Realize the A8 Smyth is a multichannel processor for headphones, compatible 5.1 up to 8 channels (8.0 or 7.1). It allows to externalize listening through any headphones using the simulation of our transfer functions HRTF (Head Related Transfer Function). The transfer functions HRTF are the frequency response curves of our heads including our shoulders and the lobes of our ears, they differ for each corner of locating a source. We have for example, for a given source at 120 ° to 0 ° axis of the center speaker (which corresponds to the left rear speaker and right rear) a loss of about 10 dB between 1.5 kHz and 7 kHz masking effect of our head.
Today there are various solutions reproduction space for the helmet. All apply the principle of correction HRTF channels incoming from general information or averaged shapes and dimensions of pinnae, heads and torsos.
Without virtualization process, a monophonic sound listened through headphones positioning the source at the center of the head, while a stereo signal sounds positioned on a line from left to right through the center of the head, but never outside as in real life. With a multichannel source, processes effectively open virtual surround sound perception and sometimes even extend beyond the helmet. The location of the rear surround channels often works well. In contrast, the frontal location is more limited, including a center channel difficult to "see" ahead. In any case, none of these methods can not claim to perfectly reproduce the location obtained from five or more speakers distributed around the listener.
Yet, in its basic principles, the Realize is similar to other systems: it applies to an alteration HRTF different ways of multichannel. Nevertheless, the Realize that and more on three critical components missing from other processes: personalization, tracking head movements and capture properties of any sound system in its original environment. Customization is done by measuring our own HRTF functions and tracking head movements is done by recalculating the parameters of these functions every 5 ms. This process is also designed to simulate a multichannel reproducing his identity and the acoustics of the room in which we make the measurement. One can imagine the transport in our multichannel processor favorite, well optimized, anywhere.
The calibration is a two preset time, we calibrate initially the system to measure our own HRTFs functions, the frequency response of speakers and room acoustics measurement, these measures are called IRB (personnal room impulse response ). In a second step we calibrate the system to fit the helmet used, the extent HPEQ to linearize frequency restitution. So we end up with two measures divided into a preset. We can then keep the IRB made and measures to change headphones or keep the same helmet and load another multichannel measured previously stored in memory. The unit has 64 IRB submissions and 64 submissions HPEQ loaded into four user presets. Realize the headset comes with a Stax SR-202 as an option but can work with any other stereo headphones.
Following the listening tests performed, the reaction of most people is to ask to cut the multichannel to start the tests ... so that the tests have already begun! The tests were done on a listening Dynaudio Acoustics AIR 6. Among the various processes for multichannel headphones already tested (Arkamys, Fraunhofer, Dolby headphones, Headzone Beyer), the Realize offers great potential in terms of spaciousness. After comparison, the locations of the various channels of the headphones correspond precisely to the locations of the multichannel speakers. The rear channels are still very impressive because they are new directions from which we really do not usually receive the helmet. We also find the identity of the speakers and the room (reverberation). The high resolution calibration (sweep duration of 12 s against s 3 in low resolution) significantly improves the external perception in frontal and acoustic rendering with better depth perception.
Realize we find in the eight most management settings of multichannel sound, the choice of angles of the speakers, setting the reference level (85 dB SPL film C or 79 dB SPL C HDTV), setting the configuration to Lfe + 10 dB or 0 dB, with the bass management crossover frequency to 80 Hz, the stereo downmix with the opportunity to learn the coefficients in the center and rear, this reduction being distributed without spatialization, the traditional way. The inputs and outputs are analog RCA, there is also a headphone output SPDIF digital, audio outputs for connecting auxiliary tactile transducers (shakers body) to enhance the sensation of vibration to the body and a professional version is expected by the second quarter of 2010 (I / O SUB25 analog and AES). Realize you can synchronize two if you want restitituer multichannel formats up to 16 audio channels allowing to explore the vertical restitution (10.2, 10 aurophonic system ...).
There are many applications both in the workplace (HD cars, sound ...) and the general public (video games, movies, music ...), but it is mostly a whole new way to listen to the helmet very attractive and more natural, comparable to the transition from stereo to listen to 5.1. Realize the reference can be identified by reminding our listening and 5.1 optimized test the compatibility of our mix. You can record the output of Realize and transport format to a mobile player to enjoy listening space. One can also imagine a box optimized for miking reportage / documentary that would last to the sound of having a spatial rendering of its decision-channel sound other than by listening to front or rear or downmix remained very limited! A test is essential!
Learn more:
