Here's some string/ flow testing pictures, below. The Slurpee cup is definetely MacGuyver, but it's almost the same size as the cylinder bore. Don't laugh!

Here you can see where the air flows the strongest through both ports, when the string (air wave) is longer--the short side flows a lot of air and the long side (opposite of where the strings are coming out of, by the spark plug) is much more difficult to move air. The SC single port heads have a much more even distribution of flow--they still want to flow quicker on the short side, but it's not as difficult to get the string to move through the long side on those heads:
But the problem with the split ports is that while you have two ports flowing air, eventually the air streams tangle up with each other. How long they tangle up with each other, I don't know--this is no simulation of an actual piston under vacuum/ pressure circumstances, nor the appropriate valve lift for the piston distance (vacuum head at the top of the Slurpee cup that can be moved around to vary vacuum source within a cylinder). But it does eventually interfere with each other at some point, when the air streams get longer. This is probably what Ford tried to address with the IMRC valves as a crude variable cam timing, but it's something that TIVCT properly rectifies:
Here's a better look at a low valve lift, and where the airflow is strongest, where there is no interference between the air being moved through the two ports, when the air wave is short (string is short):
Here's a shot of a high valve lift, and the path of least resistance that the air has to take--the straight string is the shorter, upper RPM round port, and the string that's viewed sideways here, was caught in the photo in the middle of chaotic, swirling action, as we can see from the same high lift with the same vacuum. That port is the longer, low end torque port, and this swirling action may be good for mixing with the fuel, but its unclear how much upper RPM, high valve lift damage that it may do from having a stray air stream whipping around. Again, the vacuum source isn't likely to correspond with the piston's location in relation to the valve lift (nor the actual swirl in the cylinder due to the rotation of the vacuum source as the piston moves down), but it does show that the low end torque runner's bends and design aren't as good for creating upper end horsepower. You can also see that the outer edge of the valve impedes the straightest path possible for the air to go: