How to Mix Audio Levels for Films & Movies - Introduction to Cinema Sound for Indie Films Part 01
Vložit
- čas přidán 5. 09. 2024
- here is a new video of the mixing process - • Sound Design: How to M...
Up until about the first minute, I am actually talking about the mix as played back in a cinema (acoustically), not on the digital timeline (electronically).
This first segment covers dynamic range of films and other popular media, cinema acoustics, The X-Curve, basic speaker calibration and how to estimate what your film will sound like in a theatre (theater, in US).
I will be covering LUFS, LKFS, etc soon...stay tuned!
The one video I've been looking for, for a long, long time! Thank you!
Always a nice surprise finding a Curtis Judd comment in the wild 👌🏼
So I mixed my short film first by keeping dialogue at around -12db, and my music is some times louder depending on the scary movements, However, the sub 40hrz and lower I seem to be struggling with.
I plan on putting this in film festivals but knowing full well that sometimes people will have a sub woofer and other not. So where would the sweet spot be without killing people in a theater with sub info or on the other end, loosing the power of the sub in very specific spots of my video that would loose its impact with out the sub.
@@commercialand Oh geez, sorry I missed this; I will answer shortly, even if not needed anymore because this is an excellent question!
This guy knows stuff.
Just got called to mix an indie movie for pre-public screenings. Absolute God sent!
Let me know if you need any extra info!
Good luck!
Watching this video about 9 years after it was made really makes me nostalgic for CZcams from 9 years ago. I didn't get a VPN ad read. Half the length of the video wasn't calls to action for the channel. It wasn't stretched out to > 10m just to make room for a midroll ad. The "intro" was just a quick title card and one sentence statement of topic rather than a minute long branding logo animation and annoying jingle.
Literally just a person who knows a subject talking about what I wanted to know more about. "The Algorithm" almost never recommends stuff this well made any more.
Thanks for the comment...just one note, the mid-roll ad was placed by YT, as I didn't even know you could do that at the time...because of this comment, I am going going to turn any mid roll ads off...cheers! and thanks again!!
Mid-roll ads are now off...I actually thought tey had tobe 10 minutes long for that, learning that just a few years ago. Oh well, cheers!
@@bacontrees I don't think I saw one. Yeah, what I was saying was that some current CZcamsrs will stretch 5 minutes of actual content to 10:01 in order to probably get a midroll ad. This video was over 8:00, so it clearly wasn't being stretched to get over that threshold which is nice.
@@guaposneeze When I went to my control bpanel, mid-roll ads were actually checked! I unchecked them, but I do get your point...thanks for the info!!
thanks man. i loved that last advice of putting my favourite movie on timeline and checking.
This was very useful information for improving my home theater sound experience. Having a better idea behind the mentality of the way films are mixed helped me identify and solve issues in my room. Can't wait for a part 2.
Very cool!
Very awesome and I just love the way you explain things in a way that is informative but does not feel intimidating either.
One thing to mention - HDTV signals follow the LKFS (or LUFS) perceived loudness measurement, which is another, upcoming discussion...cheers!
I enjoyed all of your classes so much. With a lot of energy and great information.
Thanks Buck.
Excellent work there! Oh and a great day to everybody reading this! :)
There's a science to this?? Wow.
thank you so much! I'm just finishing up a clip to be marked on an audio production course, little was touched on regarding levels andmixing though this gave me some direction to go in. Thank you very much!
Cool, thank you! There are at least 3 more videos in this series that get more detailed and with listening samples...glad it could help...cheers!
Very interesting, thank you. I also follow Sir Curtis Judd.
Curtis Judd has a lot of great videos on the subject of video and filmmaking!
The level of detail here compared to the common CZcamsr's tutorial is hilarious
Very well explained.
Thank you! Cheers!
Great video Buck! Felt like I was back at school! Keep making more.
Manni
This guy knows his shit
This is convoluted. All you need to know is for both dialogue and music:
For TV streaming or distribution: You need to hit -24 LUFS (long term) plus or minus 2.
For Film it needs to be: -27LUFS (long term) plus or minus 2.
Short term readings is never important.
The DBTP: Decibal True Peak, -2 is the most common specification we see. With film you might be able to get away with a higher limit, but to play safe go to -2. Use your limiter to set the True Peak max at -2. In most cases, you can’t go over it.
Well, this video was 2013, and since then I have created a few more with sound levels maps with targets of -24+/-2dB and suggested -27+/-2dB for film...I even sugest True Peak to be -5dBFS. Feature films have no absolutes, yet Netflix opts for the -27dBFS (dialog-gated) +/-2dB. So, since this video, there have been a lot of developments. Cheers!
My more recent videos have aimed to mae this unconvoluted, actually.
@@bacontrees Fair enough. Thanks, I'll check them out ✌
@@Tyrell_Corp2019 And the name Tyrell Corp....why didn't I think of that name? Cheers!
This was a great find! Very informative.
Thanks so much for this. Very clear and sensibly explained.
Richard Michalak Thanks for the comment!
Very informative, mixing my movie right now and this information helped me a lot. Thnx.
Thanks! There are more on my channel with more detailed info....cheers!
@@bacontrees which channels would you recommend? or maybe online courses you would recommend?
This is gold.
Brilliant!
Stinking brilliant! Thanks for the instruction!
