This is really a fluid dynamics problem. Anytime you have a bottleneck or flow constriction like where the vacuum hose connects to the table you will have a pressure drop. Especially if there is a long channel after the bottleneck. This means you'll have little suction at the ends of the channels (ends of the core).
The channels in this photo likely work because the cross-sectional area of the vacuum hose is not dramatically different than the combined cross-sectional area of the channels connected to that hose. The more these cross-sectional area differ, the greater the pressure drop.[/img]
mine are like this
seem to work pretty well
i usually use some double stick tape at each end just to hold the ends down better.
2 vacuum spots seems a better idea.