Red Audio - twice the bitrate for 2 channels?

[Greg Voevodsky]

I haven't heard much about the audio lately. But I'd like to know, if it isn't very difficult - to have a choice to use only 2 channels at say 24bit/96kHz rather than all 4 channels at 16 / 24 bit, 48KHz?

In other words, same amount of data rate but only using 2 channels instead of 4. Is that hard to do? If not, can we have that as an option in the menu setup please?

[David Fairbanks]

As far as I know the camera is going to have two mini XLRs. So I would assume that it will only have two channels of audio. It would be interesting if it was going to be 24bit/96kHz since I think that would be the highest bitrate of any camera. In fact, I believe that's as high as the Foxtex PD-6 and the like. But since it a 4K camera, why can't it have a nice high bitrate aduio.

[Akcelik]

high dynamic range audio & picture, yes.

tech specs audio show: 4 channel uncompressed, 16 / 24 bit, 48KHz
http://red.com/techspecs.htm

[Rob Lohman]

4 channels indeed

[Justin Anderson]

4 channels indeed

mini XLR?

[Rob Lohman]

at the moment indeed mini XLR

[Jochen Schmidt-Hambrock]

2 channel 96 Khz might be the same data rate as 4 channel 48 Khz, but its quite a different AD/DA convertor. (Different low pass filter, different clock, different limiter, etc.)

All the best,
Jochen

[Rob Lohman]

Josch: very true!

[David Fairbanks]

Four channels would be awsome. Does that mean four mini XLRs?

[Rob Lohman]

yes! (as can be seen from the renders / photos)

[David Fairbanks]

My bad, I see them now. Very cool.

Then the only question is bitrate.

[Matthew Greene]

If you're going to need audio at that high resolution I think it makes more sense to do it double system with the proper gear (great preamps, isolation from other electronic equipment, etc...)

[David Fairbanks]

I agree to some extent. Having a sound guy is essential. If he is doing a backup recording to a Diva, PD-6 or something simular. But If the sound guy can feed signal back to the camera for recording with the footage, it makes post much simpler (faster). We'll just have to make sure that we have mini XLR to XLR adapters in our kit since I can't imagine a sound guy having them.

But if we treat this more like a traditional film camera, the audio never goes to the camera, it's all done in post. It all comes down to how we want to work the total workflow.

[Greg Voevodsky]

How about a compromise? Make it an upgrade OPTION for higher bit, I'd pay for it and Im sure many fellow RED shooters who don't have soundmen following them would too. Thanks.

[Matthew Greene]

It's really hard to get great audio circuits packed in with all those other circuits and get a really clean high SN ratio/dynamic range. That's my concern with the idea.

[Blair S. Paulsen]

Is 48Khz, 16bit audio inherently more resistant to interference than 96/24? FWIW I think 4 channels of 48/16 on board is appropriate for single system field situations. Just clean with decent pre-amps is all I ask.

[Matthew Greene]

I think we're already getting 4 channels of 24bit 48Khz audio. In order to get any audible benefit from high bit depths/sample rates you do need top notch preamps and converters otherwise you don't gain anything other than a bloated data rate. If it happened, this would probably have to be an option due to economics. And it's not that higher resolution audio is directly more suceptible to interference, it's just that you need a really clean signal to benefit from it.

I think a multi channel digital audio in would be ideal, then it would be awesome to write that along with the video at up to 192Khz.

[Evin Grant]

Rob, what is the fifth Mini Xlr for? Timecode? Genlock? I don't see any BNCs.

[Gavin Greenwalt]

I would rather have a really great 48khz ADC than a crappy 96khz.

[Stuart English]

The fith mini-XLR connector is a 5 pin mini-XLR. This provides a two channel audio output which complements the 2 channel headphone output.

[Matthew Greene]

Have you guys decided what levels the outputs would be (e.g. -10dbu, +4dbu)? Also, can you play MP3s on that thing?

[GlennChan]

I would rather have a really great 48khz ADC than a crappy 96khz.

Some well-done 48khz implementations are actually (over)sampled at 96khz, digitally filtered, and then down sampled to 48khz. To avoid aliasing, you need to apply analog and/or digital filtering. For digital filtering, you need to oversample, do your filtering, and then down sample. Which route (analog or digital filtering) is taken depends on the cost/performance of the necessary parts. But if you are oversampling to begin with, it's not much of a stretch to have the option of recording 96khz also.

For the math/theory stuff (from a practical standpoint) on sampling, see
http://www.wescottdesign.com/articles/Sampling/sampling.html
"What Nyquist didn't say"

2- I do agree that well-done 16/48khz is nice to have (especially smooth sounding limiters, which is unrelated to sampling rate). 96khz might simply be a nice side effect of good 48khz.

Workflow-wise, 16/48khz is so easy to deal with.

[Matthew Greene]

My personal view on it is that for dialog higer rates are not necessary. The times when you really can benefit from recording at higher rates you're really going to have a double system setup anyways. That's just how it works out for me, not saying that someone else doesn't have the need for it. I'm not opposed to having greater sampling rates and bit depths available but by no means am I dissapointed by 4 channels of 24bit 48Khz.

Oh, and ditto with the AD/DA converter opinion/info.