mr.incandenza

Super CMIT and wireless phase issues

46 posts in this topic

I'm curious to hear how many of the people who use SuperCMITs on dialogue actually use the processed digital audio (preset 1) mostly as opposed to only using it as a special tool for complicated noisy environments. IIRC Simon Hayes used the processed audio pretty much all the time on Les Miserables but I have also heard others say that the artifacts even on preset 1 are audible in many situations, which is why they tend to use the non processed audio (equivalent to an analog CMIT) most of the time.

Share this post


Link to post
Share on other sites

If I use the SCmit I always use the processed output (record both ch) in the mix. If I want the unprocessed sound I go for the Cmit5U. The SCmit noise floor is much more audible. 

The idea was to buy a SCmit and sell one of my 5U but unfortunately its not a 1:1 replacemement.

Share this post


Link to post
Share on other sites
If I use the SCmit I always use the processed output (record both ch) in the mix. If I want the unprocessed sound I go for the Cmit5U. The SCmit noise floor is much more audible. 
The idea was to buy a SCmit and sell one of my 5U but unfortunately its not a 1:1 replacemement.

That is 100% exactly the same for me
IIRC Simon Hayes used the processed audio pretty much all the time on Les Miserables

I seem to remember the opposite. Only unprocessed...

Share this post


Link to post
Share on other sites
4 hours ago, Matthias Richter said:

got word from Helmut directly:

the Super CMIT is perfectly suited for dialog recording in film. However the mic does have different delays for different frequencies that can not compensated for when mixing the SuperCMIT with lavs unless one is using a fir filter, something normally not used by the production sound mixer.

 

 

Screenshot_20161221-083332.jpg

 

Thanks Matthias, this was exactly what I was trying to say, but in appropriate words.

Share this post


Link to post
Share on other sites
1 hour ago, Constantin said:


That is 100% exactly the same for me
I seem to remember the opposite. Only unprocessed...

By processed I meant preset 1, channel 1 as opposed to unprocessed analog CMIT sound (channel 2). AFAIK both SuperCMIT presets on channel 1 are processed.

Share this post


Link to post
Share on other sites
By processed I meant preset 1, channel 1 as opposed to unprocessed analog CMIT sound (channel 2). AFAIK both SuperCMIT presets on channel 1 are processed.

Yes I know. I got the impression he would only use channel 2, unprocessed

Share this post


Link to post
Share on other sites
By processed I meant preset 1, channel 1 as opposed to unprocessed analog CMIT sound (channel 2). AFAIK both SuperCMIT presets on channel 1 are processed.

I dug up an old thread an it turns out you were right. And you should know it as Simon Hayes was responding to you!
He does only use the SuperCMIT on preset 1, and records channel 1, although he didn't mention if he records channel 2, too.

Share this post


Link to post
Share on other sites

I hope this my question is viewed in context, not having had the chance to demo a Supercmit, I'm not trying to doubt the mic or the company (consider me a fan of Schoeps mics, having used them for over 20 years, they are irreplacable in much of the work I do) but I don't understand the  fundamental approach that Schoeps is apparently using here. I'm far from an expert on these issues but to me it doesn't seem like a good idea to have a mic with processing that isn't phase-coherent across its entire spectrum. Isn't that the problem with using excessive amounts of EQ, phase shift? How hard would it be  to have a phase coherent output across the entire spectrum? I'd rather use a mic with a greater latency but phase coherent than a mic that messes with phase vs frequency. I know, if it sounds good that's all that matters, use your ears and so forth but I'd like to understand.

 

Share this post


Link to post
Share on other sites

here is an email from Helmut, I think this is interesting. He says about using a FIR filter to re allign the phase :

"Dear Angelo,
the SuperCMIT has a frequency-dependent latency on ch1. This means that it cannot be compensated by a simple delay but only by a special FIR filter. If you want to mix wireless mics and the SuperCMIT exactly in phase, you would have to use this FIR filter. However, a small movement of the actor already would cause a change of this delay. For more info, see http://digital.schoeps.de/documents/Schoeps_Manual_SuperCMIT-DE-111007.pdf

We designed this latency so that the minimum latency is short for high frequencies so that you can listen to a live mix on headphones with some live leakage.

In the experience of our users, this property is no disadvantage. Very seldomly, people ask, and I never heard of users that don't use the mic because of this.

Best regards,
Helmut"

do you guys have any idea how about to use this filter?

