I'm struggling with a Polk unit too. I haven't had any trouble with Tivo's.
I think that it is a signal strength issue. Using a pinhole does not seem like a good idea to me because it has more effect on the incidence angle than signal strength and if the IR wraps around the edges of the mask you may not accomplish much.
Two avenues that I will try ... when I have time. First, enclose the emitter with increasingly dense tape, paper, or whatever. You can more easily avoid sneak paths if you control emission. Install a variable resistor in series with the emitter. This will allow very fine control over the IR intensity.
In my case there are multiple emitters in the cabinet and it is difficult to mask all of the emitters. It is possible that another nearby emitter is overloading the Polk, even though I have limited emission from the official emitter.
I've seen some units allow/accept enough IR leakage trough the rear of the unit to cause trouble. Simply masking the front panel is not enough.
In my Polk installation, a reflection from the wall opposite the equipment is in the loop. In some cases, if I place my body "just so", the Polk unit will work. While I am not completely satisfied with the following explanation, it may be possible that the Polk unit is sensitive to pulse edges and that the combined IR from multiple emitters and room reflections degrades the pulse edges enough to cause trouble.
As a general note, I have not had reliable results using simple tape to block IR transmission. If I want to completely block IR, I use a relatively large patch of aluminum foil. (you can then cover the foil with a more attractive material) Take care not to flex the foil too much or you may develop pinholes. You may need to cover a wide area in order to avoid sneak paths though the unit's plastic control panel. Always consider the possibility that IR may be entering from the rear. Materials that appear opaque to visible light may be IR transmissive.
Things can become very strange. In one installation I was using an IR receiver and the system had a history of intermittent poor performance. Naturally, when I was on the scene, things worked well. Finally, I took an oscilloscope to the site and observed the signal from the IR receiver. There was obvious noise on the IR receiver's output. After a while I realized that when I stood in certain spots, the noise disappeared and the system behaved. The problem also followed the sun. This was surprising because the system is installed in a shaded room on the north end of the house and the IR receiver faced north. There was a small patch of wallpaper opposite the IR receiver that could receive some indirect sunlight. After I installed a sun filter on the IR receiver the system worked well. Evidently, something on that wall was being stimulated to emit enough IR to degrade the signal to noise ratio of the IR signals.
