BMCC with Sound Devices MixPre Tests

The place for questions about shooting with Blackmagic Cameras.
  • Author
  • Message
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 7:50 am

Hi Everyone,

As requested, I've done some very quick tests with a Sound Devices MixPre-D connected to a Blackmagic Cinema Camera EF.

For all of these tests I had a MixPre-D mounted underneath the BMCC, and connected via custom made 2 x Female XLR to 2 x 6.35mm TRS Balanced Jack with a Unbalanced 3.5mm Mini-Jack to 3.5mm Mini-Jack return feed.

Here are some photos:

Image
Image

Connected to the MixPre-D was a Sennheiser ME 66 on the left channel and a Rode NT3 on the right channel, with both obtaining Phantom Power from the MixPre-D. I picked these microphones because they are common place with low-budget productions - and also because they can be self-powered if required. I've done tests previously with these microphones connected directly to the camera - so I don't see the point of doing this again.

To work out the gain structure, I connected the BMCC to a MacBook Pro laptop, and use UltraScope to adjust the levels so that the 1 kHz tone oscillator from the MixPre roughy matches -20dB in Audition. For tests done with the 1.2.1 firmware, I removed the DC Offset. The Limiter on the MixPre-D was set to OFF for all of these tests.

All of the below dialogue tests were done with the exact same microphone setup, and the exact same gain settings on the MixPre-D. For the dialogue tests, I recorded RAW on the camera, then brought the individual WAV files into Adobe Audition for trimming. No processing or adjustments were applied with the exception of the DC Offset, so you're basically hearing exactly what the camera recorded.

For the purposes of comparison, here is the same lines of dialogue recorded on a Sound Devices 552 using the same microphone setup.

Here are the results:

Test One
Camera Firmware: 1.2.1
Camera Input Level: Line 80%
MixPre Output Switch: Line
UltraScope: Screenshot
Media Express: Screenshot
Adobe Audition: Screenshot
Dialogue Test: Download

Test Two
Camera Firmware: 1.2.1
Camera Input Level: Mic 12%
MixPre Output Switch: Mic
UltraScope: Screenshot
Media Express: Screenshot
Adobe Audition: Screenshot
Dialogue Test: Download

Test Three
Camera Firmware: 1.5.1
Camera Input Level: Line 100%
MixPre Output Switch: Line
UltraScope: Screenshot
Media Express: Screenshot
Adobe Audition: Screenshot
Dialogue Test: Download

Test Four
Camera Firmware: 1.5.1
Camera Input Level: Mic 85%
MixPre Output Switch: Mic
UltraScope: Screenshot
Media Express: Screenshot
Adobe Audition: Screenshot
Dialogue Test: Download

The MixPre-D also has a Unbalanced TA3-M Stereo Mic Output (-36 dBu), which I did a quick test with:

Test Five
Camera Firmware: 1.2.1
Camera Input Level: Mic 7%
UltraScope: Screenshot
Media Express: Screenshot
Adobe Audition: Screenshot
Dialogue Test: Download

Test Six
Camera Firmware: 1.5.1
Camera Input Level: Mic 55%
UltraScope: Screenshot
Media Express: Screenshot
Adobe Audition: Screenshot
Dialogue Test: Download

The MixPre-D also has a Unbalanced 3.5mm Tape Output (−10 dBV), however, I don't really see the point of using this consumer output with the BMCC.

Hopefully the above tests and information are helpful to people. The MixPre-D is an absolutely awesome product - and has fantastic sound quality. Attaching it to the BMCC however makes the camera pretty heavy and bulky - so I'm not sure I'd recommend this product for solo run-and-gun productions.

Any questions let me know!

Best Regards, Chris!
Offline
User avatar

adamroberts

  • Posts: 4500
  • Joined: Wed Aug 22, 2012 5:27 am
  • Location: England, UK

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 8:35 am

Thanks Chris

Nice little test. Need to relisten to the sample on the studio monitors later.

What are you thoughts on the sound under 1.5.1?

Would have been nice to have some silent parts in the recordings to compare the noise floor.
Offline
User avatar

MaxEmanuel

  • Posts: 82
  • Joined: Wed Aug 22, 2012 4:12 pm
  • Location: Bavaria / Germany

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 9:13 am

Hey Chris, great to see it working with your SetUp. Meanwhile I've received my MixPre-D back from service and now it is working too. However, I'll stick with the BMC388 from Juiced Link cause in my opinion is better matching to the BMCC.
Max Biber // BIBERFILM // Germany
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 9:45 am

adamroberts wrote:What are you thoughts on the sound under 1.5.1?


I don't think anything actually changed between 1.3 and 1.5.1 unfortunately. We've been using 1.5 for all our BMCC work - and just fixing the EQ in post.

adamroberts wrote:Would have been nice to have some silent parts in the recordings to compare the noise floor.


Yeah, I should have in retrospect. There's a tiny little bit of silence between sentences that might be of help.

maxemanuel wrote:However, I'll stick with the BMC388 from Juiced Link cause in my opinion is better matching to the BMCC.


If you already own the MixPre-D I would DEFINITELY use that over the BMC388. The MixPre has much better pre-amps, "unclippable" input peak limiters, high-pass filters (80 Hz or 160 Hz), sunlight-viewable, 16-segment GaN LED output meter with adjustable brightness and it's built like a tank. The only time I'd pick the BMC388 over the MixPre-D is when you need to go hand-held and keep the rig as light and small as possible. The MixPre-D is definitely a better unit - but it's more expensive and much bulkier.
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 10:14 am

Since my first sound devices mixer in 2001 I have been using -10 line level out to my lectro camera hops. I know people call it consumer but it is a misguided term.
Nothing is wrong with -10 and unless you have ground loops or RF interference - unbalanced is perfectly ok.
So I don't see why you would not want to use -10.
I personally have never fed anything but LTC in to BMC so gain stage has never been an issue.
Last edited by Rado Stefanov on Thu Jan 02, 2014 10:22 am, edited 1 time in total.
Rado Stefanov
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 10:21 am

Recording sound in to bmc is like alchemy. You are trying to make gold out of iron.
But if you really have attach bad sound to you nice looking image always use line out of mixpred. You compromise a lot less with low grade line level analog front end and AD conversion then with low grade Mic pre-amps.
Rado Stefanov
Offline
User avatar

MaxEmanuel

  • Posts: 82
  • Joined: Wed Aug 22, 2012 4:12 pm
  • Location: Bavaria / Germany

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 10:21 am

Chris: You're right about quality and size but I did a test right now and in extreme quiet situations where you can hear any noise the BMC388 gives me the better quality - at least for me - over the MixPre-D. Could be that my MixPre-D is still not repaired probably but there is a tiny difference and so in that situation it's the 388th advantage.
Max Biber // BIBERFILM // Germany
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 10:24 am

maxemanuel wrote:Chris: You're right about quality and size but I did a test right now and in extreme quiet situations where you can hear any noise the BMC388 gives me the better quality - at least for me - over the MixPre-D. Could be that my MixPre-D is still not repaired probably but there is a tiny difference and so in that situation it's the 388th advantage.