This is a super useful video, thank you for sharing, it helped a lot. I'm doing a 5 mins long video for school assignment and it helped a lot me to find the proper levels in the mix. Thanks again!
Thank you!
Great help, so well explained. Thank you.
Awesome video. Make more post mixing videos!
Nice video, well explained and super clean, congrats!
This is a great lecture.
Very good... Thorough and informative video.
I can't believe this tutorial has been up for almost two years already...
Thanks!
I am maybe far too common example of a professional with much experience (directed, wrote, shot, edited, mixed 7 feature films), yet who knows so little still about sound mix. And now I'm getting my mixes being rejected by QC people...
Not to abuse too much of your time, I'll just stick to one of the few questions I've had in my mind forever:
Regarding dialog (which I understood should be averaged at about -12db), what do I do when the actors are whispering? Do I up the level of their whispers to the same -12db? Or should it sound a little lower than normal dialog?
Thanks for any help!
Ivan
Ivan Noel - Hi Ivan, do you mean average -12dB for the recording or mix? For recordings, I always average -12dB for average dialogue and my back-up channel is -20dB or -22dB (depends on the resolution of the meter. If they whisper, I get a mic as close as possible to average -12dB.
In the mix - I set my speakers to 85dB SPL, then my dialogue should play back around 60dB SPL (A-weighted - for clarity).....It's hard to say what that would be on the dBFS scale as compression and EQ settings can vary by the mixer. I tend to use around 4:1 Compression Ratio and maybe a -15dB to -18dB threshold (assuming the dialogue was recorded to averaging-12dB).
BEST WAY to check (and I will have a video on this soon) - compare with other films - loudest peak around -2dB and dialogue played back at around 60dB SPL....did you see part 2? I go into more tech details.....
let me know if I have helped in any way and thanks for watching!! Cheers!
bacontrees Hi, I did understand all the above, thanks.
So... I am right in understanding that whispers are heard at the same db level as dialog (only that you make sure of that at the recording stage)?
- Another question I have been holding in for a while... (for lack of anyone able to answer it):
when one hears audio in the cinemas it seems they have a MUCH bigger dynamic range than my pro monitor speakers at home. From very quiet, sometimes one gets the most god-almighty transient sounds that seem to defy all dynamic range. In my limited range at home, I can never seem to be able to get such a loud sound without it distorting.
The question is... how do they do it?
Do they have other speakers, at ready, set higher in volume, just to give you that terrible bang of a sound and make you jump off your seat?
Do they somehow benefit from a much wider dynamic range than us nere mortals?
I somehow fear the answer might not be straightforward...
- An extra bonus question (my last, promise): I am aware that 0db is the max level for digital sound. But ...does it matter very much that I have the occasional +1 from a transient sound (that doesn't sound distorted or crackle), or is it doomed in any case because on some machine it might distort?
Many thanks for your help... it's been such a very long while since I could speak to a sound pro. Audio is without doubt (with a close second being color correction) the least understood aspect of filmmaking. Even my own sound guy couldn't answer most of my questions.
Ivan Noel oh, okay, missed the whisper thing....I tend to 'cheat' whispers - make them a little louder than normal, but no quite as loud as dialogue - since whispers contain very little low frequency info, they can be almost as loud as normal dialogue, and leaving them a little lower will actually make people stop to listen more (in a cinema).
The dynamic range of a cinemas greater than monitor speaker for sure...the almost 30dB DR of a mix on a timeline is made more dynamics by the playback systems (Dark Knight was close to 30dB on timeline, and up to 45dB in IMAX cinema).
No clipping to +1dB whatsoever - a hard-knee limiter must be put on the entire mix and set to maybe -2dB (or even -1dB, but I choose -2 to be safe). Anything over 0dB is distortion and usually will be rejected.....
I agree, audio isn't understood nearly as much as picture....my take on it is that nobody sees the waveforms until post! Hope this helps, post another question or comment if you need more info...
bacontrees ok great. that was simple and clear enough!
- about greater cinema range: how then does one 'adapt'? Does one simply start with lower levels in cinema version so the loud bits sound louder?
- You say put a hard-knee limiter. Doesn't that 'crushing' of sound not make it SOUND distorted (though scientifically it isn't)? Can I therefore not simply add a limiter to all my transient sound clips so that, in the final mix, they will never be either over 0db or 'distorted'? Or should no sound clip anywhere ever be over 0db?
If mixing for cinema, dialogue would play back at normal conversational levels at 2/3 back in the theatre (60dB SPL or so at that distance). Loudest sound FX would be over 90dB SPL (and more, possibly, with the bass considered). Once the monitors are set up for mixing, dialogue should always be clear and usually at about 60dB. The cinema will add more SPL, especially in the bass.
Some mixers mix lower level sounds even lower before a big FX blast, to cheat the limited dynamic range of a soundtrack, so the sound levels have more contrast, making the louder sounds sound even louder (the art part of the mix).
A limiter is just preventing the few party-crashers from crashing the party, and will not have a very noticeable effect on the transients unless the threshold is set too low....I set my limiters to about -2dB threshold and 20:1 ratio or even infinity to 1 only to prevent clipped peaks.
If you calibrate your speakers to 85dB with pink noise, then mix dialogue around 60dB SPL, the rest is by feel. It is not an exact science but the mix will probably hold up well, with no distortion (with a limiter on). But, the EQ also has a role in the clarity and perceived loudness of the dialogue. I use the D-curve, as mentioned in part 2.
Thank you so much for your insightfull and extremely valuable knowledge sharing video. I am facing a trouble in Delhi/India (PAL transmission standard) and totally unable to troubleshoot.
My audio output sounds well on every media except T.V.
Problem is audio is loud enough in the starting but as it proceeds, it gradually gets supressed. if we revind it a few seconds, the same audio sounds loud.
No where I have found any solution as yet.
very helpful video thanks
this explains why when im watching most movies i can barely hear the dialogue but then an intense scene happens and its way too loud
Dialog is always king
Thanks for this!
Thanks a lot! Nice explanation! 👍
So I mixed my short film first by keeping dialogue at around -12db, and my music is some times louder depending on the scary movements, However, the sub 40hrz and lower I seem to be struggling with.
I plan on putting this in film festivals but knowing full well that sometimes people will have a sub woofer and other not. So where would the sweet spot be without killing people in a theater with sub info or on the other end, loosing the power of the sub in very specific spots of my video that would loose its impact with out the sub.
Okay, I will copy/paste from the answer on the other post:
"Okay, cool, thanks for asking.....
1. Dialogue, when measured for film, should be measured with a loudness meter and hit between -24 and -30 ON AVERAGE....so, some EQ and compression would help smooth it out (compression settings 4:1 ratio, threshold at -25% gain reduction as per the red, backward indicator lights). Film is between -27 and -30 (or should be for a great balance) and TV has to aim for -24LKFS.
2. Speakers should have -20dBLKFS Pink Noise on the timeline to equal 82dBSPL through the system (79dB for small room and 85 for a large room, with 82 being somewhere in the middle...I prefer 82dBSPL). Sub included.
3. If mixing in a small room, you will have some 'nodes' where bass is quiet, and 'anti-nodes' where bass is loud. Play some low frequency tones and/or soundtrack elements through your system and walk around the room and notice what content drops in volume. Double check with flat headphones.
4. Do a listen with flat headphones and compare with what your speakers and room tell you.
5. Drop other favourite movie mixes on your timeline and compare your mix with both speakers and headphones. You can get some samples from CZcams that were hopefully digital copies so the mix was not altered in any way.
6. Follow the festival website guidelines very carefully, but -27dB LKFS should work very nicely for dialogue, with loud parts hitting -2dBFS
Try this and let me know, or if you have further questions, just let me know...cheers!
"
Hello, this is very useful content. I would like to make CZ subtitles and so on. Where can I upload it?
Czech subtitles? I don't know the process, but I could post a Cz transcript.
CZ subtitles (unlisted video) and downloadable SRT file in the link in comment. Thank you for adding: czcams.com/video/WXnKeBzF_O8/video.htmlsi=F_PvKECuVpcri3Lg@@bacontrees
Wow awesome
this is why i mix at very low volume and it will fix all dynamic range problems... ..
What levels do you mix at? Please share, it would help to know....cheers!
@@bacontrees my mixes always between -20db and -24db because i use K ystem { bob kats meters such as k12 k14 k20} .+ i fader level automation to low down high dynamic peaks to keep every arragement section almost at same level.. only way to know your loudness using RMS meters not peak. also you must mix at very low volume because it is so much easier for you to hear the different which audio track is louder to have perfect balance . i only filter to remove resonance frequencies on each track and i never boost only in sub group processing using parametric eq for tonal balance to have better stereo image. it is not about plugins through it is all about balancing . good luck
@@aviozstudio4903 Cool! Thanks for the info! Very helpful.
Perfect! =)
Thank you Buck for your video.
I work in my little office (bedroom size) in my basement. I mixed in 5.1 and encoded in Dolby, made a Bluray copy and showed in 3 different theaters and it worked very well the 5.1. However, I need real help because the truth of the matter is that I am never sure if my tracks are too loud for the theater (and perhaps the projectionist had to bring the volume down).
My levels were: dialogue -6, effects up to -9 or even to 0. Ambient and Foley -24. Is this OK or not?
Second question, I am now remastering for DCP which uses raw PCM and no compression at all. Would my levels be too loud for the theaters ?
It will be awesome to know how to do it right. I want it to make it perfect for DCP and I cannot have clear answer. Thank you all.
sorry I missed this - are you still working on it? Email me if you are...cheers!
Cool insight! Im starting on the XboxOne GrandTheftAutoV tho. Thanks!
I always watch this video from time to time, very informative.. I am currently working on a 7.1 mix in a very small room (approx. 3.4 x 4), i calibrated my speakers to have the L R Ls Rs Lc and Rc on 85dB and the C on 82dB.. I did so using pink noise measuring the RMS, individually calibrating each speaker. Sometimes when I'm adding a loud SFX, it appears to be super loud.. Is this normal ? I'm pretty sure the peak level is sometimes way over 85, almost painful.. and I never go up more than - 3db on my digital levels.
This will be a detailed answer....sorry to miss this...stay tuned!
awesome tips thanks
Very nice video. I am a beginner (I almost know nothing!). In a 7.1 system, if each of three front speaker can go atmost 85dB (It will clip if we put something louder than 85dB), and each individual array of surround channels can also produce 85 dB max ( I mean 85dB max for each surround CHANNEL). Then the only way to give a quick 2 second loud effect of 95dB is by sending the signal to all three front speakers (or any three depending on the position of sound). Am I right or wrong or is this simply a very dumb question?
85dB SPL for a speaker is quite low, in fact, that should be an average loudness. There should be some headroom in any speaker reproducing a soundtrack, such as up to 105dB for a surround speaker and higher for other speakers. What is the sensitivity of the speaker (in dB)? If you have that, you can do an equation - (10log P1/P2) + sensitivity, where P1 is the average power or peak power, and P2 is 1 (equal to 1 watt of pink noise in an anechoic chamber, but let's just say it is always '1' here), then you can figure out the total SPL in dB that the speakers can handle. In short, 85dB isn't loud enough to handle peaks. Speakers for soundtrack listening should be able to reach levels higher than 85dB (for transient FX, etc).
Does this answer the question? Cheers!
I'm so fascinated by cinema sound and still new to sound mixing. I have a lot of questions :p You mention comparing our favorite movie sound mix with our own when mixing our projects. Aren't DVD's sound levels completely different than theatrical releases as in the case of TV Ads or Internet content? Lastly, the title says "independent films" so are studio films loudness different than indies?
Great questions! Only some movies are re-mixed for DVD's, but even the mixer on The Dark Night said they didn't change the mix a whole lot...only tweaks here and there but essentially the same.....and when I measured IMAX and DVD, there were very close. TV ads are much more compressed with sometimes less than 5dB of dynamic range, and the internet is the wild west when it comes to standards. Indie films and studio films are the same, or should be....in fact, some studio films are overly dynamic, while some indie's are much more comfortable yet effective. I stated 'indie' because studio people don't talk much about it and they don't always have 100% creative control over the mix, whereas indie filmmakers often do. Thanks for the comment! I hope I have answered fully.
Also, I have a few more videos on this subject, comparing other films on a video timeline. Cheers!
Yes all questions answered. Thank you for taking the time to reply.
Thanks for asking! DVD's & Blurays 'can' be re-mixed but sometimes aren't. Even if re-mixed, sometimes it's just small tweaks to keep the original theatrical mix pretty much the same. When I compare some films I have seen in cinema with BluRay/DVD, the mix is very close, almost not noticeable. The dialogue may be brought up a few dB, to be more in balance with loudest FX. When I measured The Dark Knight in IMAX and compared it with DVD, it sounded almost the same, level-wise, when I actually analyzed the DVD wave-forms.
I aimed this video for Indie filmmakers who needed a starting point and a bit of theory to be able to eventually mix like big budget filmmakers, as the latter are working their ways up in studios learning from the veteran mixers. But, the big budget teams can mix in elaborate studios and dubbing stages, or actual cinemas, with up to 3 people. So, I wanted to shed some light on the topic and hopefully help people understand the goals of a soundtrack mix, and be able to do it on smaller NLE's and DAW's, in their small project studios or homes studios.
Let me know If I have answered all questions with clarity, and thanks for commenting Cheers!.
Thanks what is the plugin called
Is there a part 2 to this? and a part 3 and 4 and 5? I think I need them.
great video.
Yes, there is a Part 2 coming very soon...probably next week. Cheers! If you have specific questions before then, I'd be happy to answer.
bacontrees No one ever shows video of the Master channel in protools during playback when various parts of a mix are playing. It would help to have a visual example of what properly mixed dial, sfx, music levels look like at various parts of a film on the Master meter and to hear you talk about it with regard to TV specs v. Theater specs. Are you using limiters to achieve specs or hovering around the suggested levels. There's tons we need to know, and we need to see how it LOOKS in protools. Thanks!
Hey, I use Vegas Pro...same concept, though, where the master audio window should be shown. I just started working on that video and am continuing today. I use limiters at the top around -2dB...my dialogue sits around -22dB....for cinema, there is a little more dynamic range, mixed to taste, really, but using dialogue as the 'anchor element'. On TV, the LKFS (or LUFS) meter can be a VST plugin to make sure your mix averages -24dB +/- 2dB...the spec is similar for UK TV...more on this in my next video...thanks for the comment!
Thank you very much for sharing this video!
Is this great little tool you developed available as a plugin for download?
Thanks for the comment! Not a plug-in, but a process...I put a soundtrack through a process within my editing system to simulate cinema sound playback as well as compare with different room characteristics.
sorry, what is the actual reference specs to relate to, and what is the main institution that fixes levels and any other parameters? Dolby? Ebu? I hope my question makes sense,
There are no actual specs for film, just guidelines as it is an art to mix films. However, there are for TV. But the guidelines for film sound can be found here:
czcams.com/video/lf5it1yEmLI/video.html
Just wondering, when calibrating speakers and DAW to 85dB SPL, do we normalise that pink noise file to 100%,
and with the track it's on, set that track volume at 0.00 db?
The pink Noise can be downloaded as -20dBFS or you can render raw pink noise to -20, but some DAWs have the setting, so no normalizing is needed. You can also try a 400Hz tone set to -20 and set speakers to a comfortable level. Other sources have recommended even using a 1kHz tone.
