Setting balanced line input on AF100

[Steve House]

I would interpret the line you quote from the manual, "the recording level for this camera is set 8db higher than Panasonic broadcasting recorders", to mean the recorders are equipped with averaging meters while the camera has a peak-reading meter. Averaging meters were/are the norm for broadcast playback equipment. (Note your quoted text does not compare the AF100 to broadcast CAMERAS, it compares it to broadcast RECORDERS.) What I think they are pointing out is the fact that a given signal will read 8dB higher on a peak-reading meter than it does on an averaging meter, even though the "real" signal level is the same in both cases. It's the same signal with the same RMS average level but the two types of meters display different values for it.

dB is a ratio comparing two values, not an absolute value in itself, and is meaningless unless the reference value is stated. "+4dBu" has a reference implied in the "u" part of the abbreviation. But -12 engraved on the camera meter simply means 12 db below whatever signal level drives the meter to full scale into clipping. Setting tone to -20dBFS means you have 20dB of elbow room before the signal clips. Setting it to -12 means you only have 12db to work with. Also, because the distribution of energy differs between sinewaves and real-world signals, peak meters respond to them differently. Lining up 0VU tone to -20 on your camera means that 0VU speech will indicate -12 on its meter.

Why do you need it to be -6dB in the mix? You don't adjust dialog levels to other mix elements. Just the reverse - you set the dialog level FIRST, then adjust all the other elements (music, FX, ambience, etc) to balance against the dialog. Dialog levels are set "to spec," for example see the Wiki discussion of "dialnorm" for current broadcast practices - dialog levels averaged over the entire program should be -24dBFS, etc. - and then the other elements can be balanced against it by ear on a calibrated monitor system. Where you should place it depends on the destination of the final release. So what is the ultimate release destination of your production?

 

[hscully]

The final release of the production is for film festivals. It is a short narrative film. What I have is anecdotal.

I'm just starting to working with my own sound gear, trying to get it as close as I can to what I've received from working with other working production sound guys. I set up the mixer as described and I noticed, in the timeline in the editor that my stuff was lower relative to what I had gotten on a few other occasions working with others into this camera. It was my perception that it was low compared to the production guy's stuff and I wanted to try to match it. The -6db dialog peaks are about what I got from them. The audio seemed to match the good stuff I'd gotten and I'd heard someplace that that was where you wanted the dialog's highest peaks to land. That's it. Not very scientific but that's really all I got. I can increase the audio level on the timeline +6db and I'm good.

All this got me wondering if I set up the line input on the camera correctly. The audio is clean, not clipped and I'm psyched about the apparent noise floor. I checked the manual again and no word on whether the level meters on the meter are peak or average. I'm sure you're right they are peak. Makes sense. So how would you set up with tone with this camera? As you say the tone is absolute at +4dbu, which is 0VU, right? Does that tone represent a signal that is representative of what would be "peak" dialog level? Average? When I was recording I was careful not to clip the meters on the mixer but they were still pretty good peaking below clipping but nice and hot, like -6 averaging between -20 and -12, you know. I thought the mic inputs were good. Hot but not clipping.

At Gary's suggestion I recorded the tone that was -12 on the camera level meter ant it shows up as -18 on the timeline, which got me thinking that I was low or at least different from what I expected. Thanks again for the detailed info.

 

[Steve House]

The final release of the production is for film festivals. It is a short narrative film. What I have is anecdotal.

I'm just starting to working with my own sound gear, trying to get it as close as I can to what I've received from working with other working production sound guys. I set up the mixer as described and I noticed, in the timeline in the editor that my stuff was lower relative to what I had gotten on a few other occasions working with others into this camera. It was my perception that it was low compared to the production guy's stuff and I wanted to try to match it. The -6db dialog peaks are about what I got from them. The audio seemed to match the good stuff I'd gotten and I'd heard someplace that that was where you wanted the dialog's highest peaks to land. That's it. Not very scientific but that's really all I got. I can increase the audio level on the timeline +6db and I'm good.

All this got me wondering if I set up the line input on the camera correctly. The audio is clean, not clipped and I'm psyched about the apparent noise floor. I checked the manual again and no word on whether the level meters on the meter are peak or average. I'm sure you're right they are peak. Makes sense. So how would you set up with tone with this camera? As you say the tone is absolute at +4dbu, which is 0VU, right? Does that tone represent a signal that is representative of what would be "peak" dialog level? Average? When I was recording I was careful not to clip the meters on the mixer but they were still pretty good peaking below clipping but nice and hot, like -6 averaging between -20 and -12, you know. I thought the mic inputs were good. Hot but not clipping.

At Gary's suggestion I recorded the tone that was -12 on the camera level meter ant it shows up as -18 on the timeline, which got me thinking that I was low or at least different from what I expected. Thanks again for the detailed info.

0VU is not necessarily +4dBu (1.228v), though that is the definition of the classic analog "studio line level." In fact, 0VU=0dBu (0.775v) is becoming increasingly more common. Check your mixer's documentation as to what standard your mixer meter is calibrated to. Think of "0VU" as the "optimum operating level" for the mixer and at that level it outputs whatever the design engineers decided it ought to output when the circuits are all operating at their peak performance settings. Whatever the mixer is outputting at 0VU, you can arbitrarily place that signal at any desired level on the recording device by adjusting the recorder's input level controls. The classic SMPTE standards were 0VU = +4dBu, lined up with -20dBfs digital. EBU standards (which are becoming increasingly used outside the EBU as well) are 0VU=0dBu, lined up to -18dBfs digital.

I know I've said it before but it bears repeating - the basic purpose of the line-up tone is to calibrate the mixer's meter to the camera or recorder's meter. If, for example, you line up 0VU TONE with -18 on the camera meter, just by watching the mixer meter you can be confident that 0VU speech is going to reading around -10dBFS on the camera meter. Lining up TONE at -12 seems too hot to me - that means 0VU speech will be up around -4, leaving you almost no room for the normal ups and downs of speech volume or sudden loud sounds. OTOH Recording SPEECH at -12 seems normal.

Was looking back over some of the posts and there may be some confusion as to what dBfs refers to. The "fs" is "Full Scale", the end of the meter where the signal voltage is translated into "111111....", the largest possible number digital, and anything that goes over that is clipped (assuming the meter is accurate). -12dBfs means 12 dB below full scale. Since "dB" is a ratio, -12dB means exactly the same thing on every meter that exists, ie it is impossible for -12dBfs on one meter to mean -20dBfs on another. Of course, the input signal VOLTAGE that translates to -12 on one might translate to -20 on another but that's another issue. If you are recording at -12 you can have sounds go up to 12 dB louder before you clip while if you record at -20 you can go 20dB louder before you clip.