Deconvolution at playback

Steps to activate the digital room correction

1. Select one of the target responses you have created before.
2. Press Apply.
3. Check Enable processing if not yet done.

Because any change requires a new calculation of the filter kernel and this causes significant CPU cycles you have to press Apply always after you finished changes.

Filter options

Filter kernel length

This parameter controls computing time versus accuracy. Larger values are more accurate but also introduce more latency. The required length depends on your room. Rule of thumb: the more reverb the longer the kernel. In most cases the default value of 49152 should be suitable.
The selected value is used for playback at 32, 44.4 and 48 kHz sampling rate. For 16, 22.05 and 24 kHz only half of the selected size is used. And in case of 96 kHz the size is doubled, well if you managed to get a 24/96 sound device working with OS/2.

Apply Hamming window

Optionally a Hamming window function can be applied to the filter kernel to reduce artifacts especially at low frequencies. This effectively causes further smoothing to the target response. You may show the difference by using another view mode (see below).
In general it is a better advice to use reasonably smooth target responses instead of a window function. That gives you more control about the result.

Filter frequencies

The two check boxes control whether frequencies below (subsonic) and above (supersonic) the domain of definition of the target response are rejected. I.e. if your selected target covers the range of [20 Hz, 20 kHz] and you enabled both boxes then frequencies below 20 Hz are damped by 3 dB per inverse filter kernel length (e.g, 0.9 Hz for 49152 kernel at 44.1 kHz) and frequencies above 20 kHz are damped by .1 dB per inverse kernel length.
While high end freaks usually dislike discarding supersonic frequencies it could protect your tweeters from unwanted inaudible noise. In fact you won't notice any difference in A/B blind test. It on you whether you favor it. But you should keep the subsonic filter active. It usually improves sound reproduction because it reduces movement of yous bass bins at very low frequencies which causes heavy intermodulation artifacts.

View mode

Besides the default view of the selected target response there are three further view modes available while a playback is active. They are intended for advanced users to control what actually is going on during playback. In the frequency and group delay view you can compare the effect of the currently applied filter at the sampling rat of the current son with the selected target response. Usually there should be only small differences at low frequencies. Otherwise you should chose a larger filter kernel.

The time domain view shows the real values of the applied filter kernel in the time domain. It should return close to zero at both ends.

The dimensions of the graphs can be controlled by the extended playback options page.