Acceptable Recording Levels for SD702T

[tong]

I'm doing an indie short film recording on a 702T this week. Just like to find out what is the acceptable range of levels on the recorder. My range was between -30 to -4dBu but it turn out soft when playback on the MacBook pro speaker.

[Marc Wielage]

I usually see sound in dailies peak a little higher than -10dBFS, but not much more than that. You need some headroom for people who yell.

Don't trust the speaker. Use headphones. If possible, do a test and let the editor check it on their system. I think getting the levels consistent day to day is one of the hardest issues to overcome.

--Marc W.

[tong]

you are right indeed! i had better do a test check this wedneday before the day begans!

[RPSharman]

You will find that Quicktime, or other such programs like iTunes, will "normalize" your peaks to an "acceptable" level.  Your peaks might be 0dBFS because of slates or other noises that peak the recording.  This normalization will then make the level of your recorded voices to appear quite low.  When brought into a DAW, all will be ok.

[justanross]

I have wave agent on my computer to check my files.  I find it works great.

[studiomprd]

" what is the acceptable range of levels on the recorder. "

acceptable to whom??

as a production sound recordist, you don't want to hit the brick wall (0 dBFS).

you don't want to waste a lot of headroom, and record down near the mud...

[WhyOne]

When Post finally allowed 24 bit files, and working with a -20dB reference = to -8dB on a PPM meter, I mix to the PPM with "visible" peaks kissing 0 on the PPM. The resulting peaks on my 788 (with a much faster rise time) will often hit -4dB, but never overload. Feels like a Nagra again...

[peter775]

nogwiat,

listening too soft on speakers has nothing to do with your recording level. You have a volume knob to set your speaker level. Recoding level is NOT to serve correct listening level.

Playback some pink noise over you speakers, set SPL to 79dB and you're good.

Most edit rooms listen WAY too soft and then slam all the levels to the max losing all the head room.

also don't listen too people who tell you to record as hot as possible "to get all the bits". This is nonsense at 24bit. The system-noise is allways the same, no matter how hot you record. It is detmined by the bit depth not the recording level. Record with your average level at around -20dBFs anod you Re good.

In a movie theater dialog recorded at around -20dBs is already reproduced at 85dBPL. If you record extremely hot the mixer wil,gain everything down anyway. As a result you have no more headroom.

[peter775]

sorry,

i meat to say to play back pink noise of -20dBFs and set that to 79dBSPL.

[Glen Trew]

I'm doing an indie short film recording on a 702T this week. Just like to find out what is the acceptable range of levels on the recorder. My range was between -30 to -4dBu but it turn out soft when playback on the MacBook pro speaker.

Original tracks recorded with a 0VU reference of -20dBfs will almost always sound less-loud than typical mastered end-product. This is because of two reasons: 1) The standard 20dB of headroom above 0VU reference allows for surprise peaks that mastered program does not have to allow for (after the original recording, there are no surprises). 2) Original tracks are usually compressed very little, allowing that subjective touch for the final mix. (compression is generally done to make the overall volume louder.)

That said, acceptable recording levels for the 702T (and any other modern digital recorder) are up to 0dBfs (max on the record meter). But when recording original tracks, you never really know what the maximum level will be, so it's best to allow a few dB for safety.

If you go your entire career without ever going beyond digital max, that means that much of your recordings have either had too much limiting/compression, or have been recorded too low. Take reasonable precautions to avoid digital max, but don't let the precautions be worse than the rare "splat".

Glen Trew

[peter775]

If you go your entire career without ever going beyond digital max, that means that much of your recordings have either had too much limiting/compression, or have been recorded too low. Take reasonable precautions to avoid digital max, but don't let the precautions be worse than the rare "splat".

I often get production audio in post that was recorded as if we were still in the DAT ages. Believe me extra-hot levels like dialog always around -10dBFs or more WILL get gained down in post. Recording average levels at 10dBFs or hotter will ONLY give you less headroom. Your sound will not sound better. The only thing you´re doing is give post a hard time. Less headroom in post for DSP processing. Constantly clipped levels after EQ is applied etc.

