5D MkIII-dual sync sound still needed?

[TheDingo]

If you use a clean preamp to allow recording in camera with a mic level of +1 above 0 while hitting -12dB, you'll get a very low noise floor: effectively no hiss

This depends on the camera. I owned a Canon 60D for a few months before switching everything over to the Panasonic GH2.

The Canon 60D has a noise-floor of -52 dB, which is too noisy for me, as there is NO room for any audio recording errors.

The Panasonic GH2 has a noise-floor of -63 dB ( same as my HMC-150 ENG camera ), which is better, but it also does not leave much room for any audio recording errors.

The Tascam DR-100 MK2 recorder in LINE or MIC ( Low and Med pre-amp setting, High is too noisy ) level has a noise-floor of -83 dB, which gives you lots of room for audio corrections.

...I do record audio directly to the GH2 when I want a quick'n'dirty set-up, but otherwise I always record to an external digital recorder. ( Tascam DR-100 MK2, Olympus LS-11, Sony PCM-M10 )

[jcs]

jcs, your video is set to 'private'

Sorry about that- video should now be available.

http://www.youtube.com/watch?v=2bWGogt2Kxg

[TheDingo]

I can't get your YouTube link to work through the office firewall ( most YouTube videos are fine, but for some reason yours won't play )

Is there any chance you can sort out your Vimeo link. When I click your Vimeo link I see the message that the video is private and thus I can't view or download it.

Sorry about that- video should now be available.

http://www.youtube.com/watch?v=2bWGogt2Kxg

[jcs]

The last change was to "Unlisted". It's now public: http://www.youtube.com/watch?v=2bWGogt2Kxg
Vimeo version now public, with download option: https://vimeo.com/39445020

[dave12781]

I care about sound quality. And, I'd prefer the dual system for recording audio and video separately... but you don't always have the time or man power.

I have a 5d3 and I use a juicedlink rm333 which I think gives me great sound directly into the camera. It has 3 xlr inputs - 2 channels mixed to the right and the left stays independent.

I had previously purchased a bunch of wireless lavs, so I just strapped the receivers to the juicedlink on my rig and BAM! Great audio with an easy solution and it is 100% wireless.

Now all I have to do is wear headphones when I shoot!

[jcs]

Did you consider a juicedlink solution over the SoundDevices? What 'noisy' mics did you use?

JuicedLink is more practical for a mobile camera setup. SoundDevices USBPre2 also works as an ultra-high quality computer audio interface (can also run off USB battery for remote work on a tripod for). SD USBPre2 can pass digital output into digital in of low-cost recorders (such as the DR100MkII). It's so clean going into camera when the camera gain is +1 above 0 I don't bother with separate audio (though 24bit 96kHz may come in handy in the future when using separate audio).

Mics used: MXLR144 ribbon and Shure SM7B dynamic. Both mics require a lot of gain (50-60+dB)- without a clean preamp, they are noisy.

From my tests, the SD USBPre2 into camera with gain of +1 is cleaner than same mics into DR100MkII stored digitally at 2496 (DR100mkII purchased as a digital bit bucket: preamps are better than H4N, etc., but still not very good compared to SD level gear).

The new JL devices have a 'bracketed' audio solution to prevent clipping, however the SD limiter is amazing and transparent. For this reason I'd pick a MicPre-D over JL gear: http://www.sounddevices.com/products.../key_features/

For 3 channels: http://www.sounddevices.com/products/302/
If JL had a decent limiter, would recommend over SD if cost an issue (additionally- I like to do Mid-Side (M+S) recording with the AT BP4029 shotgun and need both channels: not possible with JL if using their 'bracketed' output).

An older test with 5DMark II: AT BP4029 into SD USBPre2 => digital out to DR100mkII => analog line level out into Studio 1 XLR Pro => into 5DMark II: http://www.youtube.com/watch?v=ulWOV0WujuQ

[TheDingo]

Listening to the 60 MB Vimeo download version, the audio sounds muffled, like the upper range is missing. I have no idea if this is the mics or the Canon recording.

Also in the background there is a definite "fan" or AC unit hum that is not helping at all.

...As a side note, the 5D MK3 image looks great. Very nice indeed.

[jcs]

Listening to the 60 MB Vimeo download version, the audio sounds muffled, like the upper range is missing. I have no idea if this is the mics or the Canon recording.

