C#

<br />
double MyFloat = 123.456;<br />
string MyString;<br />
MyString = String.Format("{0:00.00}", MyFloat);<br />
Console.WriteLine(MyString);<br />

<br />
double MyFloat = 123.456;<br />
Console.WriteLine(String.Format("{0:00.00}", MyFloat));<br />

<br />
....<br />
MediaState mediaState = new MediaState();<br />
// hook up events<br />
mediaState.MediaCenter.Started += new EventHandler(OnMediaCenterStarted);<br />
mediaState.MediaCenter.VolumeChanged += new EventHandler(MediaCenter_VolumeChanged);<br />
// Start receiving media state information<br />
mediaState.Connect();<br />
if (mediaState.MediaCenter.Active)<br />
{<br />
   label8.Text = "Media Center is Active";<br />
}<br />
else<br />
{<br />
    label8.Text = "Media Center is NOT Active";<br />
}<br />
....<br />

• You must marshal data from unmanaged to managed to unmanaged to unmanaged to managed (that seems right) in one direction, and vice versa for the other (to call methods on the running object).
• You must P/Invoke many native functions and re-declare several COM interfaces in managed code (see ComImportAttribute for starters) manually since several of these won't interop correctly using VS.NET or tlbimp.exe (like methods that can return S_FALSE, which is still a success code)
• Your adding complexity to a design that will most likely introduce bugs
• You really end up writing a COM client and server in managed code, not taking advantage of the managed features of the CLR.