Eric, my question is who listens too music at 32 kHz,40 kHz,53 kHz ,65 kHz and 90 khz.
If I was a dog I might be a little worried. 
1. I was replying to the false statement that the amp is "just about flat from seemingly DC out to about 70KHz".
2. Each of these responses deviates from flat in the audible band starting at about 3kHz, reaching -0.5 or +1.5db at 20kHz.
3. The ultrasonic SPL into high impedance headphones will be twice the figures shown, or about 6db@53kHz. If you're listening to music at about 100db, you may be exposed to 106db of ultrasonics.
4. Ultrasonic frequencies will intermodulate with both audible frequencies and themselves, with difference products in the audible band.
5. There is some research that exposure to ultrasonic frequencies will temporarily decrease your sensitivity to their subharmonics. (eg, if you listen to a loud 40kHz sound, you will perceived the levels of the subharmonics at 20kHz, 10kHz, 5kHz, etc, at lower levels).
Now, nobody I've discussed this with knows if
- your particular headphones will block the ultrasonics, turn them into heat, or turn them into sound at any frequencies above 20kHz and below the peak at 53kHz; or
- if ultrasonics in the first few octaves above audibility will do damage to your hearing.
I won't rehash all the arguments and workarounds. If interested, please see this
thread or this
thread on headfi. If you have any insights or test gear that can measure for possible ultrasonics with headphones, please join the fray.
- Eric
PS - Tripath's own spec sheet shows a dropoff in the bass of -0.5db at 20Hz that starts at 50Hz.
PPS - For those reading this post out of context, the T-amp only puts out large levels of ultrasonics when driving loads that are significantly different from the 4/8 ohm loading that it's designed for. Most headphones have substantially higher impedance and are thus problematic.
