I think it's easy to get lost in the "rendering 2 frames vs. 1" train of thought. It's true that you're rendering 2 frames for every frame in 3D, but those two frames are identical except for the viewpoint.
So in the code, they need to:
1. Update the game world
2. Render a frame
3. Render the other frame
4. Display both frames
Versus 2D mode running at double framerate, which requires:
1. Update the game world
2. Render a frame
3. Display the frame
4. Update the game world
5. Render a frame
6. Display the frame
The point being, just rendering two frames at slightly different viewpoints takes a little extra power, but there's a lot more happening between each frame of 2D than between the Left and Right frames of 3D. So it's probably false logic to assume that disabling 3D automatically allows double the 2D frames.
At least that's my assumptions on the matter.
So in the code, they need to:
1. Update the game world
2. Render a frame
3. Render the other frame
4. Display both frames
Versus 2D mode running at double framerate, which requires:
1. Update the game world
2. Render a frame
3. Display the frame
4. Update the game world
5. Render a frame
6. Display the frame
The point being, just rendering two frames at slightly different viewpoints takes a little extra power, but there's a lot more happening between each frame of 2D than between the Left and Right frames of 3D. So it's probably false logic to assume that disabling 3D automatically allows double the 2D frames.
At least that's my assumptions on the matter.