We have not tested this, but everything should be good for the AD8152 to handle
the TMDS signals for DVI. It would have to run on 3.3V.
However, there is an issue to be aware of. The AD8152 is a 34x34 switch matrix
with full non-blocking performance. One might conceive of using it for an 8x8
This will most likely work, but this configuration does not make use of the
full matrix (not even considering the two unused channels: 32 vs. 34).
A DVI (or HDMI) link has a group of four TMDS channels: Red, Green Blue and
Clock. (technically, the Clock is not TMDS, but it is grouped into the four
signals). For an 8x8 DVI switch using the AD8152, a Red input channel can only
be switched to a Red output channel. Green input to Green output and so on for
Blue and Clock.
A Red input channel will never be switched to a Green, Blue or Clock output,
but the AD8152 has this switching capacity built in. Therefore, only 1/4 of the
AD8152 switching function will be used for an 8x8 DVI switch. This is not a
technical problem, but it will impact the solution cost, and it might be
difficult to compete for such an application.
I would like to point out that it is possible to have 100% utilization, but it
would have to be a much bigger matrix. If you want to make a 34 x 34 DVI
switch, then you can use one AD8152 to switch all of the Red channels, another
for the Green channels, etc. Then four devices can make a 34 x 34 switch with
100% switching utilization.