I just about blew that attachment off because it’s very small print but am happy I did read it.
These are screenshots from the PDF material (see pages), so you can read it more comfortably.
I also didn’t know previously about that in sensitivity 12:
-
the threshold T-level (minimal stimulation, which is yet perceived) is activated when sound has 25
-
the C-level is activated when sound has 65 dB. Or above that 65, but then it’s compressed by (infinite compression ratio, IIRC)
In sensitivity 1:
- T-level activated only at 38 dB
- C-level activated when sound is at 81 dB and above (according to the graph)
Strange, when IIDR=40, it should be 38+40 = 78 dB activating C-level.
I wonder does the graph would look like the same if IIDR (Instantaneous Input Dynamic Range) will be increased. You can ask what’s your IIDR.
You’re ignoring this member’s posts.