Even though i stopped using sound devices gear 3 years ago I am sure they still do an amazing job supporting and servicing their gear...
Rado Stefanov
Offline
User avatar

MaxEmanuel

  • Posts: 82
  • Joined: Wed Aug 22, 2012 4:12 pm
  • Location: Bavaria / Germany

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 10:32 am

Well I'm sure about that but since I'm in Germany everything goes a long way. For example I've sent in a DOA BMCC about three weeks ago and they did not even check it up to now....
Max Biber // BIBERFILM // Germany
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 11:00 am

maxemanuel wrote:Chris: You're right about quality and size but I did a test right now and in extreme quiet situations where you can hear any noise the BMC388 gives me the better quality - at least for me - over the MixPre-D.


Interesting! Are you able to post some sample audio from the MixPre-D and BMC388 along with the settings you use so that we can do a comparison?

I find it VERY hard to believe that the BMC388 pre-amps would out-perform the MixPre-D, so I'm assuming the reason you're hearing such a big difference is because of the different camera settings you're using for each device. The BMC is a strange beast as there is different frequency attenuation happening between Mic vs Line, and the attenuation is also affected by the gain setting you use. The 1.2 firmware also has the added confusion of the DC Offset.

Rado wrote:Since my first sound devices mixer in 2001 I have been using -10 line level out to my lectro camera hops. I know people call it consumer but it is a misguided term.
Nothing is wrong with -10 and unless you have ground loops or RF interference - unbalanced is perfectly ok.
So I don't see why you would not want to use -10.
I personally have never fed anything but LTC in to BMC so gain stage has never been an issue.


Nothing is wrong with the Unbalanced 3.5mm Tape Output (−10 dBV) output from the MixPre-D - it's electrically isolated from the XLR outputs and is very useful to have, especially for hooking up Comtek's and other consumer-level devices. However, when you have the option of balanced XLRs and unbalanced TA3's (both of which are professional LOCKING connectors), I really don't see the advantage of using un-lockable mini-jack's. Do they have a use - of course! Do they have a use with the BMC - not really, unless you're already using the XLR's and TA3's for something else (i.e. a three camera shoot, and a single MixPre-D).

When I say CONSUMER, I don't mean it in a bad way - but the fact remains that −10 dBV is for CONSUMER devices and +4 dBu is for PROFESSIONAL devices. When you're trying to interface a professional piece of gear to another professional piece of gear, keeping everything at +4 dBu means that all the equipment is working at their designed "sweet spot", maintaining the optimum noise floor. Mathematically, connecting a professional device to another professional device at +4 dBu will be much better than connecting a −10 dBV signal to something that's expecting something a bit "hotter". In practise however - there are so many other factors to take into consideration, that in the scheme of things it probably doesn't matter which one you use, as I doubt anyone will be able to actually HEAR the difference if you did a side-by-side comparison. That said... from a practical point, the locking XLRs and TA3's have the edge. Balanced XLRs would obviously be preferred as they reduce susceptibility to external noise.
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 8:29 pm

Chris Hocking wrote:
maxemanuel wrote:Chris: You're right about quality and size but I did a test right now and in extreme quiet situations where you can hear any noise the BMC388 gives me the better quality - at least for me - over the MixPre-D.


Interesting! Are you able to post some sample audio from the MixPre-D and BMC388 along with the settings you use so that we can do a comparison?

I find it VERY hard to believe that the BMC388 pre-amps would out-perform the MixPre-D, so I'm assuming the reason you're hearing such a big difference is because of the different camera settings you're using for each device. The BMC is a strange beast as there is different frequency attenuation happening between Mic vs Line, and the attenuation is also affected by the gain setting you use. The 1.2 firmware also has the added confusion of the DC Offset.

Rado wrote:Since my first sound devices mixer in 2001 I have been using -10 line level out to my lectro camera hops. I know people call it consumer but it is a misguided term.
Nothing is wrong with -10 and unless you have ground loops or RF interference - unbalanced is perfectly ok.
So I don't see why you would not want to use -10.
I personally have never fed anything but LTC in to BMC so gain stage has never been an issue.


Nothing is wrong with the Unbalanced 3.5mm Tape Output (−10 dBV) output from the MixPre-D - it's electrically isolated from the XLR outputs and is very useful to have, especially for hooking up Comtek's and other consumer-level devices. However, when you have the option of balanced XLRs and unbalanced TA3's (both of which are professional LOCKING connectors), I really don't see the advantage of using un-lockable mini-jack's. Do they have a use - of course! Do they have a use with the BMC - not really, unless you're already using the XLR's and TA3's for something else (i.e. a three camera shoot, and a single MixPre-D).

When I say CONSUMER, I don't mean it in a bad way - but the fact remains that −10 dBV is for CONSUMER devices and +4 dBu is for PROFESSIONAL devices. When you're trying to interface a professional piece of gear to another professional piece of gear, keeping everything at +4 dBu means that all the equipment is working at their designed "sweet spot", maintaining the optimum noise floor. Mathematically, connecting a professional device to another professional device at +4 dBu will be much better than connecting a −10 dBV signal to something that's expecting something a bit "hotter". In practise however - there are so many other factors to take into consideration, that in the scheme of things it probably doesn't matter which one you use, as I doubt anyone will be able to actually HEAR the difference if you did a side-by-side comparison. That said... from a practical point, the locking XLRs and TA3's have the edge. Balanced XLRs would obviously be preferred as they reduce susceptibility to external noise.


The -10db and +4db are approximate nominal voltage levels. they are guidelines for pick amplitude.
Mixpred Line out Clip Level is +22 dBu.
In my studio I use +19 on all my RME-Audio gear.
On my zaxcom gear I again use -10db going to Comtek TX and some cameras and +4 to other cameras.
Gain staging is something you learn with time and not something you read from manual specifications.
Especially coming from camera manufacturers....

what do you mean "the locking XLRs and TA3's have the edge"?
Rado Stefanov
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 8:59 pm

Rado wrote:what do you mean "the locking XLRs and TA3's have the edge"?


If I had the choice between a mini-jack connection or a XLR/TA3 connecting the MixPre to the BMCC - I would always pick the XLR/TA3 because the cables LOCK in place meaning I can't accidentally pull them out. The BMCC has so many wires hanging off it by the time you add on-board audio, monitoring, power, etc. that it would be easy to miss if a tiny mini-jack cable has accidentally been pulled out. Without audio meters, and the fact that the on-board headphone output is absolutely horrible (there's no way I would monitor the feed from the camera for more than a few minutes or else my head would explode from all the noise) - it would be easy to miss the fact that the camera isn't actually getting audio at all if you're not constantly double checking the return feed.

Having said that, if you're working in a studio, or have the camera on a tripod and not running around doco-style, then maybe the fact that you have a tiny mini-jack connected to the MixPre won't be an issue, as the risk of it getting pulled out accidentally is minimal.

As I said, −10 dBV connections definitely have their place, as does the 3.5mm mini-jack connector and there's nothing WRONG with using these connections for a HUGE range of different applications. However in THIS case, why would you bother connecting a −10 dBV output to a BMCC that's EXPECTING something "hotter"?

Rado wrote:The -10db and +4db are approximate nominal voltage levels. they are guidelines for pick amplitude.


