPlaneTransformStream
D3DXColorNegative XMColorNegative
D3DXColorAdd XMVectorAdd
D3DXColorSubtract XMVectorSubtract
D3DXColorScale XMVectorScale
D3DXColorModulate XMColorModulate
D3DXColorLerp XMVectorLerp oder XMVectorLerpV
D3DXColorAdjustSaturation XMColorAdjustSaturation
D3DXColorAdjustContrast XMColorAdjustContrast
D3DXFresnelTerm XMFresnelTerm

 

Hinweis

Sphärische Harmoniefunktionen für DirectXMath sind separat verfügbar. Ein DirectXMath-Äquivalent zu ID3DXMatrixStack ist ebenfalls verfügbar.

 

DirectXMath-Programmierhandbuch