CODEWITHSUNDEEP
visual-c++directxdirectx-11
Brooks
Brooks

Reputation: 167

XMMATRIX operator() working with xnamath.h but not with DirectXMath.h

I am new to directX programming and Visual C++ and I am having an issue migrating an example I found from xnamath.h to DirectXMath.h. I am using Visual Studio 2012.

The purpose of the code is simply to initialize a XMMATRIX and then display it in the console. The original code is shown below (it works just fine): 

#include <windows.h>
#include <xnamath.h>
#include <iostream>
using namespace std;

ostream& operator<<(ostream& os, CXMMATRIX m)
{
    for(int i = 0; i < 4; ++i)
    {
        for(int j = 0; j < 4; ++j)
            os << m(i, j) << "\t";
        os << endl;
    }
    return os;
}

int main()
{
    XMMATRIX A(1.0f, 0.0f, 0.0f, 0.0f,
               0.0f, 2.0f, 0.0f, 0.0f,
               0.0f, 0.0f, 4.0f, 0.0f,
               1.0f, 2.0f, 3.0f, 1.0f);

    cout << "A = " << endl << A << endl;

    return 0;
}

When I run the program it gives the following output:

A =
1       0       0       0
0       2       0       0
0       0       4       0
1       2       3       1

Press any key to continue . . .

However, when I change the headers to DirectXMath it no longer works:

#include <windows.h>
#include <iostream>
#include <DirectXMath.h>
#include <DirectXPackedVector.h>
using namespace DirectX; 
using namespace DirectX::PackedVector;
using namespace std;

ostream& operator<<(ostream& os, CXMMATRIX m)
{
    for(int i = 0; i < 4; ++i)
    {
        for(int j = 0; j < 4; ++j)
            os << m(i, j) << "\t";
        os << endl;
    }
    return os;
}

int main()
{
    XMMATRIX A(1.0f, 0.0f, 0.0f, 0.0f,
               0.0f, 2.0f, 0.0f, 0.0f,
               0.0f, 0.0f, 4.0f, 0.0f,
               1.0f, 2.0f, 3.0f, 1.0f);

    cout << "A = " << endl << A << endl;

    return 0;
}

When I try to compile I get an error with os << m(i, j) << "\t"; which says: 

error C2064: term does not evaluate to a function taking 2 arguments

When I hover over the red squiggly line under m(i, j) it tells me:

DirectX::CXMMATRIX m
Error: call of an object of a class type without appropriate operator() or conversion function to pointer-to-function type

Any advice would be greatly appreciated.

Upvotes: 5

Views: 2480

Answers (3)

NoCake
NoCake

Reputation: 84

In Direct X 11+, the ability to access a matrix directly as

matrix (row, column)

has been removed, due to performance issues. Microsoft recommends accessing the values via the r member. I suggest using 

XMStoreFloat4 (row, column)

for a 4x4 matrix, as then you do not need to worry about the data type.

ostream& operator<< (ostream& os, CXMMATRIX m)
{
    for (int i = 0; i < 4; i++)
    {
        XMVECTOR row = m.r[i];
        XMFLOAT4 frow;
        XMStoreFloat4(&frow, row);

        os << frow.x << "\t" << frow.y << "\t" << frow.z << "\t" << frow.w << endl;
    }

    return os;
}

Be careful when using _XM_NO_INTRINSICS_ as this affects more than just accessing matrix values, may have performance hits in performance sensitive code and may impact other operations. DirectXMath is a jump from XNAMath... and when upgrading old code, it can be a pain, but it is worth it.

Upvotes: 1

Nathan Beasley
Nathan Beasley

Reputation: 31

I changed the example code to use 

ostream& operator<<(ostream& os, CXMMATRIX m)
{
    for(int i = 0; i < 4; ++i)
    {
            for(int j = 0; j < 4; ++j)
                os << m.r[i].m128_f32[j] << "\t";
            os << endl;
    }
    return os;
}

This will have the same effect as the old xnamath had.

Upvotes: 3

Carlos Ch
Carlos Ch

Reputation: 403

It depends on the version you are using for DirectXMath, you can define _XM_NO_INTRINSICS_ to get the result you want. See http://msdn.microsoft.com/en-us/library/windows/desktop/microsoft.directx_sdk.reference.xmmatrix(v=vs.85).aspx for more information

Upvotes: 2

Related Questions