Exactly! In electronics, nominal level is the signal strength at which a device is designed to operate for OPTIMAL PERFORMANCE. +4dBu puts out four times the level of –10dBV gear, since professional equipment may require higher voltage levels to carry a signal over longer cable runs without degrading. In PROPERLY designed equipment, a –10dBV system will provide EQUAL sound quality to a +4dBu system. HOWEVER, troubles occur when you mismatch settings on interconnected equipment. A +4dBu output will overload a –10dBV input and cause distortion, whereas a –10dBV output will not have enough signal strength to drive a +4dBu input.

Essentially if you plug a -10dBV output into a +4dBu input the signal is coming in 11.79dB quieter than the gear was designed for... so you need to turn something up. In all cases connecting a -10dBv output to the input of +4dBU gear will increase the noise floor.
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 10:17 pm

Chris Hocking wrote:
Rado wrote:what do you mean "the locking XLRs and TA3's have the edge"?


If I had the choice between a mini-jack connection or a XLR/TA3 connecting the MixPre to the BMCC - I would always pick the XLR/TA3 because the cables LOCK in place meaning I can't accidentally pull them out. The BMCC has so many wires hanging off it by the time you add on-board audio, monitoring, power, etc. that it would be easy to miss if a tiny mini-jack cable has accidentally been pulled out. Without audio meters, and the fact that the on-board headphone output is absolutely horrible (there's no way I would monitor the feed from the camera for more than a few minutes or else my head would explode from all the noise) - it would be easy to miss the fact that the camera isn't actually getting audio at all if you're not constantly double checking the return feed.

Having said that, if you're working in a studio, or have the camera on a tripod and not running around doco-style, then maybe the fact that you have a tiny mini-jack connected to the MixPre won't be an issue, as the risk of it getting pulled out accidentally is minimal.

As I said, −10 dBV connections definitely have their place, as does the 3.5mm mini-jack connector and there's nothing WRONG with using these connections for a HUGE range of different applications. However in THIS case, why would you bother connecting a −10 dBV output to a BMCC that's EXPECTING something "hotter"?

Rado wrote:The -10db and +4db are approximate nominal voltage levels. they are guidelines for pick amplitude.


Exactly! In electronics, nominal level is the signal strength at which a device is designed to operate for OPTIMAL PERFORMANCE. +4dBu puts out four times the level of –10dBV gear, since professional equipment may require higher voltage levels to carry a signal over longer cable runs without degrading. In PROPERLY designed equipment, a –10dBV system will provide EQUAL sound quality to a +4dBu system. HOWEVER, troubles occur when you mismatch settings on interconnected equipment. A +4dBu output will overload a –10dBV input and cause distortion, whereas a –10dBV output will not have enough signal strength to drive a +4dBu input.

Essentially if you plug a -10dBV output into a +4dBu input the signal is coming in 11.79dB quieter than the gear was designed for... so you need to turn something up. In all cases connecting a -10dBv output to the input of +4dBU gear will increase the noise floor.


With good cable management using 3.5 trs is not an issue. In fact that is what I use 99% of the time going to Alexa, Reds, Sonys.


You have a good grasp on theory but I feel like you are reading to many studio recording forums. In practice gain staging for production sound is a lot more complex.
Noise floor is not an issue unless you work on a sound stage.
And you can not be "pixel peeping" when it comes to sound considering you are using a medium range mixer going in to low grade camera sound recording.
Again you are trying to make gold out of iron.
Rado Stefanov
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 10:24 pm

And if I had to choose between XLR and 1/4 TRS I would always choose 1/4 TRS for the smaller size.
Rado Stefanov
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 10:42 pm

Rado wrote:And if I had to choose between XLR and 1/4 TRS I would always choose 1/4 TRS for the smaller size.


Fair enough. Have you actually tested a −10 dBV signal with the BMCC? What firmware are you using?
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 10:51 pm

Chris Hocking wrote:
Rado wrote:And if I had to choose between XLR and 1/4 TRS I would always choose 1/4 TRS for the smaller size.


Fair enough. Have you actually tested a −10 dBV signal with the BMCC? What firmware are you using?


I have never fed sound to BMC or BMPCC. No reason to waist my time. I only feed LTC from ERX2TCD and since LTC is square wave, I don't care what level.
If one day a client who uses BMC hires me I will send a scratch track. But then I will quickly measure on spot and provide the best possible scratch track.

The BMC and BMPCC are exactly what they are called: CINEMA CAMERAS.
Treat them as such.
Rado Stefanov
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 10:53 pm

BTW just listened to your samples in my studio. They sound horrible and exactly the reason why you should not record sound in to BMC BMPCC unless you are making a test video for Vimeo/Youtube.
Rado Stefanov
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 02, 2014 11:34 pm

Rado wrote:You have a good grasp on theory but I feel like you are reading to many studio recording forums.


I find it ironic that you wrongly guess I spent too much time on studio recording forums, when I'm trying to do real-world tests, and you're arguing with me over the use of a 3.5mm connection - when you're not even using the BMCC anyway!

Rado wrote:I have never fed sound to BMC or BMPCC. No reason to waist my time. I only feed LTC from ERX2TCD and since LTC is square wave, I don't care what level.


I agree - I always recommend recording LTC and a guide track to the camera, then syncing the "proper" audio in post. HOWEVER, there are jobs where for whatever reason (tight turnaround, budget, limited crew, etc.) you have no option but to record audio directly to the camera. That's the whole reason we have these discussions on this forum, and why we do these tests - not because we're trying to "make gold out of iron" and record the worlds best audio in-camera, but because due to the production requirements we HAVE to find a way to make on-camera audio recording work. If everyone was just able to send LTC to the camera, set and forget - then there wouldn't be an issue at all. But sadly, we don't live in this world.

Rado wrote:If one day a client who uses BMC hires me I will send a scratch track. But then I will quickly measure on spot and provide the best possible scratch track.


That's the point - you can't "quickly measure" - because the camera doesn't have any on-board metering. Unless you have a SDI monitor that has audio metering, UltraScope via TB is your only other option. There is also absolutely no way to determine what firmware the camera is running - so if you're just rocking up on set with no knowledge of the camera, then you have to guess as to whether it's running 1.2 or 1.3+. You can't accurately use the headphones as a guide, because the headphone output is incredibly noisy, and if you're running 1.2 the DC Offset throws you off.

Rado, it's obvious that you have a great knowledge of audio - and no doubt you've been doing this long enough to really master your craft. But just flat out saying that the on-board audio is useless, and there's no point in using it at all apart for TC isn't really helpful.

I'm not a sound recordist - I run a production company. For a lot of our jobs we hire incredibly talented freelance sound recordists that have invested huge amounts of money in gear. For those jobs, we get them to send the camera TC & a guide track, and sync it all up in post. Works wonderfully. But for other jobs, we might need to send a single camera operator to some remote part of the world where they're required to do both camera and sound. This might be because the job only has a small budget, or the budget might actually be quite large, and there are other factors that are forcing us to only use one person. Sometimes the scope of the production requires us to try and find ways to speed things up as much as possible - an recording audio directly to the camera helps in this case.

