TOPIC:

I'm looking for guidance on how to avoid clipping in my C-DSP DL setup. I have a tune that gives truly amazing imaging, but sounds somewhat fatiguing and unnatural. This after many iterations of adjusting the mid slope and treble rolloff in Dirac. In the treble Dirac is boosting 10ish db above the measured response even though I'm rolling off starting at 6.5k and ~6db down by 10k. I tried dragging the horizontal curtain so that Dirac does nothing above 10k, and found the sound more natural sounding but the imaging much worse.

Based on the below post from the Dirac forum, I'm about to do a new calibration with the following tweaks:

- Increase analog amplifier gains so that the measured signal is above the target, on the assumption that Dirac is much better at reducing gain than increasing it.

- Turn down the gain at the final stage (just before the DAC according to the post below) to avoid digital clipping. 6-10db.

- Perhaps turn up the gain at the initial stage (just after the ADC, or in my case toslink receiver) to compensate for the above.

As the post below states, it would be nice if the documentation and/or devteam could help guide us in this. Anyone who has delved into this please comment about what you have discovered.

Here is the post I'm referencing:

www.minidsp.com/forum/dirac-series-suppo...nting-clipping#46176

- If the signal applied to the analogue inputs exceeds 4v or 2v RMS, then the ADC will clip.
- Internally, there is no headroom issue for practical purposes (I read elsewhere in the forum that the headroom is at least 100 dB internally, because the hardware uses 32 bit arithmetic).
- If any filter elements that are applied in the digital domain can result in any part of the digital signal being boosted above 0dB, then it is essential to apply a corresponding overall level cut before outputting to the DAC, otherwise the resultant analog signal will exhibit clipping (i.e., any part of the waveform that cannot be represented in 24 bits is truncated to 24 bits).

It would also be helpful to the user to understand that the RMS meters in the plugin show the level *after* any analogue signal has been through the ADC (in the case of the meters on the inputs) and *immediately before* the signal is output to the DAC (in the case of the meters on the outputs), and therefore, if you see the output levels peaking in the positive dB range, then you haven't applied enough of a cut.

Progress report for those who follow along at home:

I tried boosting the tweeter amp gain nearly to the max in order to bring it up enough that the pre-cal signal could follow my target curve. The results sounded very underwhelming after calibration, not to mention the out of control noise floor. Another thing that might have been related was that I followed the instructions in the C-DSP manual for setting Dirac measurement levels, which yielded higher signals (and lower noise) than the instructions in the right pane of the Dirac program. Even though I kept it low enough that I didn't hear distortion on the sweeps, I'm thinking it might be better to keep the output level lower so as to ensure the drivers are in the linear zone for calibration.

Up until this point I avoided doing any tuning in the C-DSP plugin other than crossovers, figuring it was best to let Dirac do the flattening. But I'm thinking my next step is to try and flatten frequency response to get closer to my target using the C-DSP parametric eq. At least to flatten a big broad 5k room response peak that exists in my dash-mounted mid/tweet system, which was made worse by the tweeter level increase.

I've find it beneficial to at least cut the large peaks in the driver's response using the plug-in before attempting Dirac. Doing so makes it much easier to have good results on the DLCT levels screen.