@@bacontrees Thanks Bacontrees. Hi from Australia. I really appreciate your reply. So we put a -20dB pink noise or other appropriate tone onto a track that is set at its default of 0.00dB. Then position the dB meter on a stand in the sweet spot, and measure each of the 6 monitors separately, calibrating each to 85dB with the surround volume manager plugin (I have the Waves M360 one). I guess we use a 5:1 noise file on a 6 channel track for surround.
I usually set the Ls and Rs at -3db Less.
I've been confused in the past about what internal attenuation the actual noise/source file should have. You may have cleared that up here so thanks heaps. If any of the above is incorrect please do correct it.
I’m doing features these days, and sometimes they don’t have money for a mix in the proper space, so I always manage to talk them into at least hiring a cinema for a viewing. I take notes then come back here and tweak. Better than nothing!
@@quatermass8 Thanks for the query! In a smaller room, I would suggest 82dB for the front speakers and 3dB less for surrounds. Even smaller rooms would be 79dB for front speaker, but I think 82dB would be a good mid point. Any test tone or pink noise is best at -20 on the timeline. In fact, try pink noise, 400Hz and 1kHz and compare all 3. I have always questioned why different sources say different things.
BUT, I also believe in putting a well know film soundtrack on the timeline and compare with that, if the dynamic range isn't too out of whack. And, yes, each speaker should be tested separately.
Thanks again, and this comment makes me want to update this subject, as I feel I have left some details out. Cheers to you down-under! (I am not crazy about that expression, BTW, as I saw a map of the world with Australia up top and it changed the way I see the world.) Anyway, let me know if this makes sense as it has been a long day of teaching and soundtrack editing, and I am burning the midnight oil...(one of my fav bands, BTW!).
@@bacontrees We're very okay with the term Down-under. It would be impossible to render it insulting in any context, don't worry !!
But I know what you mean. Our solar system is hurtling around the galaxy head on at a 60 degree inclination to the galactic plane, then all galaxies are topsy turvy to each other so ultimately there's no up or down.
You advice makes sense and it will help in the next few weeks. Thanks so much for your time. Have a good 'rest of the day' !
@@quatermass8 Cheers and thanks!!
Hi. Can anybody or you, plz help me/others with a unique topic, not explored very often:
making a film w/out audio crew?
Here's my thoughts so far:
For dialogue,
1. Give each actor a hidden ZoomH1 and their own wired lav. OR
2. Place Zoom H1's or lavs hidden on set, or just off camera. But option 1 is better i assume, no?
Other ideas/reactions?
Next, For the sound effects
1. You could "plant" lav mics in strategic places,
2. You could "plant" field recorders by themselves in strategic places,
3. Or tell me what you think of this idea:
what about doing a take twice, and the second time not doing any video, but repeating scene
as exactly as possible and having lavs attached to actor's legs (for footsteps) or arms (for doors etc.) unconventional i know, but could it work?
any other ideas? thanks for any help. i think a lot of this could benefit from creative ideas regarding running sound w/out any crew.
thanks again!
rory
***** - sorry, I didn't see this...it went to 'awaiting approval' and I missed it....too late to help?
not too late. i'm still curious, as others might be. any thoughts would be appreciated. thanks!
1. The Zoom H1 with wired lav sounds good, as long as a slate is being used and Zoom files recorded onto a 'sound report' so you can find and sync files later - the camera should have it's mic on so the waveforms can be matched for perfect sync.
2. I prefer to do sound FX in post, at home, to have total control....the reason is that dialogue and on-set FX cannot always be separated. I have a Foley FX at home video that can give ideas.
3. Recording 'wild lines' (dialogue recorded without video) is always helpful....find a quiet room (or closet) or the same shooting room if quiet. That way, you can avoid/minimize and ADR.
4. Don't forget to get some 'room tone' - about 1 minute of the set alone recorded for filling in gaps when editing dialogue (aka 'fill', which can also be found from clips with longer sections of no dialogue in a scene).
Sorry for the delay, been very busy as an instructor of filmmaking and audio and haven't had much time to respond...but I made a point to do it today...let me know if this helps!! Cheers!!
thanks so much for your very thorough response! much appreciated!
No problem! Hope it was helpful.
Hi , Great Video ! Thank you ! At around 7:30 you talk about the cinema Emulation with Reverb and aFlat pair of headphones ? Would you be able to share some screenshots of reverb settings ? And also I am not sure about my Headphones, they are sennheiser 650's .. I saw earlier in your presentation , how you flatten the speakers ... At what point in my Channel Chain Do I doo this ? ie. last or without any plugins turned on ...Could I do this on Headphones on Master channel and then apply reverb Plug in afterwards ? I tried to find any Emulation Plug In that could do this but diddn't find one, any suggestions ? Hope this all makes sense. Thank you for your time !
Yes! Apply the X-curve EQ, then the reverb at .85 seconds to the master FX bus. Only to listen to the result of what it would probably sound like in the cinema, then take off the reverb and EQ before rendering the final project. No need for an emulation program, just do the EQ and reverb to compare original signal to modified signal.
bacontrees brilliant ! Will do thank you again !