Rado wrote:BTW just listened to your samples in my studio. They sound horrible and exactly the reason why you should not record sound in to BMC BMPCC unless you are making a test video for Vimeo/Youtube.


Again, this is exactly why we do these tests - and this is exactly why I don't offer an opinion on these tests - you can make up your own mind as to what you think.

Filmmaking is all about compromises - regardless of what budget the production has and at what scale.

We've done plenty of jobs now with the BMCC with audio that's recorded directly to camera, that's been delivered for broadcast with no complaints. Many other people on this forum have the same experience. You CAN make the audio work if you do the research, and spend a bit of time and care in post.

Rado wrote:The BMC and BMPCC are exactly what they are called: CINEMA CAMERAS. Treat them as such.


We record audio directly to the Alexa and EPIC all the time without any issues. Again, it all depends on the scope of the job. Would we prefer to record audio separately on a 788T, Nomad, Cantar-X2, etc - of course we would! But audio is only a couple of line items in a budget - and there are so many other things to take into consideration.

The BMCC & BMPCC aren't just CINEMA CAMERAS - they're $1K and $2K cinema cameras. They're in a league of their own.

Rado, I have many good friends that are passionate and very talented audio gurus - so I know exactly where you stand, and I think your opinions are spot on and very valid. Sometimes however, you just need to get the job done, however possible.

My 2c.
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostFri Jan 03, 2014 12:00 am

Nobody records sound in Alexa or Red and use it in the final product. Nobody. Well maybe people using RED make that mistake.

You can measure how to give "good enough" level sound without clipping if you have a basic understanding of gain stage.

I never said the onboard audio is useless. You can use it for a scratch track, LTC and clap/wave matching sync.


You can not compromise sound. Bad sound is something that everybody notices. The way the human voice sounds - the intonation , the timbre , the level - is what triggers emotional response a lot more then picture.

People always blame clients when they sacrifice sound. Can you elaborate on the clients that hire you and your bmc and what exactly is your role?
Last edited by Rado Stefanov on Fri Jan 03, 2014 12:23 am, edited 1 time in total.
Rado Stefanov
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostFri Jan 03, 2014 12:08 am

You have to understand where I am coming from.
Even though my colleagues dismiss the BMC cameras I personally see the potential. The BM cameras produce an amazing picture but it has to be matched with great sound.
Also I work in broadcasting/TV. There are certain standards that we all follow.
Now days even reality TV shows have a good solid sound because they understood few years back how important sound is.

People get in to BMC because of the DR, color science, filmic look, Raw and I am sure they want to create something beautiful. I just don't want them to get lost in pixel peeping, lenses , Follow focus , rigs and miss that sound is 50% of the final product.
I apologies but It is ridicules that I am trying to hold you guys to the same standard as reality TV.
Rado Stefanov
Offline
User avatar

Peter J. DeCrescenzo

  • Posts: 2243
  • Joined: Wed Aug 22, 2012 6:53 am
  • Location: Portland, Oregon USA

Re: BMCC with Sound Devices MixPre Tests

PostFri Jan 03, 2014 12:31 am

Rado wrote:… You can not compromise sound. Bad sound is something that everybody notices. … People always blame clients when they sacrifice sound. ...


Clients are absolutely responsible for bad sound if they knowingly refuse to pay for good sound. Or good anything else for that matter.

It's OK for crew to deliver some reasonable percentage of quality above expectations, budget, and schedule, but beyond that is unsustainable.

I agree that audio (quality, music, effects, etc.) is at least as important as visuals (cam specs, composition, lighting, graphics, etc.). But I don't agree that "everybody notices" the quality of the audio. Sure, they notice if dialog is unintelligible, but beyond that? On a good day maybe.

If a production pays for a good script, direction, acting, sound, lighting, photography, operating, etc. they have every right to expect it and perhaps a bit more. However, if a production doesn't pay for high quality, it's the client's responsibility for the poor quality result.

-
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostFri Jan 03, 2014 12:50 am

Rado wrote:Nobody records sound in Alexa or Red and use it in the final product. Nobody. Well maybe people using RED make that mistake.


That's not true. Just last week we did a corporate job shot on Alexa with audio recorded directly to camera. We did have a backup on the 552 - but for final delivery, we used the audio directly from the camera. We've done many jobs in the past where the audio recorded directly to camera (both EPIC & Alexa) was used for final audio.

Rado wrote:Can you elaborate on the clients that hire you and your bmc and what exactly is your role?


We do all kinds of jobs - TVCs, broadcast, theatrical, branded content, additional DVD/BluRay content, etc. I normally produce - but for smaller jobs, where we can only bring a two or three person crew I'll also jump on sound. Previously I worked as a Post Production Supervisor and many years ago now I worked as a Lighting Designer in Live Productions.

Rado wrote:You have to understand where I am coming from.


I do. Good sound is absolutely vital. I completely understand it's importance, and also the important of a highly skilled and experienced sound recordist. I get it. We work regularly with some awesome sound recordists, and incredible sound studios that do amazing work.

Regardless of whether we're using a freelance professional, or doing a job entirely in-house using our own gear and crew, we never deliver "bad sound". Yes, if you're an audio professional with a trained ear, you might be able to tell the subtle difference between the jobs we've done where we've recorded straight to camera, and jobs we've done when everything has been using an external mixer - but most people, including the broadcasters doing the QC tests won't.

The world needs people like you that stick up for the craft of quality audio. But at the same time, testing this new and innovative gear like the BMCC and BMPCC is incredibly important, not just for camera operators and cinematographers out there - but also for the freelance sound recordists who don't have the time, knowledge or resources to go out and test every camera on the market.

I'll shut up now.
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostFri Jan 03, 2014 1:29 am

While I agree that Alexa sounds better then all other cameras I must say you should have used the 552 files.
BTW with 552 you can stamp external time code from Alexa. While 552 does not have TC generator it is still a time code recorder.
Rado Stefanov
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostFri Jan 03, 2014 1:31 am

And you are underestimating your clients. Sometimes they even notice the difference between good sound and great sound.
Rado Stefanov
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostFri Jan 03, 2014 1:55 am

Rado wrote:While I agree that Alexa sounds better then all other cameras I must say you should have used the 552 files.


Unfortunately in these cases, timeframe and workflow prevents us from being able to go back to the 552 files - we keep them on archive, but we don't actually use them for final delivery. If we could we would (and yes, I'm aware that you can stamp the 552 files with timecode). Of course there are other plenty of jobs we do where we'll definitely use the 552 (or 744T or 788T or Cantar or Deva) and timecode sync the recording device to the camera(s) or vice versa - but for most of the work we do with the BMCC, it's either low budget, or low-profile run and gun.

If this was an Alexa forum or REDuser we would be having a very different discussion. Remember, this forum is for a line of cameras that are very inexpensive and fairly unique in their strengths and weaknesses.

To my ear however, I can BARELY notice the difference between the 552 and the Alexa - the Alexa has wonderful audio recording functionality. The EPIC... not so much, but it's useable. Again, as always, it's a matter of balancing compromises.

Rado wrote:And you are underestimating your clients. Sometimes they even notice the difference between good sound and great sound.


