comparing Neumann 81i/Sanken CS-3e

[Philip Fass]

I need a good, versatile shotgun that works decently indoors. I've read lots of comments that the CS-3e is the closest you can get to an "all-around" shotgun. But listening to the samples in Dan Brockett's article --
As I Hear It - Choosing the Right Microphone -- these are my impressions:

On-axis, both indoors and out, the Sanken seems somewhat muffled compared to the Neumann.
Off-axis rejection and frequency response sound very similar with both.
I hear less handling noise with the Sanken....with the Neumann I hear a rumble, which might be avoided with low cut filter or blimp.

So I'm wondering what sorts of situations that aren't covered in the article would make the Sanken more of an all-'rounder.

[Rick Reineke]

The Neumann KM81/82 are 'interference tube' type mic, the CS-3e is not. A good shotgun can sound wonderful indoors... but not in all instances, depends on the room, and then you would want an alternative. The CS-3e would certainly be my choice in a 'one mic fits all ' scenario.
Despite it's versatility and though in the minority, not everyone is so enamored with the CS-3 for the very reason you describe.

[Philip Fass]

Being hundreds of miles from the nearest pro video dealer, I have to decide between what I heard in the samples (and go with 81i) and what I've read in comments (CS-3e). I'm hoping I can't go too far wrong either way.

[Jon Fairhurst]

You might try adding some HF boost on the CS-3e samples to see if you can get a less muffled sound.

The thing about EQ is that it's easy to add in post. Echo removal and phasing is pretty much impossible to remove in post. For that reason, I'd steer clear of any interference tube mic for indoor use - if you can EQ the CS-3e to your liking.

[Philip Fass]

Good point, Jon. I know from my own editing experiences.

[Jon Fairhurst]

Philip,

You might know this already, but it might be helpful for other readers...

The voice has a few key frequency ranges:

* 200-300 Hz - This is the fundamental. If the voice sounds boomy or thin, try optimizing this. It varies for each speaker. For dialog, I often put different talent on different tracks so I can tweak this separately.

* 1-1.2 kHz - This is where the consonants live. Boost to help intelligibility, but not so much as to sound unnatural.

* 2.4 kHz region - This is the area for nasal sounds. Boost if you want a dull voice to sound more lively. Cut if the voice sounds sharp and annoying. It works on musical instruments such as the oboe and viola as well. This definitely varies by voice/instrument. It also helps us differentiate similar voices. Cut this frequency and it's hard to tell an oboe from a flute.

* 5-15 kHz - This is the "air" around a voice. Boost for a lighter, more open sound. But be careful not to boost noise (hiss) and sibilant tones.

I'd be curious to know what sounded lacking in the CS-3e. Of course, it might be due to the recording space or the talent. My guess is that it was either due to less "character" in the 2.4 kHz range, and/or less air. The "character" part can be due to the talent, while the "air" is more dependent on the mic/preamp.

[Philip Fass]

Jon, I've never seen the frequencies broken down that way. A great help to me, as well as others I'm sure!

Could you take a quick listen to the CS-3e and 81i samples in the first group on this page?

As I Hear It - Choosing the Right Microphone

They're about 1/3 of the way down the page.

I'd just like to know how you'd compare them. (Maybe my hearing is off.)

[Jon Fairhurst]

Hi Philip,

I did a quick comparison on the Male and Female on-axis interior samples. On this system, I'm using a PC with a basic soundcard into Sennheiser HD280 Pro headphones. Note that these cans don't have extended bass, so I don't hear the handling noise and rumble you mentioned. I'm sure that my 15" full sized monitors at home would say otherwise.

On the female voice, the levels are VERY different. I had to boost the Neuman track by about 8dB to get a similar feel. The Sanken has a much stronger fundamental tone on her voice, and that makes it sound boomier and heavier in the low mids. Also, because I have to boost the Neumann so much to match the fundamental tone, it becomes louder in the higher ranges, and loud sounds are always preferred to our ears.

Without EQ, the Sanken sounds a bit scooped, while the Neumann sounds flatter. I'm wondering if the distance from the mic was different between the recordings, since the Neumann level is so much lower. We might be hearing a proximity effect on the Sanken. In general, when we turn the overall volume down, scooped sounds better. When we turn it up, flat sounds better. It's hard to really get an apples to apples comparison.