Buck, I have this nagging question about the different monitoring/calibration levels quoted for different room sizes. For example, the 'standard' is 85 dBC, but I've also read/seen directions to use 82-83 for mid-size rooms, and even 79 dBC for small home studios. So, I guess my question is, why use lower values for smaller rooms, even with nearfield monitors? Isn't 85 dBC still "85dBC" at the listening position regardless of the room. After all, that IS what you measured from your monitors when calibrating them at the listening position, right? One video said that for small rooms, using 79 will ensure that your mix will fit into main mix at larger studios. Why calibrate to anything other than 85? I don't understand that rationale. Cheers!
+Shane Taylor Hi Shane, other sources, such as the Dolby Surround Mixing Manual, had stated pink noise at 85dBC for every speaker for cinema style mixing or 83dBC for mixing in a smaller room where speakers are closer to you. Then 79dB for home video releases so dialogue will be more consistent.
It has to do with a couple of things - namely: high frequency attenuation in the air in larger space and perhaps psychoacoustics. 85dBC is what cinemas do, but it is not an exact science and as long as the speakers are close and the dialogue plays back at normal levels...and we compare to other mixes, there is a tolerance range where most people won't be offended by the mix. I tried 85dBC and then compared to others with very good results. Cheers for now!
As a production sound recordist there has never been much explanation what happens to the audio I record after I hand to over in terms of level.
I had a scene the there day with such whispering that I had pushed the gain very high. to be able to get an average level of between -20 and -12dbfs or PPM 4 to PPM .6 Now when monitoring off the 664 at 0db/line It gives a horrific sounding back ground noise and a very live sound, you can hear the room a almost too much.
However, turn the headphone monitor down to -12db and it suddenly sounds lovely, clear dialogue and a a dry silent background.
So my question is, how is the audio standardised so on TV at home or the cinema you don't hear the noise floor as you do when monitoring production sound or the reflections of the room? Comparatively what level is it set to? how do the levels transcend?
great question - Iclean up the background noise in of of 2 ways - use a noise gate (if the signal-to-noise ratio is good) to get rid of the noise between dialogue chunks, then add room tone just under all chunks of dialogue to make a seamless background sound.
Or, if the background noise is very steady, I can do a noise removal by sampling the noise for a few seconds (ideally) then removing the noise form the waveform - but, this is tricky if the noise floor is fluctuating a lot in tone. Then room tone is added under all dialogue so they sound like they are actually having a conversation in one location.
Outside dialogue, well that is generally more work but ideally the voices were recorded as close to the mic as possible to minimize the BG noise.
But how are the levels standardised for broadcast? Monitoring the levels only sounds 'normal' when you reduce the recorded dialogue by -20db. I suppose this is because at this level you are not hearing it at an amplified level. and then through a TV set levels are generally quite low anyway so the high Signal recorded is reduced anyway hiding all the artefacts.
I persoanlly hate cinema sound... far too dry
The overall broadcast signal would be measured with a loudness meter and would have to be, on average, - 24dBLUFS +/- 2dB.
So, I record my dialogue as loud as possible without distorting (aiming for - 12dB) then would mix it at - 24dB +/- 2dB, with the loudest sound FX to - 4dB or lower - depending on the style of mixing transients (I just put a new video up about this).
Also, since I usually don't use any compression on dialogue, I would EQ, then compress slightly for more controllable levels and parity (then, I would do a print master to tweak levels of words and sounds here and there). I hope this helps. Cheers!
czcams.com/video/Rcn1sMFiCOE/video.html
I don't know why there are lines through words where the '-' minus sign should be...oh well.
New Dune film brought me here.
Nice to know, thanks for sharing!
That is Dolby 7 level. If I put an effect at -2 level, then it will come out at 105 - 2=103dB. Right? And according to that, the dialogue should be mixed at -45dB (105 - 45 = 60dB). 😨 Isn't that tooooo low? Even if I have a loud 85dB scream, still I need to mix at -20 dB. Please tell me if what I wrote is right or not.
Not exactly - dBFS does not completely correlate with dBSPL. This is because the acoustic space and loudspeakers (the B-chain) tend to add SPL. For example, one movie I measured had dialogue scenes mixed at around -35dBFS (with an LKFS measurement in Orban Loudness Meter). The acoustic result was dialogue was just under 60dB and loudest sound FX measured 105dBSPL A-weighted, and 110dBSPL C-weighted (C = 'bass included'). So basically, the dBSPL added 15-20dB in the B-chain.
So, you can just mix dialogue around approximately -30dBFS (using calibrated near-field monitors) and you will get more loudness during playback in a larger acoustic space (depending on what is in your mix, of course, as some people don't reach -2dBFS with their mixes). Hope this helps!
bacontrees Wow! That was a quick and perfect reply. Thanks. That really helped.
Awesome! Thanks for the comment, I really enjoy working in the art and science of soundtracks. Cheers!
whats the name of this program he's talking about
Hey there...it's just a couple of settings in any editing program. It is a pre-set I came up with but not a program by itself.
thanks for the reply! is there a way for one to get this preset? i use adobe premiere, and also do they mix the sound twice, one for theater and for dvd?
Just use any graphic EQ and set to 'X-curve' (3dB/octave roll off below 63Hz and 2kHz), then a .85 second reverb. And some movies are mixed for both theatre and DVD/bluray, some are not. BUT, the mix should not have to be changed too much, only tweaked a bit for dvd/bluray.
hi ,I make some shorts and i would like to ask you which is the best way to write audio of a convertation of actors?the boom shot gun or the wireless jacket microphones?which is the advantages and disadvantages of these two ocassions?from a shot gun i noticed that the sound is more normal(as it concerns the deepness of the space) but i lose quality when i am far from actors and earns environment sounds .on the other hand the wireless mics are good but it looks like the actor talks next to your ear at post production .what is your opinion?