On one of my last projects I had to manually and destructively gain down every single clip because It wasn´t possible to work on these files because they were already maxed out all the time. Result: Dialog was back at normal levels but lots of wasted space (headroom) that could have been used if it had been recorded 10dBs lower in the first place.

Leave your average level at around -20dBFs and allow room for surprises instead of slamming everything into your limiter.

It´s a general misconception of many picture editors that hot tracks equal "good" sound. It´s complete nonsense. If they want more volume they need to grab the knob on their monitor system.

Hot tracks equal less headroom and more limiting.

Keep in mind that in cinemas the monitoring is calibrated to 85dBSPL@-20dBFs per speaker. Average dialog level on a final mix is around -20dBFs. It doesn´t make sense to record 10dBs hotter because it WILL be gained down in post anyway otherwise you will end up with bleeding ears.

Again: You gain nothing from recording extra-hot.

What I usually do before a new project is to go to the edit room. Set monitoring levels at around 79dBSPLand tape down the volume knob and ask editors to work with this level.

Asking the recordist to record hotter to compensate for lousy monitoring setups is a bad idea.

[RPSharman]

I tend to record in the way Glen has described.  I record pretty hot, and haven't ever been told by anyone at post that this is a bad thing.  I always call the post sound guys (these have included award winners Paul Ottosson and Chris Aud), and have not been told once that my levels are too hot.

It would be great if some other post guys would chime in like Peter has to offer their thoughts.  I'd like to know if recording hot really makes no difference at all, is a bad thing, or is a good thing.

Robert

[peter775]

I  I'd like to know if recording hot really makes no difference at all, is a bad thing, or is a good thing.

Like I said. The hotter you record the smaller your head-room, the more compression you have to apply in order to keep loud parts below full-scale. This is a technical fact.

For theatrical there are calibrated relations between absolut digital level and listening volume. If you record a voice with an average level of -10 to -5dBFs it WILL be turned down in the mix. otherwise it will sound at around 105dBSPL. This is a technical fact no matter how many Oscars your re-recording mixer has won. No one will ever need dialog averaging at 105dBSPL so why record it that high and waste all the nice head-room?

Of course you can later turn those levels down but it will leave you with a sound with low dynamic content since you gave away your headroom for your hot levels. This is not an issue when you record a stable narrator´s voice but for production sound where you encounter sudden peaks it´s not great to record super-hot because your peaks will end up slammed down by the limiter.

Give your tracks reasonable head-room for post. Don´t slam your tracks. This won´t be re-doable in post whereas reducing unwanted dynamics later is easy. Leave your tracks some room to "breathe".

But let me ask from the other side: What in your opinion is the advantage of recording super-hot?

[John Blankenship]

My standard is still -20dBfs nominal average with peaks to -10dBfs, and with occasional peaks 2 or 3 dB above that as needed and only barely hitting the limiter (set for about a -4dBfs) with an occasional errant peak.

This may be adjusted a bit depending upon what I'm mixing.

To the best of my knowledge, the "network standard" is still -20dB average with peaks to -10.

John B., CAS

[RPSharman]

Thanks, Peter.  I didn't mean to imply these post guys were more qualified or knowledgeable.  I just wanted to show that they were indeed qualified.

Another thing to consider is that while sitting in a final mix session at Warner Brothers, I learned that on "bigger" projects they discard the mix and ALL dialog is remixed from the ISO tracks.  This brings up the question of recording level for ISO tracks, which has been previously discussed.

I suppose that some of my films have been remixed from the ISO tracks, which are typically an average of 10dB lower than the mix.

Robert

[Eric Toline]

I think some of the "recording hot" theory is a hold over from analog days where you recorded hot to overcome tape hiss and other analog/mag track/tubed mixing console noise issues. Since that is no longer a factor, keeping dialog peaks to a -10dbfs level is a good idea as it still leaves 10db of headroom.

Eric 

[peter775]

Thanks, Peter.  I didn't mean to imply these post guys were more qualified or knowledgeable.  I just wanted to show that they were indeed qualified.