Our clients are fantastic and extremely intelligent, and they trust us to make these judgement calls based on our knowledge and experience. Depending on budget, timeframe and workflow we will use the best tool for the job. Audio is important, but so is camera, lighting, grip, production design, costumes, makeup, and so many other factors that all affect the overall production value. Just like having really good catering can actually really have an effect on how the crew perform (and as a result, have an impact on the actual final product) - just because you're using the very best audio gear available, won't necessarily help the end product.

I'm not disagreeing with you Rado. Audio is really, really important, and you should always aim for the best. But if you're not willing to compromise when required, then some really amazing and important stories, might not be able to be told.

The whole point of these forums is to discuss and explore the technology that Blackmagic produces and sells. We invested in some of Blackmagic's cameras because they fill a very small need for SOME of the work we do. Because we have made this commitment, we want to make sure we are completely across the technology and understand it's advantages and limitations. If we don't push the limits of what the hardware can do, and offer Blackmagic ideas and advice on how the hardware and software can be improved, then that's bad for everyone.
Offline
User avatar

adamroberts

  • Posts: 4500
  • Joined: Wed Aug 22, 2012 5:27 am
  • Location: England, UK

Re: BMCC with Sound Devices MixPre Tests

PostFri Jan 03, 2014 8:56 am

These debates will always be heated when people who are passionate about their craft are involved. And rightly so.

I totally agree that sound is important but I'm also with Chris on this one. If the budget is limited then there are compromises that are made at every level. I'd love to shoot on the Alexa for every job and have a dedicated sound crew but then I'd be neglecting many many clients needs who have far smaller budgets.

While we have the ear of a few passionate sound people... Here is a question that will no doubt bring more debate...

Considering sound is 50% of the product. Considering the low price point of the BM cameras and lets for a moment exclude lenses, mics, lighting, grip, etc...

Many are looking at this thread to improve the quality of audio recorded to the BMCC (or even BMPCC for that matter) and in most cases adding a SD MixPreD or JuiceLink is an option.

Looking at other options, what would be a good budget recorder for the BMCC that would improve on the quality of sound over a preamp and in camera but still fit within the small budget?
Offline
User avatar

adamroberts

  • Posts: 4500
  • Joined: Wed Aug 22, 2012 5:27 am
  • Location: England, UK

Re: BMCC with Sound Devices MixPre Tests

PostFri Jan 03, 2014 9:23 am

Geoff Baxter wrote:The Zoom H6 and the Tascam DR60D come to mind, the specs on both are quite good. And both are in the $300 category.


Agreed. Those 2 are often mentioned. I'm interested in what the sound guys think tho...

Lets look at it as a 50/50 split.

If we look at the BMPCC being $1000 - what $1000 field recorder with decent pre-amps and monitoring would you choose?

If we look at the BMCC being $2000 - what $2000 field recorder with decent pre-amps and monitoring would you choose?

Or even simple adding a recorder to the MixPreD.
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostFri Jan 03, 2014 9:34 am

I would go with the TASCAM DR-60D if I was just starting.
For $2000 I would get an used 702t or 552. But still a better value bang for your buck will be Zaxcom maxx.
For $2995 I would get the zaxcom maxx. It is the top of production sound engineering and you can use it for the next 100 years without upgrading.
Rado Stefanov
Offline

Alexander Arndt

  • Posts: 160
  • Joined: Sun Nov 04, 2012 7:33 pm
  • Location: Germany

Re: BMCC with Sound Devices MixPre Tests

PostSat Jan 04, 2014 2:07 am

hey Chris Hocking!

nice little test test test you got going on here. in fact for people like me i ill guess its kind of a guidance threw this hole jungle technical stuff thing, or you just spend more time with it, nevermind, indeed
if you want a descent sound out of the black magic, which one would you prefer ...just for point of view to get usuable sound out of the camera ...and also to be adjusted later on if necessary...or a must?

or is it even better to switch to a portable recorder...but this means that when you shoot alone in a documentary style its kind of tricky or nearly impossible to adjust both at the same time.

thanks-
Offline

Christopher Barry

  • Posts: 65
  • Joined: Mon Jan 28, 2013 2:56 am
  • Location: Australia

Re: BMCC with Sound Devices MixPre Tests

PostSun Jan 05, 2014 6:48 pm

Chris Hocking, thank you for conducting the test and providing the sample audio files.

After listening to all the recordings, without scientific analysis of my own, it sounds like the audio recordings generally reflect Robert/juicedLink findings who prepared his own analysis between firmware v1.2.1 and 1.3. The graphs in the following video show v1.2.1 had better frequency response, especially the subs, yet suffered a DC offset problem, which became worse with more attenuation of the BMCC input level (Line and/or Mic?). While DC offset can be corrected, the offset from centre reduces signal potential before clipping. Whereas v1.3 solved the DC offset problem, then introduced a roll off of the subs, like a High Pass filter. While EQ can add back the subs in post, sonically it is unlikely to technically be as good, and may introduce increased noise floor as a result of the post EQ boost, etc. It was suggested by Robert that perhaps the subs may haven been cut out to reduce the sound of the BMCC fan noise when using the internal BMCC microphone, which makes no sense to affect the recording of professional sound being input into the BMCC. Who knows why BMD rolled off the subs in v1.3+?



Chris, the sound recordings may generally match the chart analysis in the above video. The following is my generalised opinion, after a quick listen through headphones on a laptop:
* v1.2.1 Line at 80% sounded the most full compared to all the other recordings, if such a description makes any sense to anyone else.
* v1.5.1 Line at 100% was probably the second best, yet it lacked the sub frequencies of the aforementioned v1.2.1 recording. Not sure if the noise floor became a problem in this recording, even though I did sample the break in the sentence, the silence for a moment, I was not clearly sure of the noise floor result in comparison to other recordings.
* the balance of audio recordings that were input into the BMCC via Mic input setting, they sounded thinner and more 'brittle', generally. Although some where better than others.

Chris, what I did notice, the v1.2.1 audio clip did not suffer DC offset issues. I was expecting to observe this problem. Did you correct the DC offset for this file?

Not sure if I would revert back to v1.2.1 or run v1.5+ on a BMCC. The answer to my DC offset question is one factor to consider. Also, I am interested to test the MixPre-D Tape Out (-10dB) to BMCC Mic input. I do not think it will help with v1.5+, it will require Mic input on the BMCC, and the graphs in the video appear to show the Mic input suffers more roll off of the subs, compared the Line input on the BMCC, if memory serves.

Views may vary, and I may be wrong with some observations. I will listen again to the recordings later this week. Thanks, Chris.
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostSun Jan 05, 2014 10:25 pm

Thanks for your detailed comments Christopher!

Christopher Barry wrote:Who knows why BMD rolled off the subs in v1.3+?


I believe it's simply a bug/mistake.

Christopher Barry wrote:Chris, what I did notice, the v1.2.1 audio clip did not suffer DC offset issues. I was expecting to observe this problem. Did you correct the DC offset for this file?


Yes, for all the 1.2.1 clips, I removed the DC Offset in Adobe Audition for easier comparison.

