Canon 5D Mark II Audio Exposed; with BeachTek DXA-5D and juicedLink CX231

Tramm Hudson · June 20th, 2009, 10:03 PM

Quote:

Originally Posted by Jon Fairhurst

To summarize, the low-noise of the active juicedLink is excellent. I can turn it full bore and get better than 16-bit performance at most frequencies. It's a good match for the 5D2 with Magic Lantern firmware, though we might need to add some gain in the camera to capture low-level sounds or a dull mic. The BeachTek5D combos are noisy, but smooth. The MicroTrack II really needs an active preamp to mask its patterned noise.

That's great news and gives us some further direction for things to try. One easy change is to set the in-camera mic gain to the first step of +10 dB (MGAIN2-0=100). I'm really pleased to hear how successful your results have been in testing the new firmware!

Jon Fairhurst · June 20th, 2009, 10:26 PM

Quote:

Originally Posted by Tramm Hudson

That's great news and gives us some further direction for things to try. One easy change is to set the in-camera mic gain to the first step of +10 dB (MGAIN2-0=100).

I think that's totally viable. Right now, the noise is below the 16-bit threshold, yet, the 12 dB of digital gain is throwing away two bits. 10 dB of analog gain should be perfect. If we get a good balance, then the final firmware could control the digital gain only.

Quote:

I'm really pleased to hear how successful your results have been in testing the new firmware!

It's been challenging to do right, but seeing and hearing the results is exhilarating!

Jon Fairhurst · June 21st, 2009, 12:16 AM

Here's a 480x270 video that demonstrates my early results (right click to download):
http://p3pictures.com/audio_5d2/5DMa...rstResults.m4v You can even play it on your iPhone.

And here's a zipped 24-bit wave file for critical listening:
http://p3pictures.com/audio_5d2/5DMa...rstResults.zip

As mentioned earlier, I calibrated everything with the NT1-A and ran some noise tests. This evening I went outside and recorded some real-world stuff with the calibrated settings, using an Audio Technica AT815b long shotgun above my head. (This is one of the few times when I want the boom mic in the shot!)

In post, I added a 19.740 Hz notch to the BeachTek/5D2 pilot tone audio. I also felt that it was on the hot side, so I lowered the level by 3.5dB in post. That reduces its noise a bit. I also thought that the juicedLink/MagicLantern result might be a bit cold - I think the mic was a touch further from my mouth - so I boosted it by 2dB. I then boosted the whole project by 6dB, so we don't have to turn our amplifiers up quite so high to hear the details.

As expected, the juicedLink/Magic Lantern combo wins the day. The noise is wonderfully low.

BTW, I didn't get the MicroTrack II direct recording to work, due to operator error and failing light. Trust me that it's not much good without a preamp to give it a hand.

I'll work with Tramm to boost the gain in the Magic Lantern firmware. That will let me calibrate again with the BeachTek at full tilt, which will put it in its best light.

Anyway, download and watch this early result. Listen to the uncompressed audio. Let me know which combo you prefer - and if you think it is of pro quality.

Chris Barcellos · June 21st, 2009, 12:26 AM

Jon, you have linked to the same file twice, sure that was not intended. Both mv4, and no zip link.

I listened to the tests via the .mv4 files and it is clealy looking like we are going to be able to get pretty clean sound out of the Juiced Link Magic Lantern combo.

Thanks for the test results.

Jon Fairhurst · June 21st, 2009, 12:35 AM

Quote:

Originally Posted by Chris Barcellos

Jon, you have linked to the same file twice, was that intended. Both mv4.

Fixed. Thanks, Chris!

Jeremy Nicholl · June 21st, 2009, 02:35 AM

Quote:

Originally Posted by Jon Fairhurst

Here's a 480x270 video that demonstrates my early results
Anyway, download and watch this early result. Listen to the uncompressed audio. Let me know which combo you prefer - and if you think it is of pro quality.

Wonderful. To my non-pro ears the juicedlink/magic lantern combo was easily the best, and if I understand correctly it seems like you can get it even better. There was some noticeable rumble in the juicedlink/magic lantern section that started as you crumpled the paper: I assume that was caused by the breeze.

Thanks very much for the tests.

Tom Daigon · June 21st, 2009, 08:06 AM

And here's a zipped 24-bit wave file for critical listening:
http://p3pictures.com/audio_5d2/5DMa...rstResults.zip

Wow.....what a dramatic difference of noise levels in the tests.. The juiced link / ML sounds spectacular. As an editor I have no problem with the audio generated by this combo
of equipment. It exceeds my expectations of performance.

Matthew Roddy · June 21st, 2009, 05:22 PM

That was awesome! Great Real-World example. So simple, even I understood it, and What a difference!!!
I'm SO glad I waited before buying the Beachtek. You guys saved me a LOT of money. I owe ya!
Thank you for all your efforts!

Chris Barcellos · June 21st, 2009, 08:26 PM

Jon Fairhurst. Send Me PM so I send you information on monitoring.

Sean Seah · June 22nd, 2009, 09:31 AM

that is great work. I wished u could get a Zoom H4n for the test. I am wondering if it is possible to record the zoom onboard mics via the zoom and 5D2 directly. As of now, I understand most ppl are capturing independantly and syning in post (99.9%) but I am not sure how to do this in Sony Vegas yet. So the best way would be to go directly into the 5D2 for my one man ops.

Jon Fairhurst · June 22nd, 2009, 10:10 AM

A Zoom H4n would be awesome. If anybody has one and would be willing to ship it to me for a one day test, I'd be happy to pay for overnight shipping to get it back to them. I'd love to see how it compares with the JL/ML combo, as well as JL/Microtrack II. And I'd like to know if the Zoom works well enough alone, or if an active preamp is needed to provide a clean front end.

The challenge with not having it here is calibration. It's critical that the signal levels match for the comparison to make any sense. Not to mention that the noise tests are a PITA! I left my cell phone on during one run, and I had to start over. Trucks, planes and a bit of rain delayed things as well.

BTW, I need to re-do the "beer test" now that the Magic Lantern firmware has new gain levels. I should have something to post by mid-week...

Brian Valente · June 22nd, 2009, 05:24 PM

Jon

PM me and we'll send you our H4N for your tests.

Brian

Jon Fairhurst · June 22nd, 2009, 05:54 PM

Quote:

Originally Posted by Brian Valente

PM me and we'll send you our H4N for your tests.

Fantastic! PM sent!

Kamau Bilal · June 25th, 2009, 04:36 PM

So i checked out your video Jon, clearly the juicedlink is the best, i've not yet ordered the mark II just doing research. If I want the kind of setup you spoke of in your video where can i find everything i need?

thanks.

Brian Valente · June 25th, 2009, 04:44 PM

Kamau I suggest waiting until you see jon's latest results with H4N and his latest round of tests.

Brian

June 21st, 2009, 12:16 AM	#18
Jon Fairhurst Inner Circle Join Date: May 2006 Location: Camas, WA, USA Posts: 5,513	First results Here's a 480x270 video that demonstrates my early results (right click to download): http://p3pictures.com/audio_5d2/5DMa...rstResults.m4v You can even play it on your iPhone. And here's a zipped 24-bit wave file for critical listening: http://p3pictures.com/audio_5d2/5DMa...rstResults.zip As mentioned earlier, I calibrated everything with the NT1-A and ran some noise tests. This evening I went outside and recorded some real-world stuff with the calibrated settings, using an Audio Technica AT815b long shotgun above my head. (This is one of the few times when I want the boom mic in the shot!) In post, I added a 19.740 Hz notch to the BeachTek/5D2 pilot tone audio. I also felt that it was on the hot side, so I lowered the level by 3.5dB in post. That reduces its noise a bit. I also thought that the juicedLink/MagicLantern result might be a bit cold - I think the mic was a touch further from my mouth - so I boosted it by 2dB. I then boosted the whole project by 6dB, so we don't have to turn our amplifiers up quite so high to hear the details. As expected, the juicedLink/Magic Lantern combo wins the day. The noise is wonderfully low. BTW, I didn't get the MicroTrack II direct recording to work, due to operator error and failing light. Trust me that it's not much good without a preamp to give it a hand. I'll work with Tramm to boost the gain in the Magic Lantern firmware. That will let me calibrate again with the BeachTek at full tilt, which will put it in its best light. Anyway, download and watch this early result. Listen to the uncompressed audio. Let me know which combo you prefer - and if you think it is of pro quality. __________________ Jon Fairhurst

June 21st, 2009, 12:26 AM	#19
Chris Barcellos Inner Circle Join Date: Nov 2005 Location: Elk Grove CA Posts: 6,838	Jon, you have linked to the same file twice, sure that was not intended. Both mv4, and no zip link. I listened to the tests via the .mv4 files and it is clealy looking like we are going to be able to get pretty clean sound out of the Juiced Link Magic Lantern combo. Thanks for the test results. __________________ Chris J. Barcellos

June 21st, 2009, 08:26 PM	#24
Chris Barcellos Inner Circle Join Date: Nov 2005 Location: Elk Grove CA Posts: 6,838	Jon Fairhurst. Send Me PM so I send you information on monitoring. __________________ Chris J. Barcellos

June 22nd, 2009, 10:10 AM	#26
Jon Fairhurst Inner Circle Join Date: May 2006 Location: Camas, WA, USA Posts: 5,513	A Zoom H4n would be awesome. If anybody has one and would be willing to ship it to me for a one day test, I'd be happy to pay for overnight shipping to get it back to them. I'd love to see how it compares with the JL/ML combo, as well as JL/Microtrack II. And I'd like to know if the Zoom works well enough alone, or if an active preamp is needed to provide a clean front end. The challenge with not having it here is calibration. It's critical that the signal levels match for the comparison to make any sense. Not to mention that the noise tests are a PITA! I left my cell phone on during one run, and I had to start over. Trucks, planes and a bit of rain delayed things as well. BTW, I need to re-do the "beer test" now that the Magic Lantern firmware has new gain levels. I should have something to post by mid-week... __________________ Jon Fairhurst

June 22nd, 2009, 05:24 PM	#27
Brian Valente Major Player Join Date: Jan 2005 Location: Hollywood, CA Posts: 293	Jon PM me and we'll send you our H4N for your tests. Brian __________________ ========================== Brian Valente Redrock Microsystems

June 21st, 2009, 08:06 AM	#22
Tom Daigon Regular Crew Join Date: Jun 2009 Location: Las Vegas, NV Posts: 170	And here's a zipped 24-bit wave file for critical listening: http://p3pictures.com/audio_5d2/5DMa...rstResults.zip Wow.....what a dramatic difference of noise levels in the tests.. The juiced link / ML sounds spectacular. As an editor I have no problem with the audio generated by this combo of equipment. It exceeds my expectations of performance.

June 21st, 2009, 05:22 PM	#23
Matthew Roddy Major Player Join Date: Jan 2006 Location: Riverside, Ca Posts: 307	That was awesome! Great Real-World example. So simple, even I understood it, and What a difference!!! I'm SO glad I waited before buying the Beachtek. You guys saved me a LOT of money. I owe ya! Thank you for all your efforts!

June 22nd, 2009, 09:31 AM	#25
Sean Seah Trustee Join Date: Sep 2005 Location: Singapore Posts: 1,498	that is great work. I wished u could get a Zoom H4n for the test. I am wondering if it is possible to record the zoom onboard mics via the zoom and 5D2 directly. As of now, I understand most ppl are capturing independantly and syning in post (99.9%) but I am not sure how to do this in Sony Vegas yet. So the best way would be to go directly into the 5D2 for my one man ops.

June 25th, 2009, 04:36 PM	#29
Kamau Bilal Tourist Join Date: Jun 2009 Location: Columbia, MO Posts: 1	So what do i need to get? So i checked out your video Jon, clearly the juicedlink is the best, i've not yet ordered the mark II just doing research. If I want the kind of setup you spoke of in your video where can i find everything i need? thanks.

June 25th, 2009, 04:44 PM	#30
Brian Valente Major Player Join Date: Jan 2005 Location: Hollywood, CA Posts: 293	Kamau I suggest waiting until you see jon's latest results with H4N and his latest round of tests. Brian __________________ ========================== Brian Valente Redrock Microsystems