Originally Posted by adidino
The XLR does have about +6db gain for the rca in my experience.
The XLR output has 6dB gain as it's two signals in opposite phase of the same gain as the RCA. The XLR input loses that 6dB, so the overall system gain ought to be the same using XLR-XLR or RCA-RCA. That's the theory, anyway. A 90dB sensitivity speakers ought to need about -5dB trim to get to 75dB, so all looks textbook correct there. (Speaker sensitivity is measured with 1-watt wrt 85 dB SPL at 1 meter. AV cal noise looks for 75 dB with a 10dB lower drive signal at the listening chair.)
However, +9db on the sub sounds awefully high. What did you use to calibrate output levels? If you're using the internal tone generated by the SSP800, I wouldn't recommend that. Try using Avia or THX audio setup from most THX dvd movies. Also, the surrounds are about 3db too hot relative to the mains if you calibrate through the 800 speaker level menu.
I'm assuming the total amount of bass sounds correct. Only the knob trim settings seem too far off. If that's the case, raise the gain on the F112, and turn down the trim in the SSP. Same sonic result, less trim range.
When I use the SSP's internal tones, my sub reads about the same as with a THX Optimode signal. And both of those read a few dB on the hot side wrt some other test discs on hand. This is due to there being differing standards for filtering the bass signals, and the fact that my room's bass response is not ruler flat.
Originally Posted by hifisponge
I'm just doing it by ear right now because I know that the tone generator is not working properly. Which by the way is something that they haven't gotten right from day one. Doesn't seem like a difficult fix.
From where I sit, the tone generator in the SSP has been fixed since 0062. Code 2.0 0064 has it right, also. I've checked it with external voltmeters on the line outs (to avoid any question of SPL meter accuracy), and vs other test discs. One of our compadres at "another forum" where we're discussing these matters reportedly has a hardware problem causing the noise level error. I have not seen any other evidence that the SSP signals are wrong in the latest code rendition. Yes, in the 0058 preliminary code shipped with early dual DSP boards, it was wrong. Might have been wrong in the single DSP, too but that's before my time.
I'll run the test disc route tomorrow, but I'm pretty certain the levels won't change more than a couple of dB. Even before the upgrade, I had a spread of 9dB, which I've always thought was odd, but just didn't mention it until now. I just put a Y splitter going into the sub which gave me 6dB, but that still leaves with with a 9dB difference. I wonder if it is a problem with the sub?
Where is the master gain set now? If Fathom would have published a sensitivity spec, or had a calibrated THX setting, we'd know if it was behaving correctly or not. Since they did not do that, all you can do is keep turning up the gain until the trim offset is mostly neutralized.