 

Share this post


Link to post
Share on other sites
here is an email from Helmut, I think this is interesting. He says about using a FIR filter to re allign the phase :
"Dear Angelo,
the SuperCMIT has a frequency-dependent latency on ch1. This means that it cannot be compensated by a simple delay but only by a special FIR filter. If you want to mix wireless mics and the SuperCMIT exactly in phase, you would have to use this FIR filter. However, a small movement of the actor already would cause a change of this delay. For more info, see http://digital.schoeps.de/documents/Schoeps_Manual_SuperCMIT-DE-111007.pdf

We designed this latency so that the minimum latency is short for high frequencies so that you can listen to a live mix on headphones with some live leakage.

In the experience of our users, this property is no disadvantage. Very seldomly, people ask, and I never heard of users that don't use the mic because of this.

Best regards,
Helmut"
do you guys have any idea how about to use this filter?
 

There is no one FIR filter. FIR is a type of design for various kinds of filters, but it's the only filter design that is phase coherent. So if you must absolutely stay in phase, such as in this case, you must use a FIR filter to split the signal into a number of frequency bands. Then you can phase-align the various bands independently. The easiest way would be to split of the unaffected higher frequency band and then move around the lower band until it fits. However, I don't know if just two bands are enough.
But as Helmut mentions, as soon as one of the signals changes distance, and thus phase, you'd need a new section with different alignment. As is being discussed in the other thread about phase, it is not very likely to be successful in any reasonable amount of time as the variance is far too high.
Probably best to work with other methods, such as levelling...

Share this post


Link to post
Share on other sites

I was pointed to this thread about the SuperCMIT latency and I'm happy to comment.

When we designed the SuperCMIT in 2010, we opted for a minimum latency for ch1 at higher frequencies and accepted the compromise that the latency got frequency-dependent. In practice we found no real disadvantage (loudspekers also have a frequency-dependent latency), but nevertheless we told our customers to keep it in mind.

It has always been more a theoretic issue, no practical user identified it as a problem. Actually, there has been one sound engineer that wanted to experiment with the SuperCMIT, ch1, in M/S, where you really need exact phase alignment. For this guy we produced the attached compensation FIR filter. 

Remember that the normal CMIT signal on ch2 of the SuperCMIT has a normal, frequency-independent latency. So whenever you need to mix a lav with the Super, simply choose ch2!

When using the SuperCMIT, in my guess 90% of the users use ch1, Preset1, 5% use ch1, Preset2 and 5% ch2. Usually ch2 is only recorded as a backup when the situation makes you doubt the algorithm or when you want to offer an option to post. 

SCMIT_inv_Groupdelay_FIR.wav

Share this post


Link to post
Share on other sites

Thank you Helmut for your post --- very clear explanation directly from the source! Beyond all the numbers, the science, what really nails it for me is your statement: "It has always been more a theoretic issue, no practical user identified it as a problem" --- I'm into all the theory, of course, but I'm also aware that the most important thing is the sound --- the art of critical listening and creative use of all these amazing tools is what matters.

Share this post


Link to post
Share on other sites

Thank you very much Helmut, I'm so glad you appeared in this thread. Your explanation really satisfied me. I also agree with mr. Waxler that the most important thing is the way you use your tools but this is really helpful to have more answers. 

On 31/12/2016 at 5:12 PM, Helmut Wittek said:

I was pointed to this thread about the SuperCMIT latency and I'm happy to comment.

When we designed the SuperCMIT in 2010, we opted for a minimum latency for ch1 at higher frequencies and accepted the compromise that the latency got frequency-dependent. In practice we found no real disadvantage (loudspekers also have a frequency-dependent latency), but nevertheless we told our customers to keep it in mind.

It has always been more a theoretic issue, no practical user identified it as a problem. Actually, there has been one sound engineer that wanted to experiment with the SuperCMIT, ch1, in M/S, where you really need exact phase alignment. For this guy we produced the attached compensation FIR filter. 

Remember that the normal CMIT signal on ch2 of the SuperCMIT has a normal, frequency-independent latency. So whenever you need to mix a lav with the Super, simply choose ch2!

When using the SuperCMIT, in my guess 90% of the users use ch1, Preset1, 5% use ch1, Preset2 and 5% ch2. Usually ch2 is only recorded as a backup when the situation makes you doubt the algorithm or when you want to offer an option to post. 