Christopher Barry wrote:Not sure if I would revert back to v1.2.1 or run v1.5+ on a BMCC.


Originally when 1.3 first came out I did some quick initial tests, and decide to stick with 1.2 - however, after further testing and playing, I determined that the frequency attenuation was non-destructive (as far as I could tell without opening up the camera and analysing the hardware) - so I decided that the 1.3+ gave us the best results after post production.

Christopher Barry wrote:Also, I am interested to test the MixPre-D Tape Out (-10dB) to BMCC Mic input. I do not think it will help with v1.5+, it will require Mic input on the BMCC, and the graphs in the video appear to show the Mic input suffers more roll off of the subs, compared the Line input on the BMCC, if memory serves.


I just did some very quick tests with the -10dB Tape Out from the MixPre-D. As I predicted - it's not terribly useful. On the 1.2.1 firmware this connection is completely unusable, as the signal is too hot when set to Mic, causing the camera to automatically switch to line (and then switch off audio on the camera completely after a few seconds requiring a restart), and too low when set to line. I could use some in-line pads to solve the Mic issue, but that defeats the whole purpose of the small mini-jack connector (i.e. if we were only using the mini-jack connector to reduce the size of the kit).

On the 1.5.1 firmware the signal is still too low for line level - however you can get it working on Mic at 7%. Given the
of the camera is around the 80% mark on this firmware, if you do decide to go down this route, I'd suggest using in-line pads to attenuate the signal so that you can get it into the 80% range.

Test Seven
Camera Firmware: 1.5.1
Camera Input Level: Mic 7%
UltraScope: Screenshot
Media Express: Screenshot
Adobe Audition: Screenshot
Dialogue Test: Download

Rado wrote:You compromise a lot less with low grade line level analog front end and AD conversion then with low grade Mic pre-amps.


Although this makes sense in theory - there are cases where you'd actually pick Mic over Line. For example the BMC388 uses a amplified mic level output - not line level (explained in more detail here).

Alexander wrote:if you want a descent sound out of the black magic, which one would you prefer ...just for point of view to get usuable sound out of the camera ...and also to be adjusted later on if necessary...or a must? or is it even better to switch to a portable recorder...but this means that when you shoot alone in a documentary style its kind of tricky or nearly impossible to adjust both at the same time.


If you want the BEST quality results - hire someone like Rado who obviously seriously cares about his craft and has access to good gear. Use the 1.3+ firmware and record a guide track to the Left Channel of the camera and timecode to the Right Channel.

If however the job doesn't allow for an audio recordist, and you need to do everything yourself, then the next best option would be to use an external recorder (the best you can afford), but also record a guide track to the camera for easy syncing.

If however, the job doesn't allow for a sound recordist or the ability to use an external recorder - then use the best pre-amp you can afford and record straight to camera. Regardless of whether you use 1.2 or 1.3+ you'll still need to do processing in post - it's just a matter of deciding whether you fix the DC Offset, or the Frequency Attenuation issue. Personally - I've been using the 1.5.1 firmware for jobs that require us to record directly to camera.

adamroberts wrote:I totally agree that sound is important but I'm also with Chris on this one. If the budget is limited then there are compromises that are made at every level. I'd love to shoot on the Alexa for every job and have a dedicated sound crew but then I'd be neglecting many many clients needs who have far smaller budgets.


The really important thing to note here is that it's not always about budget. People often blame lack of funds for these decisions - but a lot of the time, there are so many other factors that go into these decisions - such as transportation limitations, crew limitations, space limitations, client restrictions, etc. It's not always about the dollars.

adamroberts wrote:Would have been nice to have some silent parts in the recordings to compare the noise floor.


I did some more VERY QUICK tests. I connected a Rode NT3 to the MixPre-D, with the MixPre-D supplying the phantom power. I then recoded RAW on the camera, with a few seconds of tone coming off the MixPre-D, then "silence" from the NT3 (although I recorded it in the office, so there is a fair amount of background noise unfortunately - planes overhead, fridge, etc.), then I turned the volume completely down on the MixPre-D and recorded a few seconds of that. Here are the results:

- XLR out of the MixPre-D, Line 80% on the BMCC running 1.2.1
- XLR out of the MixPre-D, Mic 12% on the BMCC running 1.2.1
- XLR out of the MixPre-D, Line 100% on the BMCC running 1.5.1
- XLR out of the MixPre-D, Mic 85% on the BMCC running 1.5.1
- Tape Out of the MixPre-D, Mic 7% on the BMCC running 1.5.1
Offline

Loch Phillipps

  • Posts: 45
  • Joined: Wed Oct 31, 2012 8:32 pm

Re: BMCC with Sound Devices MixPre Tests

PostMon Jan 06, 2014 12:35 pm

Chris, I really appreciate all the work you're doing to help us all deal with BMCC sound. I'm a run and gun shooter and not terribly smart with sound. I wonder if you could give me your take on which settings to use with the MixPre-D with 1.5.1.

And you also say the MixPre is pretty heavy for us run and gun guys and it's true that my rig weighs too much already. Is there another mixer and/or preamp that you'd recommend? I currently strap a DR-100 on top of the battery at the back of my rig and record on that. There are some aspects of it I really don't like: another switch to hit to start/stop recording, the fact that it is not clear through the headphones when its battery is shot, that battery life on it is short, etc. And my employers would really love it if they didn't have to pluralize everything I shoot.
Loch Phillipps
Offline

Christopher Barry

  • Posts: 65
  • Joined: Mon Jan 28, 2013 2:56 am
  • Location: Australia

Re: BMCC with Sound Devices MixPre Tests

PostMon Jan 06, 2014 1:02 pm

Chris Hocking, you've been busy. Thanks for the reply, explanation and for posting the latest WAV files.

If the resulting subs being rolled off in v1.3+ is a bug or a mistake, I am perplexed as to why BMD have not corrected this via firmware update yet. It has been over 6 months. Not complaining, it has not effected me, it will from now. I don't think I will roll back to firmware 1.2.1. Post fix is not a problem for me, and I do want to record audio into the BMCC for most of my work, if not all of it, if possible. As much as there is limited communication from BMD in this sub-forum, we know that it has been said that the BMCC has at least one more firmware update coming, in due course, so hoping audio is tweaked get it about as good as the hardware will allow. Thinking positive here.

Listening to your 'silent parts' recordings, v1.5.1 100% Line input sounds like it may have slightly less noise floor creeping in, while the v1.2.1 80% Line input picked up a lot of the ambient rumbles and subs, which it should, it has good frequency response, yet suffers DC offset. v1.5.1 is like have a High Pass filter on all the time, so you can probably not compare them so much. 80% input level vs 100% input level, the later may likely cause an increase in noise, a hiss? What I did see is that you did not apply DC offset correction on the two files recorded with firmware v1.2.1, and the DC offset was huge, known to occur more if you increase the level of the BMCC input. With such a big DC offset, you can see how narrow the range becomes under v1.2.1, and clipping will occur quickly, perhaps with low audio levels being feed into the BMCC with high BMCC input value added, in this case, 80%. 80% = ouch under v1.2.1, whereas 20-30% is said to be the optimal level for that firmware version. As you know, you can not always dictate the input level you desire, as opposed to that which is required.