Great question! I prefer to do both if possible - shotgun and wireless. My shotgun mic is long and has a great reach, but the wireless lav has closer sound, less ambience.
Shotgun mic - must be as close as possible to actor(s), room must be as quiet as possible (some room ambience is fine, because there has to be room sound, or 'room tone', anyway to sound natural). If an actor can't spit on it, it is too far away. I would always dip the mic into the frame to establish a frame-line, and just stay right above the frame-line.
Wireless lav - if no wardrobe rubbing or RFI/EMI is happening, and the sound is clear, you can always add some fake room reverb to give it proper 'sound perspective' and make it sound more natural.
bacontrees thanks a lot for your fast and excellent answer.now propably i use a mix of the two ways!
Cool!
can u upload a video of how mixing the audio on a film (practically) thnaks
Sure...I will figure out a screen capture or something...great idea!
great , i wish no do this not just in theory , but in practice... thanks a lot
Hiya bacon tree. How would you say the levels are just as you see them on the mixer. What level would dialogue, music and effects be. And what range? I heard dialogue is somewhere between -6 to -12 dB. But I'm not sure if I'm correct.
Hi there, although dialogue can certainly sometimes peak at -12 and up to -6dBFS, the average for a film would actually be -27dBFS, down to 33dBFS, but more accurately, those average levels would be measured in LKFS (Loudness Units, K-weighted, as measured with a loudness meter). TV has been mandated to be -24LKFS, +/-2dB in North America. Internet dialogue is often very compressed and made as loud as possible.
Since dialogue can be very dynamic, it is suggested that it is EQ'd and compressed to be less dynamic, then the peaks would be even less that -12dBFS. I made a 'sound map', seen in this video ( czcams.com/video/ft13z53P0ks/video.html ) at about 2:25, which shows the levels of everything. I am working on a more detailed version as a handy reference.
Loudest FX 'transients' can reach -2 in a film, and ambience can be as low -40. Foley FX and music should be mixed around dialogue. Let me know if this is helpful, or confusing.......cheers! :-)
Oh, and as for the mixer, the levels would fluctuate evenly above and below the target point (like as much dB above the point as there is level below the point).
Oh, and TV for Europe is -23LUFS (same thing almost as LKFS) at +/-1dB.
bacontrees thanks. yeah I'm a sound post production film students and we don't have treated rooms so I jjust have to rely on the numbers I can read on my protools sessions. so thanks for the info.
You can also use headphones if the room is too reflective, and also, what I do is sometimes get the soundtrack of a famous movie and put it on my timeline to compare. Cheers!
Great!! Thank You!! :)
What exactly is this cinema simulation? software? Can we find it somewhere? Thank you
Oh, that is putting an X-curve EQ and warm reverb on the master output temporarily to compare what a cinema may sound like with your soundtrack playing in it (without actually being in a cinema). It's an approximation and would give the worst case scenario. Then, before rendering the movie, the simulation is take off the master output.
thank you!
What about mixing for documentary?
Documentaries would best be mixed at TV levels - with an average of -24dB on the digital timeline. I measured an Oscar-winner, though, and the dialogue was more like an average of -20db (LKFS - as measured with a loudness meter).
-24, where a feature film would be -27 to -30 average on the timeline (so, -27 to -30LKFS, when measured).
if Dialog is at 60 dB and transients are 100 that's 40 dB and you've lost the ability to make anything such as foley a lower volume while keeping with the original stated 40 dB of dynamic range.
Foley can be around 50dB or just a bit hotter sometimes. Also, the transients are only 100dB when played back in a cinema or loud home theatre system. The DAW shows close to 30dB and the room adds the rest, but mostly C-weighted. When I analyze The dark Knight electronically, the dynamics, on average (from dialogue to loudest sound FX) are just under 30dB, but acoustically, around 40dB. I actually sat in the IMAX and measured at the optimal distance from the screen and it was over 40dB for the loudest parts. I confirmed the tests by playing on various TV's/sound systems and measured again.
bacontrees at 01:02 the measurement sheet from IMAX can be seen for Dark Knight. Matrix Reloaded is shown on the green SPL Graph periodically.
but you don't have a cinema master of your fav movie only the DVD master, which is probably not the same?
Actually, not everyone has the luxury of re-mixing for DVD/Bluray. Also, I did read an interview with the mixers of The Dark Knight, and they said that although they had the luxury of a re-mix, the re-mix was pretty much the same with the exception of a couple of tweaks. What I tend to do is watch a film in cinema, buy the DVD/Bluray (if the movie is good), and compare my cinema measurements with the discs and the the findings are the same. I've done this a lot. I am the nerd with the sound meters in the drink cup, to the annoyance of my friends. :-)
Hey,
I am new in film making. I want to have audio similar like we have in Hollywood or Bollywood. The films made here in Pakistan don't have that dynamic sound like we have in Hollywood &Bollywood. Kindly suggest me sound equipments.