SCMIT_inv_Groupdelay_FIR.wav

The attached file is a 2 KB file... is that correct? 

Share this post


Link to post
Share on other sites

Yes, thank you very much, Helmut, that was great!
I am curious though, if money and time were no issue, would it be possible to have ch. 1 completely phase coherent across the entire spectrum? At least on preset 1?

Share this post


Link to post
Share on other sites

Angelo,  Just came across this thread.

What you are experiencing is far beyond a phase issue.  It is an incompatibility issue.

Using the AES CH.1 side of the SuperCmit you are using the DSP processing which is quite complex and in my opinion should never be utilized as the normal preferred output of this mic for exactly the issues your editor is experiencing.   My personal approach with this mic is to work the CH.2 -unprocessed audio and use DSP only when the exceptional noise reduction is required. 

When I do use the DSP in the production mix it is ISO'd as I always keep the CH.2-unprocessed on the dedicated Boom track at all times.  I also do my best to keep all wires out of the mix at this time with the exception of maybe an off camera line or to snatch a line that would be lost otherwise.

Being a digital microphone it takes extra care to mix with the wires.  The DSP makes it extremely difficult if not impossible to do so.  I have been known to crossfade between processed/unprocessed as a scene progresses to make it compatible with wires.  It is touchy.

The SuperCmit in my opinion is the most superior mic in my inventory and the quietest smoothest sounding shotgun I know of.

Share this post


Link to post
Share on other sites

The SuperCMIT is anything but quiet. When the scene gets quiet, and one has to crank the gain up, the SuperCMIT is noisy, which is a great shame. If you want a smooth and quiet shotgun then have a listen to the Neumann KMR81D (note the D there folks, which is D for Digital), which sounds gorgeous, but.... it doesn't do the SuperCMIT 'trick', which is really clever.

Kindest, sb

Share this post


Link to post
Share on other sites
1 hour ago, Bash said:

The SuperCMIT is anything but quiet. When the scene gets quiet, and one has to crank the gain up, the SuperCMIT is noisy, which is a great shame. If you want a smooth and quiet shotgun then have a listen to the Neumann KMR81D (note the D there folks, which is D for Digital), which sounds gorgeous, but.... it doesn't do the SuperCMIT 'trick', which is really clever.

Kindest, sb

Neumann/Sennheiser AES42 is really quiet. I have been using mkh8060 with mzd8000 and it works great

Share this post


Link to post
Share on other sites

Interesting! JWSound is such a wonderful site to improve your sound knowledge!

 

I experienced even phase issues between two analogue wireless transmissions of the same type, in the higher audio frequency range - older Sennheiser HiDyn stuff. With Lectro SMb and SRb it was from about 5 kHz. The higher the frequ, the more the issue. I splitted a signal to two transmitters, made a stereo recording and examined the results on Audition.

Exception: Sony DWX/N wireless (the newer ones) are in phase until at least 15 kHz. I suppose that Zaxcom and Lectro D4 will also do good when multiple signals are transmitted on one carrier.

Share this post


Link to post
Share on other sites
On 2.1.2017 at 4:31 PM, mr.incandenza said:

The attached file is a 2 KB file... is that correct? 

Yes, it's FIR filter which you use with a FIR convolution plugin.

On 2.1.2017 at 9:18 PM, Constantin said:

Yes, thank you very much, Helmut, that was great!
I am curious though, if money and time were no issue, would it be possible to have ch. 1 completely phase coherent across the entire spectrum? At least on preset 1?

Yes, this is possible in theory (in future versions) when you accept an overall latency of > 6 ms.

 

Share this post


Link to post
Share on other sites
On 8.01.2017 at 9:42 PM, RadoStefanov said:

Neumann/Sennheiser AES42 is really quiet. I have been using mkh8060 with mzd8000 and it works great

How do you connect the digital microphone to the Nomad Rado? regular XLR to AES input? How do you set the gain on that mic? Can you adjust the gain remotely from Nomad or there are two gain stages - one in the mic, one in the Nomad's digital input?

Share this post


Link to post
Share on other sites
4 hours ago, resonate said:

How do you connect the digital microphone to the Nomad Rado? regular XLR to AES input? How do you set the gain on that mic? Can you adjust the gain remotely from Nomad or there are two gain stages - one in the mic, one in the Nomad's digital input?

Aes42 in

xlr to xlr

No setup on mic

No adjusting

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now