Another thing to consider is that while sitting in a final mix session at Warner Brothers, I learned that on "bigger" projects they discard the mix and ALL dialog is remixed from the ISO tracks.  This brings up the question of recording level for ISO tracks, which has been previously discussed.

I suppose that some of my films have been remixed from the ISO tracks, which are typically an average of 10dB lower than the mix.

Robert

Aha. OK. That makes sense. The ISO-track thing didn´t ocur to me since the OP was talking about a 702T ;-)

[peter775]

My standard is still -20dBfs nominal average with peaks to -10dBfs, and with occasional peaks 2 or 3 dB above that as needed and only barely hitting the limiter (set for about a -4dBfs) with an occasional errant peak.

This may be adjusted a bit depending upon what I'm mixing.

To the best of my knowledge, the "network standard" is still -20dB average with peaks to -10.

John B., CAS

Exactly! Actually TV has moved to more and more to loudness metering instead of peak levels (see CALM act). But the network standard is about master-tapes AFTER being mixed.

The points I made are about the raw production tracks. No one expects us to record production tracks to final master level specs. It´s just that reasonable head-room gives post much more options to work with compared to already slammed tracks from set.

This is something very hard to understand for picture editors. My experience is that if you explain to them the level/volume relationship and gain-staging of their edit-systems they won´t complain about "too low tracks". In fact they actually enjoy the fun headroom they also gained for all their explosions and music.

If they monitor too soft and slam all the levels inside the NLE all they gain is a flat compressed rough-mix with no lows and highs. 

I´ve worked on lots of feature films in post and my experience is the first thing I do when I get the OMF transfer from the editor is turn down the levels by at least 6 sometimes 10dBs to be able to listen to them in a calibrated room.

[John Blankenship]

Peter, I'm not disagreeing with any of your points by saying this, just adding my experiences to the discussion:

I've had several networks tell me they like to see their field tracks come in at the -20 average/-10 peaks standard.  Of course, if this is observed correctly (with adjustments depending upon the situation) there's little need for hitting the limiter very often.

When wearing my post hat, that's about where I like to see them.  In the edit suite, a little breathing room is good -- but, too much can be a bother.

Naturally, mix levels are relative -- like you say, it's a point many video editors don't get.

John B., CAS

[peter775]

Peter, I'm not disagreeing with any of your points by saying this, just adding my experiences to the discussion:

I've had several networks tell me they like to see their field tracks come in at the -20 average/-10 peaks standard.  Of course, if this is observed correctly (with adjustments depending upon the situation) there's little need for hitting the limiter very often.

When wearing my post hat, that's about where I like to see them.  In the edit suite, a little breathing room is good -- but, too much can be a bother.

Naturally, mix levels are relative -- like you say, it's a point many video editors don't get.

John B., CAS

Agreed.

[Petros Kolyvas]

Acceptable recording levels is a wonderful subject that I wish came up more often; and not simply in sound-for-picture discussions but in all audio production.

Here are my thoughts, most of which fall on the user-interface side of things.

Metering in DAWs and NLEs is quite poor, especially for those who have little knowledge about the intricacies of digital audio (people starting out or many video editors). Now, before I go any further, I am all for "using your ears." However, in an increasingly visual, "trust the machine" world it's often a big step for the uninitiated. At this point I'd hope that metering would help, but it doesn't. Far from it in my view.

Take, for example, this default metering setup of Logic Pro:

To the newcomer, who might lack an understanding of headroom and of absolute-scale values as they correspond to the real world experience thereof, these signals all look unacceptably low. Admittedly, they only "look" low, but the deception is often enough. More than 3/4s of the meters are devoted to the last 3-bits of resolution (at 6dB per bit for uncompressed PCM audio). People, for whatever reason, like to "fill things up" or "get the most out of it" - not realizing that tracking average levels like these do help one "get the most out of" 24-bit digital audio. (Bare with me, I realize I'm also preaching to the choir here.)

A quick right-click of the mouse and the metering scale can go from exponential, to a more real-world reflective "sectional dB-linear" scale:

Better, but not there yet.