The levels for the male voice are MUCH closer. The Sanken is still heavier on the fundamental on the male voice. When the bass rolloff is engaged, the tone is closer.

My next step will be to normalize the fundamental tones by removing some of the 2-300 Hz content of the Sanken, then I'll match levels. Until I do that, it's hard to really compare the sounds in detail.

That said, the Sanken seems to have a bit more "bite" in the 2.4kHz range while the Neumann is more technical sounding. Personally, I like a bit of bite as it gives a voice presence. I have a couple Sanken COS-11D lavs, so this Sanken sound is very familiar to me.

Regarding the fundamental tones, I think it's fair (for me anyway) to EQ them to be similar for a second level test, given that I almost always EQ this for real-life dialog. Different fundamentals can make a huge difference in our initial feelings about a mic test and in getting tracks to match. As long as a mic isn't truly lacking in the lows, it can be okay. Clearly, neither mic is lacking in the lows, given the rumble that you hear on the Neumann.

In fact, the extended frequency response of the Neumann points to its musical recording roots. Musical mics need to go deep. Sanken comes at this from the Japanese broadcaster perspective, so it's about voice. No surprise that they emphasize the fundamental, but don't have as low a reach.

Some extended or boosted HF tones can also be good on a shotgun, given that we often put them in wind protectors that cut the highs. Again the Sanken is scooped, so once I pull down the fundamental, I expect the HFs to be on the high side. We will see...

My initial feeling is that these are both very high quality mics, but they are definitely tuned differently.

I'm fairly busy tonight but might be able to play with some EQ on my home system on Wednesday...

[Philip Fass]

Jon, your comments are so valuable to me that I'm almost tempted to frame and hang them on the wall!

Just to expose my ignorance a little more, as I understand from Googling around:

scooped= emphasizing bass and treble more than mid-tones
fundamental tone=the lowest frequency in the sound you're hearing

About right?

Many thanks!

[Jon Fairhurst]

Thanks Philip!

Yes, your understanding is correct. On the fundamental tone, it's basically the lowest pitched tone, but there can be lower unpitched sounds, like plosives. If one were to sing a steady note, then, yes, it's the absolute lowest tone.

BTW, those frequencies I gave earlier are not only helpful for understanding how to EQ voice, but they're also helpful for understanding how to mix in music. The 600 Hz range isn't as important for voice, so you can put music there. Cut the music at 1.2 kHz so it doesn't compete with the consonants. Balance music and voice at 2.4 kHz - if the voice is dull, you need to give more to the voice. If the voice is nasal, you can cut there and give more to the music. Music can almost always get more air above 5kHz without getting in the way of the voice. And you can push the music below 200 Hz, since that's below the voice fundamental.

Do this well, and you can get the music a bit louder without overpowering the voice.

[Philip Fass]

I think the single biggest lesson I've learned is that what sounds best to me in a set of samples isn't necessarily what's best to use. It's like choosing to eat broccoli when there's chocolate cake sitting next to it.

[Garrett Low]

Philip, thanks for starting what began as an interesting thread and has turned into a greatly informative thread.

Jon,thanks for the great information and explanations on a subject that is very foreign to many of us who concentrate on the technical cinematography side but often neglect to educate ourselves about the audio side. I belong to an independent film cooperative and would like to ask your permission to quote some of your posts in this thread on our groups web forum. I think it would be of great value to our members in helping them understand and appreciate the need to truly educate themselves on how to improve the sound of their films. I would of course give full credit and acknowledgement to you for your wisdom.

-Garrett

[Jon Fairhurst]

Hi Garrett,

Please feel free to quote this. Also, please give a nod to DVInfo. I'm the writer. Chris is the publisher. :)

BTW, I got the initial information from an article by Bill Rose, and who knows where he learned it. I've put it into use many times with good results.

[Philip Fass]

Any relation to Jay Rose, who wrote one of my audio-for-video books? None of the Bill Rose's I searched look like the right person.

[Jon Fairhurst]

My mistake! Jay Rose! (I have a friend named Bill Rose, but he's not an audio guy.)

Thanks for patching my memory. :)