You missed a BIG one imo. Reduce saturation in the highlights, and shaddows and up the saturation in the mid tones. Film desaturates in the highs and lows
i agree with you on the overuse of the "cinematic" term. Every single video that i see on yt claim that they will give you a cinematic look just by having a bokeh and slowmo color corrected footage... and now we see this "cinematic" footage on every single commercial add on the web
Blech. Not to mention "cinematic" referring to the horrid orange and teal trend, etc. It's generally synonymous, as used in pop culture, with bland and generic.
First I thought it was the super cheap ear plugs I found yesterday on the street. Yeah, I would have fixed that in post. Even something as simple as Audacity would do wonders. But time, etc.... A number of ASMR videos here on youtube could really benefit from rolling off the bass in post. Very distracting when you're trying to lull people to a calm place.
I was literally going to say 'the first thing you should do is mix your audio'. But I thought he was going to change it and make a great point... and then he didn't....
Excellent post. Really emphasizes why this quest for overly-sharp Video is really a not helpful in creating truly non-video looking shots. I intentionally focus on adding a little softness to my shots and look for the most natural skin tones out of camera, but it seems everyone is on a quest for overly-sharp, fake-looking 4K these days.
Rolling shutter also contributes to the video look. I am surprised how people don't notice the not so subtle motion rendering differences between rolling shutter and global shutter. Even video cameras with the fastest sensor readouts such as the Arri Alexa has a degree of rolling shutter that gives this unstable motion rendering look (albeit much less than other rolling shutter cameras). I can instantly tell which movies are shot on film and which movies are shot on an Alexa from looking at how smooth the motion rendering is. Once you know what to look for rolling shutter becomes very noticeable, even on an Arri Alexa. (I am not referring to the obvious rolling shutter effects from panning and fast motion, I am referring to all motion.)
Thank you for the precise and fact based explanation as always I learned from your video, the way you broke down ideas sticks on brains! Keep the good work!
Great tips and well thought out. I always thought Nat Geo photographers/filmmakers shot on some crazy special purpose cameras. Thanks for debunking that. It's all about chasing the light. Your channel is one of the very few legit filmmaking tips channels. Kudos to you!
It's all about chasing the light indeed. I've never seen another medium where you could point to one factor that was so important, usually it all matters, but badly lit images really can't work.
What I've noticed is that on video highlights are thin looking and cyan tinted while shadows are cloudy, noisy and warm tinted. Its like a colour temp shift across the dynamic range. This would be comensurate with harmonics in audio recording; A trumpet getting louder doesn't really get much louder but it does produce a lot more treble information. So if a highlight had more upper end information it might appear more blue. This could be accounted for in the poor sensitivity of CCD sensors and operational parameters of tube cameras. Just a thought! :o)
Nice video! Can you make one on how to make video from a bit older full frame DSLR like Nikon 610 look as good as the new cameras good at shooting video.
4:29 omg.. this lines made my day.. i alway think like this... i mean if we saw some people without eyebrow.. our mind train to rember like that guy is ugly... but actually he or she is not ugly .. but our mind think that way... but u explain very well.. hope u guys understand what i mean.. my english bad
For motion blur I use the 180 degree rule. Basically film at any frame rate you want and choose a shutter speed as: shutter speed = 1/(2*frame rate) So if you’re shooting at 24fps as your base frame rate, the shutter speed would be 1/48th of a second or the next closest, 1/50th of a second. Then if you want to add b roll in slow motion shoot at 60fps or higher and then slow down by 40% to fit the 24fps timeline, or speed ramp depending on what fits the creativity best. In the 60 fps example I’d be using a shutter speed of 1/120th of a second. In either case using the 180 degree rule as a general guidance will give you good motion blur that’s not too much and not too little. To achieve this in all scenarios I use a very high end variable ND filter with step up rings so that I am not over exposed in bright conditions and keep a relatively constant iso in all situations.
Very useful tips, thanks! Why is it better to reduce sharpening in-camera and not in post-production? My thinking is that that if you capture unsharp footage with fewer details then it's more difficult to fix in post?
who thought the one at 2:35 actually looked better? The filter used for the majority of the video looked too grey and washed out to me. Now I'm even more confused
Not only skin tones, but I found that edges in the image also give away a video look. As suggested by another creator, I went into the profiles of my GH5 and turned the Sharpness down to -5. After doing this, and shooting test footage I was quite amazed at the improvement. It turns out that the default Sharpness settings in the GH5 are too high. Working at IMAX, we enhanced the imagery by reducing the grain and adding just enough sharpness to improve the look - to put it simply. But, when there's too much sharpness, the edges of the image show a halo effect around the edges, especially where there are specular hits. Reducing the sharpness may seem like a subtle change, but to me its actually quite dramatic. I could see the improvement in the footage from my GH5 without having to zoom into the image to look at the edges. Discovering this simple change was almost a spiritual experience in itself. Excellent video. Thanks for the tips.