BMD audio firmware request for next BMCC update, please provide:

* the audio frequency response of firmware from v1.2.1, ie, give us back the subs frequency, and don't bring back the DC offset that requires fixing in post, and limits the audio recording potential;

* correct (increase?) the input levels via the firmware circuit programming (Chris, is this correct, you had to set v1.5.1 at 100% Line input to get that audio signal level into the BMCC, whereas the v1.2.1 was set at 80% Line level, so what is the ideal Input level we should be applying through the BMCC menu, between 1-100? And will firmware changes to the audio circuit have an optimal range? I know it is difficult to speculate, I don't know of the audio hardware inside the BMC, and whether firmware changes should be made in digital or analogue circuits, etc.

Chris, thanks for your contribution to the BMCC audio threads. BMD?
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostMon Jan 06, 2014 11:54 pm

loch@offrampfilms.com wrote:I wonder if you could give me your take on which settings to use with the MixPre-D with 1.5.1.


If you HAVE to record audio to camera and you have an external preamp I would suggest Line Level using the 1.5.1 firmware.

loch@offrampfilms.com wrote:Is there another mixer and/or preamp that you'd recommend?


The BMC388 is pretty small and compact.

loch@offrampfilms.com wrote:There are some aspects of it I really don't like: another switch to hit to start/stop recording, the fact that it is not clear through the headphones when its battery is shot, that battery life on it is short, etc. And my employers would really love it if they didn't have to pluralize everything I shoot.


It's a tough decision. Using an external recorder will DEFINITELY get you better results. It's up to you whether you want to take the compromise or not.

Christopher Barry wrote:If the resulting subs being rolled off in v1.3+ is a bug or a mistake, I am perplexed as to why BMD have not corrected this via firmware update yet.


I don't think we'll ever find out the answer to that. All I know is BMD is very much aware of the issues, and have said they definitely want to try and address this problems at SOME STAGE.

We know that they can TECHNICALLY do it, as the 1.2. firmware didn't have the frequency attenuation issue, and they fixed the frequency attenuation issue on the 1.4 firmware for the Pocket Cinema Camera.

My only guess is that they either already have too many other things on the priority list, or they've run into some major issues with the BMCC hardware overall (limited memory?), which is forcing them to rethink some things - delaying all updates to the camera until these bigger issues can be resolved.

Christopher Barry wrote: What I did see is that you did not apply DC offset correction on the two files recorded with firmware v1.2.1, and the DC offset was huge, known to occur more if you increase the level of the BMCC input.


Yes, that's correct - I supplied the "noise files" without any processing what-so-ever.

Here are the files again with processing applied:

- XLR out of the MixPre-D, Line 80% on the BMCC running 1.2.1 (DC Offset Removed)
- XLR out of the MixPre-D, Mic 12% on the BMCC running 1.2.1 (DC Offset Removed)
- XLR out of the MixPre-D, Line 100% on the BMCC running 1.5.1 (EQ'ed)
- XLR out of the MixPre-D, Mic 85% on the BMCC running 1.5.1 (EQ'ed)

Christopher Barry wrote:With such a big DC offset, you can see how narrow the range becomes under v1.2.1, and clipping will occur quickly, perhaps with low audio levels being feed into the BMCC with high BMCC input value added, in this case, 80%. 80% = ouch under v1.2.1, whereas 20-30% is said to be the optimal level for that firmware version. As you know, you can not always dictate the input level you desire, as opposed to that which is required.


Yes, on the 1.2 firmware - you're much better off using Mic Levels than Line Levels - with the "sweet spot" between 20-30% gain level in the camera. HOWEVER, due to the fact that the camera "freaks out" when it receives a hot Mic level signal on 1.2 (i.e. it automatically switches to line), this firmware release is pretty useless for audio. If you're using an external pre-amp, I think I would steer clear of 1.2 and stick with 1.3+ (and just fix the frequency attenuation in post).

When we first started testing the 1.2 firmware we used a Sound Devices 302, set the camera to Mic Level 25%, and then applied -40dB XLR Attenuation using the 302 menu, which gave us the best performance.

Christopher Barry wrote:the audio frequency response of firmware from v1.2.1, ie, give us back the subs frequency, and don't bring back the DC offset that requires fixing in post, and limits the audio recording potential;


Even with the 1.2 firmware - it seems like there's still some frequency attenuation happening - either that or it's just bad hardware. Hopefully with future firmware they can really fine tune the audio so that it sounds a bit closer to the input it's receiving.

You can read through some more tests we did comparing firmware versions here.

Christopher Barry wrote:correct (increase?) the input levels via the firmware circuit programming (Chris, is this correct, you had to set v1.5.1 at 100% Line input to get that audio signal level into the BMCC, whereas the v1.2.1 was set at 80% Line level, so what is the ideal Input level we should be applying through the BMCC menu, between 1-100? And will firmware changes to the audio circuit have an optimal range? I know it is difficult to speculate, I don't know of the audio hardware inside the BMC, and whether firmware changes should be made in digital or analogue circuits, etc.


My suggestion would be, instead of having a 0-100% slider (which is basically useless unless you have a Thunderbolt enabled laptop hooked up to the camera or an external monitor with meters) - there is just four options for both left and right inputs:

1) Off
2) Internal
3) Line
4) Mic (-40 dB)

That way there's no confusion, and anyone that wants to use the camera's audio can just easily plug in an external preamp/mixer - and away you go.

Anyway... the whole purpose of this thread was to prove that the MixPre-D COULD in-fact work with the BMCC (following on from earlier conversations here) - so I think we've proven it definitely can.

Happy Filming Everybody!
Offline

Christopher Barry

  • Posts: 65
  • Joined: Mon Jan 28, 2013 2:56 am
  • Location: Australia

Re: BMCC with Sound Devices MixPre Tests

PostTue Jan 07, 2014 6:36 am

Chris, thank you for further clarification of matters discussed. It confirms my approach to record audio via MixPre-D into the BMCC, namely Line input under firmware v1.5.1 (v1.3+) pending the next firmware update to be released by BMD that will hopefully address our BMCC audio concerns addressed in this thread and numerous other threads in this forum.

In my last post, I forgot to mention that while we can compare the 'silent parts', based on different BMCC input configurations, and input levels from 1-100% value, what we may be missing is reference to what may be consider the result of standard professional audio recording equipment? To reference the silent recordings in comparison, would it make sense to compare the BMCC's results to, say, signal from the MixPre-D into your Sound Devices 552 and provide that WAV file for comparison? While I had to turn sound up to maximum to clearly hear the floor of the BMCC silent parts recordings through my headphones, indicating that perhaps the recording was not too bad for certain work or projects, my thoughts are to complete the comparison would be to include a recording of same scenario with dedicated professional audio equipment, replacing the BMCC in the audio chain with, say, the SD 552? It is not necessary to conduct such a test, just mentioning that I have no reference point to compare the BMCC audio results. Your comment and observations will suffice, I am not asking you to conduct yet another test. Thank you, Chris.
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostTue Jan 07, 2014 8:30 am

Christopher Barry wrote:In my last post, I forgot to mention that while we can compare the 'silent parts', based on different BMCC input configurations, and input levels from 1-100% value, what we may be missing is reference to what may be consider the result of standard professional audio recording equipment?


