Homebrew PCBs

Many of the of "professional" lab/prototyping PCB exposure units seem to 
be fitted with a vacuum assisted frame. I'm wondering if vacuum has much 
advantage over conventional sandwiched rubber foam and glass.

These commercial exposure units are built using fluorescent lamps at 
point blank range. i.e light goes all directions. I could not think of 
better way to promote light undercut. I'm thinking the vacuum on the 
more expensive units is put there to help combat non-collimated light.

If the light source is reasonably collimated then does vacuum provide 
much advantage ?
Photoresist is about 50um thick. So I guess ideally the photomask should 
sit flat on the PCB with a air gap no greater than some fraction of the 
resist thickness.