Also in the background there is a definite "fan" or AC unit hum that is not helping at all.

...As a side note, the 5D MK3 image looks great. Very nice indeed.

The Canon's 48kHz signal is not low-pass filtered in-camera. Both mics are 'smooth' by their design (especially the ribbon) and not bright as condenser mics. 99.99% of listener's can't hear the hum- no reason to spend time removing it.

This condenser mic version is brighter (and I cut high frequencies in post): http://www.youtube.com/watch?v=ulWOV0WujuQ

Jon Fairhurst posted demos on Cinema5D a few years ago showing that the Mark II with a good preamp sounded better than any of the sub $500 (and perhaps sub $1000) external recorders: http://cinema5d.com/viewtopic.php?f=32&t=3727

See this chart to see the noise levels of many common external recorders: http://www.avisoft.com/recordertests.htm

The USBPre2 has the same hardware topology as the 722 (top of the chart). I also have the DR100mkII- its AD converters are very noisy compared to the SD ADs.
In terms of noise performance and cost, the Fostex FR-2LE is the best deal.

[TheDingo]

99.99% of listener's can't hear the hum- no reason to spend time removing it.

Yes, for typical bad laptop or desktop speakers I would agree with this, but as a standard practice I try and produce the best audio I can every time I shoot.

Jon Fairhurst posted demos on Cinema5D a few years ago showing that the Mark II with a good preamp sounded better than any of the sub $500 (and perhaps sub $1000) external recorders

Personally, I think Jon wasn't using an optimal set-up. I doubt that the 5D MK2 or 5D MK3 has a lower noise-floor than a Tascam DR-100 MK2, which I know from testing has a RMS noise-floor of -83 dB, which is the best I've tested for recorders under $500. The newer 5D MK3 might be as good as my GH2 which has a RMS noise-floor of -63 dB, but I have yet to test it to see what the actual RMS noise-floor is.

See this chart to see the noise levels of many common external recorders: http://www.avisoft.com/recordertests.htm

I am not a fan of the Avisoft noise-floor testing, as their numbers don't correspond well with RMS noise-floor levels, and I have found big discrepancies between their numbers and what I am getting with the same recorders. Under certain conditions you can actually hear the high noise-floor hiss ( when it's in the -50 to -59 dB range ), which was one reason why I ditched my Canon 60D. ( I could hear the high noise-floor when I shot with it )

I am in the process of publishing an 8 recorder comparison for my first blog post, but work keeps getting in the way. The technical tests are finished, but I haven't had time to arrange a live musical recording test to see how good the built-in mics are.

The USBPre2 has the same hardware topology as the 722 (top of the chart)

Yes, Sound Devices gear is definitely among the best out there. I've been using a MixPre field mixer for the past 4 years, and recently upgraded to a MixPre-D which is a nice step up from the old unit.

I also have the DR100mkII- its AD converters are very noisy compared to the SD ADs.

In my own tests, the DR-100 MK2 shows a RMS noise floor of -83 dB for LINE level and MIC level as long as you stay away from the HIGH pre-amp MIC setting. ( this setting kind of ruins the audio capability of this recorder )

I know that the Sound Devices recorders are better than -100 dB for RMS noise floor levels, but for practical purposes most people won't notice any noise once you've crossed the -70 dB threshold in your finished audio tracks.

In terms of noise performance and cost, the Fostex FR-2LE is the best deal.

I haven't had chance to actually test this unit. I've heard that it's pretty good, but I have no idea how it actually stacks up. Some manufacturers fib a bit in their specs, so I trust my own real-world tests more than anything I read printed.

For example, the Sony PCM-M10 is a great recorder, as long as you stay away from the upper end of it's mic pre-amps. The M10 pre-amps are very noisy once you cross the mid-range of it's gain setting. Where the now discontinued Olympus LS-11 recorder, outperforms the Sony quite a bit, holding to a -75 dB RMS noise-floor throughout it's gain range. ( Olympus had no idea how to market this recorder, so sales weren't great and they dropped it )

[jcs]

The trick with the 5D2 and 5D3 is to manually set gain to +1 (lowest level), then run a high-quality preamp (SD or JL) to hit -12dB while recording. The Sennheiser G3 also works very well in this way- noise level is not an issue. If you get access to a 5D3, you might want to test with the SD MixPre-D in this way.