It all starts with great microphones to record all sounds required, followed by a sensible mixing strategy. The mixing equipment needed would be a good set of near-field monitors (or far-field if in a large studio), a good set of headphones to double check and compare, and a decent editing system (I use Sony Vegas Pro 11 right now, but Final Cut Pro, Premiere or Avid are also very good).
But why don't we get the same effect as we get in films. The sound will be clear but it will not look like we are watching a film with that audio
I'm not sure I understand....can you give me more info? I'd love to help.
bacontrees I am saying that after mixing sound we didn't get the dynamic sound which is required in a film. It more looks like a drama show going on. Suppose in hollywood we are shooting a dialogue. It will sound different for a show ,film and drama respectively. We have to make difference in them by sound and picture quality. Now coming to the question that If I am saying an angry dialogue. So how it can look more dynamic in film.
Okay give me few tips which I can apply while film making
Is there any other information about the x-curve? Maybe I missed some details somewhere, but I'm very curious as to what that would look like!
+Luke Covert - yes, right at 3:00 a large picture....I will also being doing a third video with the sound of it....cheers!
Why does everyone talk in listening volume instead of digital level? Yes, 85 dB should be -20 in the DAW, but why are we always talking in relative volumes instead of absolutes? Why do we have to be so vague?
Fair question - the DAW can't give us any indication of 'perceived loudness', only electronic levels. And, there are simply no absolutes, only averages; the average dialogue, when measured A-weighted, comes in at about 60dB (I design my tracks that way and I have measured countless cinemas and TV', even live speech and 60dB seems to be the magic number). BUT, when dialogue is EQ'd for enhancement and compressed just the right amount (depending on the source material), bringing the level a few dB more tends to sound the same as when not EQ'd, only clearer. The dialogue EQ curve actually affects the dynamics thus maintaining perceived loudness. So, dBSPL measurements, for the end user, is most important when playing back in a room.
Just so you guys know....there is so much more to this. Don't think after watching this video that you know everything. There is so much more complex analysis than you think. You need to do a lot of research. The information, unfortunately, is so scattered and there are no finalized set standards yet.
I know, there are other videos I did that follow with lots more info....I put this up because nobody else was at the time. I know there are no standards but the info is solid. This is an introductory video with actual measurements and results. This does not cover the entire gamut, but is a start for further investigation. This is part 1 of a series of videos on this subject.
bacontrees I agree, besides this video is setting the basics and is more than enough in order to get anyone started. After that comes the experience, nothing you can put on a CZcams video. I believe at some point it becomes more of a gut feeling :) cheers
C'mon man, where's the next video ??
I just haven't been around to finish it...working in and out of town doing video, audio, etc...on a dizzying schedule.before I do, what would you most like to hear about?
I did 2 more and another coming this week. Cheers!
Saying that conversation have to be at 50-63db????... - that's a total nonsense.
This is amateur not a pro audio, very confusing explanation of the subject.
In to words: from -12 up to even -9 must be your very highest level, the limit/peak of loudness. That's your "jump out of your seat" moments. And based on this limits you "build" your overall sound stage, your medium levels, and softest passages. That's your taste and creativity.
Dynamic RANGE is much wider in film than TV.
Small rooms calibrated using the pink noise to 85dB SPL.
Actually,
this is professional, and I don't recall saying dialogue should be as
low as 50dB SPL, (ambient FX are usually around there). It is 58-63dB
SPL on average from all the cinema and TV measurements I have ever
done (countless measurements) which is generally where normal
conversational levels are (as measured many, many times by myself and
others). I have also mixed for all media. Dialogue is mixed for TV
around -24LUFS, as confirmed by the Orban Loudness meter I use (grab
any random TV show and put it to the test) (-23LUFS for EBU).
As you stated: "Dynamic RANGE is much wider in film than TV."
- Yes, of course, just as mentioned in the video. "Small rooms
calibrated using the pink noise to 85dB SPL."
- Yes, I know this, but smaller rooms can also be calibrated to 79dB
(according to the Dolby Surround Sound Mixing Manual, et al).
Jump scares can be anywhere under 0dBFS, actually, for film. BUT, the
longer they last, the lower they should be. -2 should be the highest,
but I personally think (and agree) that the loudest peaks should be
around -10dBFS. I am just revealing what I had measured. I just
analyzed a scene from CSI and the jump scare (gun shot) was -2dBFS,
where the dialogue was, on average, just under -20 on the timeline
(measured to be -23.1dBLKFS in Orban).
This is one of a series of videos on the topic. Later, I get into EQ and
compression for dialogue to make better use of the dynamic range of
the timeline (NLE, or DAW). This is not theory, these are actual
measurements with professional tools, so explain why you disagree in
order to have a conversation about it. But, it appears we do agree on
a few things. Cheers!
@ 1:28 - "Anywhere from 58 to 63 DeciBels..." I spoke fast, but that is what I am saying.
But why you keep the dialogue SO LOW for home/Stream releases? Its not fun watching movies and constantly tuning the volume! The Batman 2022 is atrocious for its bad dialogue mix. Do I really need 5.1 system to watch movies now? Soundbar isn't enough?
Dialogue for feature films in cinemas should be mixed at -27LUFS and FX True Peaks at -2dBFS. When playing back, dialogue should be at conversational levels of around 60dBSpl. Dialogue at less than -30LUFS is generally too low, but especially when streaming movies.