Good point - I didn't even think about taking the signal out of the MixPre-D and feeding it into another recorder. I no longer have the MixPre-D around unfortunately, but I'll upload a clip with the NT3 + 552 later this week as that will give you basically the same results.
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostTue Jan 07, 2014 10:04 am

Chris Hocking wrote:
Christopher Barry wrote:
I no longer have the MixPre-D around unfortunately, but I'll upload a clip with the NT3 + 552 later this week as that will give you basically the same results.


Not even close...
Rado Stefanov
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostTue Jan 07, 2014 10:25 am

Geoff Baxter wrote:I don't want to keep on about this, but I would just like to make the point that most pieces of professional audio equipment have published specifications, but the BM cameras do not, unless there is a source I am unaware of. And for that reason, I am reluctant to trust them with my audio.


Most camera manufactures have very limited information about their audio specs. They normally just state what type of connection the camera has, sampling rate, bit depth and codec.

Rado wrote:Not even close...


You really think the MixPre-D's preamps will sound completely different to a 552's with a $270 microphone connected? I have my doubts, but if I get a chance I'll do two tests, and you can guess which is which.
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostTue Jan 07, 2014 10:32 am

Chris Hocking wrote:
Geoff Baxter wrote:
Rado wrote:Not even close...


You really think the MixPre-D's preamps will sound completely different to a 552's with a $270 microphone connected? I have my doubts, but if I get a chance I'll do two tests, and you can guess which is which.

Instead of doubting me and making assumptions, listen to my real life experience.
I use 3 mkh8060 on regular basis. High sensitivity boom mics. Work fine on 552. On mix pre I can not go past 8 a clock without overloading.
Gain range and maximum input before distortion is very low on mixpreD.
Rado Stefanov
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostTue Jan 07, 2014 10:37 am

Rado wrote:Instead of doubting and making assumtions listen to my real life experience.
I use 3 mkh8060 on regular basis. High sensitivity mica. Work fine on 552. On mix pre I can not go past 8 a clock without overloading. Gain range and maximum input before distortion is very low on mixpreD.


Yes, but remember we're recording "silence" for the purposes of these tests. We're not exactly going to be pushing the pre-amps.
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostTue Jan 07, 2014 10:38 am

Very low for standard sound mixing equipment.
Rado Stefanov
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostTue Jan 07, 2014 10:40 am

Chris Hocking wrote:
Rado wrote:Instead of doubting and making assumtions listen to my real life experience.
I use 3 mkh8060 on regular basis. High sensitivity mica. Work fine on 552. On mix pre I can not go past 8 a clock without overloading. Gain range and maximum input before distortion is very low on mixpreD.


Yes, but remember we're recording "silence" for the purposes of these tests. We're not exactly going to be pushing the pre-amps.

If your mic clips at 8 a clock you have no control over gain stages. Mix pre is already crippled by having only one gain/level knob. Having 1 centimeter of control over your gain compared to normal operation with the 552 is a huge difference.
Last edited by Rado Stefanov on Tue Jan 07, 2014 10:44 am, edited 1 time in total.
Rado Stefanov
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostTue Jan 07, 2014 10:41 am

Chris you are way over your head.
Rado Stefanov
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostTue Jan 07, 2014 10:59 am

Rado wrote:If your mic clips at 8 a clock you have no control over gain stages. Mix pre is already crippled by having only one gang/level knob. Having 1 centimeter of control over your gain compared to normal operation with the 552 is a huge difference.


Yes, I'm very much aware that the MixPre-D has it's limitations, and obviously you're going to get much better results and control on your Nomad - but I still think this is completely irrelevant to the question at hand. It's all relative. If the BMCC is your recording device - then there's no point using pre-amps and microphones that cost hundreds of times more than the camera itself. Whether you should record audio onto the camera at all is a completely different discussion.

The whole purpose of this thread was to talk about the MixPre-D, and let's be honest, users that are connected a MixPre-D directly to their BMCC aren't exactly going to be using MKH 8040's and MKH 8060's. I could have done all these tests with the very best gear available, but if you're using a Nomad & 8040's - then you're definitely not going to be recording directly to camera, so there's no point using this level of gear for these tests. Originally I was using the CS3e for a lot of my tests - but because the NT3 and ME66 are so much more widely available, it just made more sense to use gear that the people browsing these forums know and actually use.

Rado wrote:Chris you are way over your head.


Haha... probably. That's why I hire professionals like your good self for jobs that require it. Like I said though, there are plenty of jobs where an audio recordist CAN'T be used, which is why we do these tests. (I'm actually sending a cinematographer with an Alexa to Indonesia tomorrow recording audio directly to camera - it happens).

If I ever end up in Nevada, I'll shoot you an message and we can catch up in person for a coffee/beer and chat about Zaxcom vs Sound Devices, and Sennheiser vs Schoeps.

I'll shut up now.
Offline

Christopher Barry

  • Posts: 65
  • Joined: Mon Jan 28, 2013 2:56 am
  • Location: Australia

Re: BMCC with Sound Devices MixPre Tests

PostTue Jan 07, 2014 3:31 pm

Chris, thanks again for all your efforts, testing and offering to provide that additional reference audio file, in due course. I appreciate that the BMCC audio may never be as good as professional audio equipment, typically part of a dual audio recording setup, however, there are also times when one may not always be able to engage a professional audio expert with experience and matching professional audio equipment for a number of reasons and/or recording situations. You have clearly made that point in previous posts.

I also appreciate that you are trying to bring to BMD's attention some of the anomalies observed with the BMCC recorded audio, noting the result of firmware changes from v1.2.1 through to the current v1.5.1. To my knowledge, no BMCC audio firmware updates from v1.3 has been reported in the release notes, which you have noted in your other BMCC related audio thread.

So while BMD reference the BMCC audio on their website literature as "professional", and many people may disagree, I understand the purpose of this thread is to bring some observations to BMD's attention, with a view/hope that the next firmware update will correct and make the audio recorded into the BMCC as good as it can practicably be.
Offline

Rado Stefanov

  • Posts: 169
  • Joined: Mon Aug 26, 2013 10:26 am

Re: BMCC with Sound Devices MixPre Tests

PostTue Jan 07, 2014 4:13 pm

Chris please do. Cheers.
Rado Stefanov
Offline

Alexander Arndt

  • Posts: 160
  • Joined: Sun Nov 04, 2012 7:33 pm
  • Location: Germany

Re: BMCC with Sound Devices MixPre Tests

PostThu Jan 09, 2014 2:04 pm

thanks for the replies, great info.

will get back to you when needed!

cheers
Offline
User avatar

Chris Hocking

  • Posts: 659
  • Joined: Wed Aug 22, 2012 4:23 am
  • Location: Melbourne, Australia

Re: BMCC with Sound Devices MixPre Tests

PostMon Jan 13, 2014 10:09 pm

This video may be of interest to some people:


Return to Cinematography

Who is online

Users browsing this forum: Brad Hurley, Robert Niessner and 21 guests