> Durden[SMTP:[EMAIL PROTECTED] wrote:
>
> I dunno...I'm thinking that optical tempest is probably bullshit 99% of
> the
> time, but what do I know? My Optical specialities are ultrafast and
> optical
> networking.
>
> But I still don't believe that specular reflection of smallish type from a
>
> monitor will have anything that is recoverable. Of course, this is going
> to
> be dependent on the quality of the wall material, but for most not-so-even
>
> plaster/drywall painted surfaces, I just can't believe the appopriate
> spacial frequencies of the image are not scattered after that kind of
> reflection.
>
> The conspiracy theorist is telling me there's some reason they floated the
>
> optical tempest story, though I can't quite figure out what that reason
> is...
>
> -TD
> >
> >
> >BTW, m-o-o-t uses a randomised virtual keyboard with TEMPEST (both EM and
> >optical) resistant fonts. It's okay for inputting keys, but it's a hassle
> >for inputting text.
> >
> >Which means that your keys might be safe from keyloggers (both hardware
> and
> >software), but your plaintext isn't. Sigh. I'm trying to improve it by
> >putting the "senhorita" letters in one block and the rest elsewhere (not
> >for
> >key input obviously), and you do learn where the keys are after a while,
> >but
> >it's still a hassle.
>
Please don't top-post.
Optical TEMPEST of the type we've been discussing has
nothing to do with specular reflection.
As your CRT screen is raster-scanned, the point at which
the electron beam is pointing at any given moment is much
brighter than the rest of the screen. That the screen seems
uniformly illuminated is a result of your persistance of vision.
Optical tempest records the brightness of the light reflecting
off the wall behind the user as a function of time. This can be used
to reconstruct the brightness of each pixel on the screen, since
they are refreshed sequentially in raster order.
Peter Trei
