IRIS

Iris generates correction filters that are processed either with Coda or a 3rd-party convolver. Each of these options is available in VST3 and AU formats for use in all major DAWs. Correction filters are always placed at the end of the signal chain, at the end of the very last bus in the mixer. To keep the filtered sound out of your final renders or bounces, the correction plugin must be disabled prior to DAW export. Alternatively, the correction plugin may be placed on a monitor or listen bus (if one is available in your DAW) to keep the filtered sound you hear away from your active mixer exports.

Basically, the differences between the two are definable in terms of a trade-off between quality and latency.

Frequency domain correction focuses entirely on the magnitude response strength (SPL) measured by each loudspeaker at the listening position, and knowingly ignores the room's phase distortion in the calculations. Only the phase response of the direct sound is considered. This process allows the filter to act only on direct frequency data from the speakers while leaving the room's phase distortions untouched; a condition which leaves the arrival of frequencies in time to the discretion of the room. Thus, it avoids latency at the expense of correction quality. This type of filter is often referred to as "minimum phase". It is a basic filter type that is typically most effective for live tracking scenarios or watching movies, where the quality of the correction is ultimately less important than synchronization of the audio stream with real-time playback of other sources.

Time domain correction runs the full gamut in its calculations without compromise. The room's phase distortions are corrected in accordance with its frequency magnitudes, which adds varying amounts of latency in order to align the time-arrival (time-coherence) of all frequencies at the listening position. This advanced filter type offers superior accuracy of details and clarity of soundstage for listening-focused tasks in which synchronization with other playback sources is secondary.

Yes. Iris generates FIR (finite impulse response) correction filters for use in external processors, such as Coda and other convolution engines. All Iris room correction filters are designed for real-time performance with varying degrees of latency, depending on the filter type. To ensure coherence to latency requirements, it is advisable to use the format tailored to your specific needs:

• Zero latency = minimum phase frequency domain correction: "freq" procedure type.
• Medium latency = raw time domain (RTD) correction: "T" type.
• Maximum latency = raw time domain (RTD) correction: "N" type.

No. The .coda filter is a proprietary 64-bit floating point format that can only be processed by Coda. If you want to use Iris correction in another processor, export your filter in the 32-bit wav+cfg zip format.

Typically, no. There are exceptions, however. Coda, like every other convolver, only processes the output of one filter at a time. Generally speaking, only one instance of your convolution processor should be handling the audio at a time throughout the entire signal path. Two should never be used in series. However, if two would be used in parallel, the scenario would be one in which the first instance contains the filter for a pair of main loudspeakers and the second instance contains the filter for a subwoofer, or a stereo pair of subwoofers. In that scenario, an audio interface with multiple output routing is required to send the mains' filter to the mains and the subwoofer filter to the subwoofer (or subwoofers).