I actually blur my GH5 footage in post and add diffusion in front of the lens even when the camera is set to -5. Not necessary on RED and ARRI cameras ;)
@@AllThingsFilm1 at -5 sharpening the image is still overprocessed. So some slight blurring helps with hiding sharpening artifacts and makes the footage looks less digital and harsh.
the problem is that I love the so called video look of GH5. A director once told me in many contemporary cinema, people are too much forced to look at one isolated item/person the director wants you to see. This effect is often done by using editing, (analytical versus non analytical editing), but I guess the camera is an important factor too. I can see when everything is sharp like GH5, people can pick up the thing they want to see.
Thanks for the tips Sareesh. I love your videos because they are "In Your Face" tutorials. I have the Panasonic tz90. Here is a question. Can you make a tutorial with live examples of "Avoiding the Video Look" with point & shoot cameras? I Find that that's the thing that is mainly missing with low-end cameras without log/raw profile even though that many of them have features that didn't exist 10 years ago.
My master as I fondly say here, I've been tackling some shoot and I'm getting better as a cinematographer. But I want to task myself on directing. Could you please make some practical directing video and goes into what directing is.. Camera height and coverage and when to use which in what situation warrants what... Want to learn from your experience boss. And I must say a big thank you for all I've gained and they're practical in real life situation on location thank Sareesh sudhakran (wolfcrow team).
Artificial motion blur may be only needed when recorded with less than 180 degree angle. But there is not good reasons to do that unless not knowing about the artistic purposes in advance/ND filter not at hand. Otherwise, let's say, 100fps recorded with 1/200th of exposure and slowed down to 24fps still has the proper amount of blur - you have to capture half of the movement on picture, our brain does the rest. If the object in frame won't move, we should not make it blurry - if it moves we always capture half of its path to the next frame with a 180-degree rule.
Is there DR settings and other settings on an android phone that I can use. I have got the cinematic look through online or apps by uploading the video. Are these still good resources?
Thanks for your great tutorials. What do you think aboat sharpening settings in camera according to noise reduction in postproduction. Is it good to set this parameter in settings (for example in sony a 7 iii it is named "detail") to 0 and have some "sharp bumper" for future noise reduction. Or is it better to set it to -7. Is 0 base paremeter of sensor and -7 is cooked by cameras processor? Or maybe -7 is base and 0 is result of sharpening by cameras processor? Is settengs of detail affect noise produced by camera? Maybe less sharp setting produce less noise? And we dont need to do with it in postproduction. And second question is about noise reduction settings in camera does it affect video mode or just photo? And if it does, is it good to set high low or turn it off? Is it better to do noise reduction by bad camera processor but before compression or by good programm and good computer processor but after compressing?
Following this logic, should be able to get the “cinematic” look - in studio - where you only need to control the lighting to ensure there are stops limited to a cameras dynamic range (I assume the Z6 is on the low end... 12 stops?) so shooting studio with a more expensive DR is probable a waste of money... by the same line of thought, shooting the - limited - ProRes RAW of the Nikon/Atomos Ninja is just fine... since it will deliver on the “skin tones” maintaining the Nikon color science (?)
hey great video, can you try to test how different cameras and cinema cameras register motion? I feel like the high end cameras like the alexa shoot a smoother 24fps while the sony a7 series are choppy, but I'm not sure. Please make a conparison and explanation, would be great!
You need more than just set the fps to 24fps, you need to set the shuttle speed around 1/50 seconds, if the image is too bright, you also need ND filters to reduce the amount of light hitting the sensor.
@@MarcoZhang689 hi marco we battle with this somtimes, i try to stay close to the native iso of the camera, using nd filters when we need to for example with our c200 we will shoot at our native of 800 and as it gets darker we will increase our iso until about 3250 or 4000 then add 2 stop internal nd filter, ive also shot some vid with our nikon d850 which has a native iso of 64 and some with our sony a7r3 which 100 native iso also i think the sensor has alot to do with it, im not an expert but i had to do alot of trial and error to get it right also what are you outputting to BD or online (but mostly i think its setting up the camera correctly) im always learning so let me know if you have any other ideas
Of course it is right that you are looking for motion blur rather than fram rate. However - how can I have a shutter speed of 1/50 but a frame rate of 1/60?
Ultimatly don't use iFrames like 50i, 60i ect. Probably everyone here knows but just in case to clear out the difference between "full frame" and "interlaced is the read out of the frame. Interlaced footage will almost always look videoish and they do come from old TV times. So avoid those as well.
