Direct sound

Taking advantage of the sensitive reactivation of film production in Colombia, we will deal this time with the issue of direct sound for what has been called "digital cinematography", resorting to a fresh filming experience, interesting for the rest, thanks to the technological context within which it was developed. At the moment, the project is in the process of being assembled. This is the feature film "The River of Time" by the French-Swiss director Barbert Schroeder (Barfly, The Mystery of Von Bulow, Single Woman Seeks).

This film was shot on video, with two cameras, with Sony's HDVS technology (High Definition Video System) to be transferred later to 35 mm.

The selection of the material for the sound recording required taking into account the need to have 2 basic configurations: The first would be "heavy", in the sense of being able to manipulate 5 wireless microphones plus the signal from a cane microphone. In addition to this, to be able to ensure the mixing and sending of signals to the two cameras, to the DAT recorder and to the hearing aid monitoring system. The second configuration could be called "lightweight" and would involve only 2 wireless microphones and 1 cane microphone whose signals would go directly to the DAT. For this, a Fostex PD-4 DAT recorder with Time Code (TC) was used, which offers a 3-input mixer in its structure and allows sound recording in production conditions similar to documentary production where a single camera and low lighting are used.

The "sound cart" was also made up of a Cooper Sound CS-106 console with 6 monophonic inputs and an MS stereo, two outputs (L, R) and two monitoring outputs A and B. This console with excellent electroacoustic characteristics made it possible to filter the signals thanks to semi-parameter equalizers at 3 bands (low, medium and high frequencies).

The versatility of the console consists in the feasibility of sufficient handling of microphone inputs or lines, and, in addition, the possibility of transporting line signals to the cameras and to a Comtek wireless monitoring system, for constant review with headphones by the director, the person in charge of continuity and even for an assistant director.

The signals sent to the two cameras were the result of a monophonic premix recorded on channel 1 of the Sony HDW-700 cameras (16-bit PCM recording). This premix is a reference sound to be used in image montage on an AVID system. In the DAT the registration was made in "bipista", that is, in two different channels, the left (L) with the information coming from the rod and the right (R) with which it came from the wireless microphones (HF). The premix of the cameras is the result of mixing these two information and sending them via cable or via HF when one of the cameras was operated on a Steady Cam camera stabilizer. The L and R channels of the DAT will be used in the conformation of the Time Code (TC) that corresponds exactly to the Time Code in the cameras (Free Run@ 29.97 ndf).

For the recording of environments (wild tracks) the MS stereo input of the console was used or sometimes directly to the FOSTEX DAT. This allowed monitoring by an MS decoder built into the hearing aid circuit.

The wireless (HF) chosen were the Audio Limited (TX-2000 transmitters and DX-2000 receivers) UHF 600 Mhz and "Diversity". To facilitate the connections of the receivers with the console, the rack of 4 Audio Limited DK-2000 receivers was used in which the DX-2000 receivers can be "embedded" to be powered by a single power source and work with a single pair of external antennas.

In total, we used 5 HF, maximum 4 for the actors and the rest to send signal to the camera with more mobility (Steady Cam). In addition to the HFs, we use condenser microphones on the cane: outdoors Neumann KM81i and indoor Schoeps CMC-5 with the hypercardioid capsule MK-41. All hearing aids were Sennheiser HD-25. Finally, a DAT HHB Portadat recorder was available for security.

The project "El Río del Tiempo" was recorded in Colombia recently, in the city of Medellín, under quite worrying ambient noise conditions. The recording of direct sound in these circumstances was very complex, which determined the selection of wireless microphones to capture the voices in proximity, thus ruling out the use of the cane technique as the main capture. This would have "picked up" much more noise than HFs.

The typical case was that of two actors, each with HF and the reed supporting the scene to help rescue a certain naturalness of the environment, thanks to a weighted mixture sufficient to not mask the dialogues.

The microphone capsules used with the HF were initially Tram TR-50 but after presenting breaks in the cables at the level of the connectors it was necessary to replace them with the Sennheiser MKE-2 (cylindrical unlike the TRAM that are flat), which we kept as a reserve. The permanent use of HF transmitters hidden in the locker room can cause breaks in the cables due to the constant movement of the actors.

The lead actor had a fairly low voice register that caused a strong accentuation of the low frequencies at the time of pickup with a microphone stuck to his chest. It became necessary to filter it to achieve a more homogeneous balance with respect to the actors who did not present the same problem but who, on the contrary, had very bright voices, loaded with a lot of "seseo" (increase in the sound of the letter "s").

The tonal balance of the voices is easier to obtain when using a microphone on the reed directed from above the characters. In our case, thanks to the equalization of the Cooper console, we were able to homogenize in an acceptable way the vocal register of all the actors.

For the room scenes (semi-naked characters), where it was impossible to use HF, and taking into account the poor acoustics of the location, we chose to work with the Schoeps CMC+MK-41 microphone of excellent performance in difficult conditions of reverberated spaces, with little sound absorption.

The modulation of several HFs and the microphone in the reed involved a great concentration and knowledge of the texts on the part of the sound engineer. The modulation technique requires that only the microphone of the speaking character should be fully open and the remaining ones modulated approximately 15 decibels (dB) below or completely closed. In this way it is avoided to capture too much noise from the environment and also, the effects of "comb filter" (Comb filter in English) that alter the timbre of the voices are avoided. The L channel with the information of the rod allows to maintain a "stable" environment so that the constant movement of the modulation that occurs in the R channel is imperceptible.

The recording of direct sound with the aforementioned equipment and conditions allows to obtain an optimal quality and texture so that the process of premixing dialogues complements and takes advantage of the great expressiveness that is achieved through the technique of cinematographic direct sound.