Audio Performance of the Red Camera (Red One)

Last updated
Since it began shipping, Sound Devices has received numerous inquiries concerning audio interconnection with the Red One camera and our field mixers and recorders. We elected to bring an example of the Red camera in-house for audio testing. We are fortunate to have friends at Fletcher Chicago. Fletcher has an extensive arsenal of cameras and they were kind enough to arrange for a Red for our testing needs. We picked up unit #529 with build 15, v2.2.5 on May 21, 2008. This unit appeared fresh out-of-the box. Disclaimer: While we believe our measurements to be accurate and empirical, note that we do not guarantee that the performance we observed is the actual performance of every camera. You may want to do your own tests to verify performance for your specific application.

Microphone Preamplifiers

Based on information received directly from Red and from the many user reports on the Red forum, Sound Devices elected to concentrate audio tests on performance of the camera's line-level inputs. Our brief listening test of the microphone inputs verified that they are not suitable for anything but a scratch track, especially when phantom is activated. Whether microphone preamplifier performance is a hardware issue or something that will change in the future with a firmware revision is unclear. It is no doubt that Tom Fletcher appreciates that we didn't take apart the brand new camera body to see for ourselves. Sound Devices recommends connecting to the Red One at line-level only.

Interconnection and Cables

The Red has four, locking TA3-F connectors, each of which can be switched between line-level or mic-level. In the mic-level position, gain is menu-adjustable between 20 dB and 54 dB. In the line-level position input sensitivity is fixed with no adjustment. The Red Camera is supplied with short XLR-to-TA3 adapter cables. These cables have resistors built-in to change impedance and reduce signal level. When used without these cables, the Red inputs can be characterized as "-10" level (full scale on the meter = +8.2 dBu) with 600 ohm input impedance. With the supplied cables, the inputs can be characterized as accepting signals just below full line-level (full scale on the meter = +19.4 dBu) with 2000 ohm input impedance.

Line Input Specs (without Red-supplied special cables)

  • Input impedance: 600 ohm
  • Full scale input level: +8.2 dBu (2.0 V rms)
  • Nominal input level (for 20 dB headroom): -11.8 dBu (200 mV rms)

Line Input Specs (with Red-supplied TA3-XLR padded cables)

  • Input Impedance: 2000 ohm
  • Full scale input level: +19.4 dBu (7.2 V rms)
  • Nominal input level (for 20 dB headroom): -0.6 dBu (720 mV rms)

Line Inputs (with or without Red-supplied cable)

  • Frequency response:  flat, -2 dB @ 20 kHz, -2 dB @ 8 Hz, referenced to 1 kHz response (see graph)
    • Frequency Response
  • THD+N: 0.01% (22 Hz-22 kHz BW) best case, input to output at +5 dBu input, padded cable (see graph)
    • THD Sweep with Impedance Balanced Input
When the inputs are driven without equal voltage on each leg of the connection (impedance-balanced sources), the THD+N starts increasing 6 dB sooner than when driven with equal voltages. The chart above shows the THD+noise with impedance-balanced sources. The chart below shows THD+noise driving both pin-2 and pin-3 with signal.
    THD Sweep
  • Dynamic range, input to .WAV file, best case:  89 dB (flat weighting, 22 Hz-22 kHz), 91 dB (A-weighting), 14.8 bits equivalent
  • Dynamic range, input to analog output: 95 dB (flat weighting, 22 Hz-22 kHz), 97 dB (A-weighting), 15.8 bits equivalent

Analog Outputs:

  • Full scale output level: +8.2 dBu (2.0 V rms)
  • Nominal output level (for 20 dB headroom): -11.8 dBu (200 mV rms)
  • Output impedance: 200 ohms

Headphone Output:

  • Full scale output level: +2.2 dBu (1.0 V rms)
  • Nominal output level (for 20dB headroom): -17.8 dBu (100 mV rms)
  • Output impedance: <10 ohms

Level Meters

Audio level metering appears in the lower corner of the viewfinder output (and HDMI) screen as four 20-segment, tri-color meter bars. There are no numerical markings. In our tests we have determined that the 0 dBFS position corresponds to a line level signal of 20 dBu and the lowest reading corresponds to a line level signal of -14 dBu when using the Red-supplied XLR-to-TA3 cables. There is a solid line shown at an arbitrary -8 dBFS and a very faintly emboldened line at -20 dBFS. The chart below shows the meter scale. Red Camera Meter Levels

Use and Setup with Sound Devices Mixers and Recorders

The schematic below illustrates an interconnect cable between Sound Devices mixers and the camera's line-level input. We recommend this connection to properly pad the Red camera TA3 input to handle the line-level outputs of Sound Devices products. This circuit attenuates the Red input a few dB more than the supplied Red cables and provides a better match between the SD outputs and Red input. With this cable and the settings shown below, the meters on the SD products can be used to meter the program audio with confidence. Padded Cable for Use with Red One Inputs With this cable, SD products will drive the audio circuits of the Red to their theoretical best and will give acceptable fidelity. Following are the settings for each SD product:
  • 442: use either the XLR, multi-pin, or TA3 out, set output to line-level
  • 302: XLR output: set attenuation to 0 dB (default)
  • MixPre: XLR output: no settings necessary
  • 788T: set Analog Output Gain to -2 dB
  • 744T, 722, 702T, 702: set Output Attenuation to 0 dB

Conclusion

For most dialog applications, the Red One's near 16-bit audio performance is similar to many of the digital pro-sumer and pro cameras we have tested. This is perfectly acceptable for dialog, especially when hit with a good, clean line-level signal. Sound Devices recommends dual-system sound for critical applications.