What I've always wanted to see in DAWs and NLEs was the option to set a reference level for audio when you start a new project. Then the meters could be toggled between a referenced scale, where the user sees 0 (and aims for it - with the headroom displayed more appropriately) and a dBFS scale when required.

Of course, none of this negates the fact that professionals usually know this type of minutia inside and out, but the up-and-commers and cross-over hobbyists might not. Tutorials for audio and video editing applications should include a whole section on signals and reference levels, once people get it, it can change everything.

I'm constantly reminded of a conversation I had with someone working on their English doctorate. I was being my usual self, complaining about how people are too uptight about creative use of grammar. She replied that it was fine to break the rules as required, but you had to know what the rules were first! That stuck with me. Well that, and she claimed said my piano was "too plinky" at the top.

Anyway, just my cranky 2c.

[berniebeaudry]

Peter, I'm not disagreeing with any of your points by saying this, just adding my experiences to the discussion:

I've had several networks tell me they like to see their field tracks come in at the -20 average/-10 peaks standard.  Of course, if this is observed correctly (with adjustments depending upon the situation) there's little need for hitting the limiter very often.

When wearing my post hat, that's about where I like to see them.  In the edit suite, a little breathing room is good -- but, too much can be a bother.

Naturally, mix levels are relative -- like you say, it's a point many video editors don't get.

John B., CAS

Most video editors I work with would rather have to turn down the production tracks than have to gain them up.  I run about the same as you John.  If you have to gain up too much eventually ambient sound can become an issue.  If you have a strong but not overly hot levels and you have to come down then the room tone is decreased as well. 

Bernie

[Alan H. Chang]

Most video editors I work with would rather have to turn down the production tracks than have to gain them up.  I run about the same as you John.  If you have to gain up too much eventually ambient sound can become an issue.  If you have a strong but not overly hot levels and you have to come down then the room tone is decreased as well. 

Bernie

As most NLE's fader and gain settings are set default to unity gain, editors would need, want, and favor turning down their levels than having to turn up.  There's not a lot of makeup gain after unity unless you insert a limiter of some sort to gain an add'l 30db+ for those quiet recordings.

The ambient sound level, however, would NOT be an issue if proper recording and booming techniques were executed on set.  The whole Signal-Noise-Ratio rule.  Even if the recordings were low on tape, the ambient levels will be buried underneath if done properly.  Should the mic become more distant, such as the onboard mics of the Canon 5D, then you should start worrying about the ambient level becoming an issue because the proportion between the dialogue/ambient levels are ever reaching 1-to-1 ratio.  Otherwise, no.

[berniebeaudry]

As most NLE's fader and gain settings are set default to unity gain, editors would need, want, and favor turning down their levels than having to turn up.  There's not a lot of makeup gain after unity unless you insert a limiter of some sort to gain an add'l 30db+ for those quiet recordings.

The ambient sound level, however, would NOT be an issue if proper recording and booming techniques were executed on set.  The whole Signal-Noise-Ratio rule.  Even if the recordings were low on tape, the ambient levels will be buried underneath if done properly.  Should the mic become more distant, such as the onboard mics of the Canon 5D, then you should start worrying about the ambient level becoming an issue because the proportion between the dialogue/ambient levels are ever reaching 1-to-1 ratio.  Otherwise, no.

I agree with that opinion.  What I was referring to was when shooting in noisy environments ie: road side, sporting events, factory etc. where you don't have control of the ambiance.  If the distance from direct sound to mic is correct then gaining down the track reduces the background sound and the voice (direct sound) becomes more prominent.  Usually in such a situation the talent talks louder to hear themselves so its easier to get good strong levels and minimize the background. 

[Alan H. Chang]

I agree with that opinion.  What I was referring to was when shooting in noisy environments ie: road side, sporting events, factory etc. where you don't have control of the ambiance.  If the distance from direct sound to mic is correct then gaining down the track reduces the background sound and the voice (direct sound) becomes more prominent.  Usually in such a situation the talent talks louder to hear themselves so its easier to get good strong levels and minimize the background.