Can you have a situation where when you are reviewing your work on you video editor it is more that acceptable yet when it is downloaded to the public on your online site it looks awful. Can a site such as RU-vid have even less HDR than what would normally be considered websites acceptable with other professional video editors/processors?
Even the footage of alexa or red look like wedding video they do something in post other than color correction to the movement of footage to make it look like film
Another and probably more important difference is interlaced video, no movies are shot interlaced. Also 30fps and incorrect shutter, you need 1/48s to avoid video look. It would be a challenge for me telling just because of skin color, i'm not that good i guess :)
Cinematic look 1.Increase DR, Shoot during golden hour 2. Skin Color, skin tone, color grade, color profile, camera preset, Lumia GH5 good camera, low contrast setting 3. Motion blur, not increase stutter speed, ND filter, post production add motion blur 4. Soft setting, no sharping
Many very good tips here, and in return I have a tip for you. When you are doing lectures like this, the audio is important. I'm not talking about the volume, that is fine, but the quality of the voice. Using a lapel microphone without adding EQ is not how it should be done. Lowering a bit around 400 to 700 Hz and increasing a bit in the higher end, lets say from 3000 and up to 5000 Hz would make the spoken words much more easy to understand. Or even better. Skip the lapel completely and use a (super)cardioid microphone just outside the picture frame pointing at your mouth. No EQ needed, and perfect sound :-)
The good thing in post is you can control it rather than it being added in-camera. Like you can maks out areas of sharpness you dont want to what you personally want.
I’m doing live music shows . Working on creating a filmic and or cinema look . We want this over the high res action look of what most are doing for action and music . I realize most people in this field want that sharp 4K look . (We) don’t . I just bought a bunch of old Nippon (Nikkor) from the 60s and 70s and a couple ais lenses. We will attempt to control flare the best we can. Do you think these lenses will work for us ? Will this help with the cinema look ? We are going for a vintage look as well. Not sure if we should gear that look to the (blue) spectrum of light . Like the oldest Nikkor lenses will have . Your thoughts
1:31 Technically - not always. BMPCC4K has fixed 13 stops, but stops for above and below middle grayy will vary from ISO to ISO. For 100 ISO You'll get just 2 stops above, but 11 stops below. On ISO 1000 you'll get 5 to 8 stops. Also - dual native ISO is factor here, 1250 resets the dynamic range.
Depends on the camera. I leave my saturation at normal settings so my sensor is getting the full range of colors, i can always take some out in post but adding is post looks bad most of the time
I don't know if that makes it less "video looking" but I do tend to like how that looks with my Sony cameras. To me, it seems like the highlight rolloff for most gamma curves in Sony cameras over-saturate highlights so I almost always desaturate them.
@@makepaladinsmehagain8493 Yes I have the same with my Panasonic. These digital sensors seem to try and push as many colors in the highlights and shadows, untill they are completely white or black. That's why you always see a very unnatural yellow ring around bright lights like the sun, it goes in a very harsh transition from yellow to completely white. I read in an article that the Arri Alexa only reaches maximum saturation at about 35IRE and starts to decrease saturation again after 65IRE. I've been trying to simulate this look in resolve and got good results so far.
@@Crawling8888 No I actually leave the camera saturation at zero so the footage gets the full range of colors. It's better to have them all in camera and maybe take a bit out in post than take them all out in camera and find out later you'd want more saturation. I desaturate the shadows and highlights in Resolve.
@@Aviciiz Thanks for your reply! And which picture profile do you use? I got the GX85 that means that I can only use standard profile with decreased contrast or Cinelike D to achieve a flat footage with maximum DR. Unfortunately there's no log-profile...
Having only my experience with the first Canon 5D Mark II, even way back before the manual override software became available, I went through the learning curve of shooting with sharpness on, (landscape) and then trying it off. But I was frustrated in that if shooting a grassy field with the sun behind me, for best contrast, is the grasses never looked sharp either with sharpness on or off. It wasn't a depth of field or frame rate question. But I knew even the non L lenses for Canon were sharper than what I was seeing on my monitor. Thus, the experiments with sharpening. Back then noise in the shadows was pronounced, so, I would shoot with tripod, then in addition take stills of the scene, then layer the still over the noisy buzzing shadows and be able to control about 80% of the frame that way. A simple solution using Photoshop CS 4 Extended and it's animation timeline. For example, my short film, Bum Rap. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-PH7vdg8m9MY.html Note the alley at 7:43.
I'm not sure it's difficult to notice what the "video look" is? At least partly. All the examples you showed of television feature studio sets and harsh, high kelvin overhead lighting, none of which is really found in real life. Films tend to have much more lifelike lighting, and if it's not lifelike they do all they can to make it look lifelike while exaggerating to whatever extent they wish.
