Leaderboard

There are several variables that combine in skintone; color separation, saturation model, skin color, light color spectrum, white balance, lens coloration. Color Separation: Good cameras have color separation. Less ideal color science will have 'accordioned' hues, compressing and expanding arbitrarily throughout the hue circle. For skintone - under quality light, the Alexa will have great separation between the pink aspects of the face (lips, blush) and skintone, which rides the line between too green and too magenta under flat spectrum light Saturation Model: Photometric saturation models (like Alexa LogC+R709 or EC LogC+R709) have less saturation (the video model). Saturation amplifies hue difference, and photometric models can make skin tones seem more alike. The EC variations emphasize saturation (film backend), which foregrounds skin hue Skin Color: The skin of the talent may be slightly more olive (dark and yellow in hue) or pale with pink as the primary hue accent (all vary towards magenta with emotion). Here is a recent still from the film I've been shooting, with two people in the same frame that had divergent skin hues: Light Color Spectrum: On the Alexa (and cameras generally), filming under light that has a spikey color spectum (Leds, Flos, Sodium Vapor, etc.) can cast skintone distinctly to green or magenta, even with correct - White Balance: A fully neutral WB is vital for EC, not only in skintone, but the whole of color space accuracy. Skintone is one of the reliable indicators that WB might be off on the green/magenta axis when under reference light without a white card reference (should anchor on the skintone line of the Vectorscope under sunlight and halogen). A common occurrence is that WB will be set for direct sun on the GH5 (possibly with the WB sun preset), and then the shoot will move into the shade, which will shift the WB towards blue/magenta (needs a new WB, or correction in post ahead of the conversion) Lens Coloration: Interrelated with WB is lens coloration, which is like a WB that variably shifts throughout the grayscale [For EC: it is optimal to WB around middle gray, or a little brighter for the most salient part of the image]. This can vary from minimal coloration (Sigma 18-35) to heavy coloration (for example, Xeens). For NDs - the reference EC ND is the Firecrest FSND, which evenly cuts the full spectrum (visible/invisible). Non FSNDs (even IRNDs) will have a varying ratio between IR/UV and visible, which can unpredictably shift sensor response to a given light (interacting with the camera sensor's native cut). Variable IR ratio particularly acts on the green/magenta axis Notably, EC Tungsten does have a slight magenta bias in the skintone region towards magenta compared to the Alexa. I've figured out what in the data was causing this problem (IR leak every third exposure slice), and this will be solved for the next release.

Hi Everyone I finally managed to get my first short underwater clips with the GH5 And I am using Cine Like D and the results are pretty good. However those videos are with rich ambient light in other situation I need to go in deeper water and there will be less light I am wondering if VLOG is a better option. Dynamic range is not the issue there isn't a lot at depth but low noise is. I will be shooting with predominant artificial LED light Would love some theoritical pro and cons of Vlog vs Cine like D in this circumstances (optimise noise not range( Thank you

Panasonic DMC-GH4 Voigtlander Nokton 25mm f/0.95 SLR Magic Anamorphot-50 1.33x Anamorphic Adapter

Hey @Mattias Burling, I shot a comprehensive image test with the Canon RP. If you want, I can send you some original footage. I don't have any footage of the 6D MKII to compare it to but I have the feeling it is not going to be much different.

I look like that and all I do is watch my screen. And deceive myself.

So, let me see if I caught everything there: It's an 8K camera that can't record 8K Canon don't even have an external 8K recorder They made an 8K video but we can't watch it in 8K A YouTuber with a RED posted an 8K video on YT over two years ago, but Canon hasn't worked out how to do it yet? Did I miss anything?

The truth about running people don’t want to hear https://www.stuff.co.nz/life-style/well-good/motivate-me/111240962/the-truth-about-running-people-dont-want-to-hear Basically, if you want to run a marathon you start out by doing 2km jogs around the block. Then doing 5km races, then try out a half marathon, etc etc This is true for everything in life!

Hey thanks to all for the awesome grades you did. Its making mine feel too "Pinky" in comparison. P4k Shot it back in Dec I think 2 or 3 updates ago... It was a very loose random "shoot" just messing around. was mainly trying to test out a $900 Lumix G II 12-35 2.8 that had a scratch on it. I scored for $200 on the ebay... you can only see it well when in certain light and it just looks Kinda blurry. Id say its acceptable "Character" for a $700 discount. Here is the full Vid if Interested.... Shot it 1080p and uprez'd to 4k for output.

What about the 6.1 firmware version + saving 63.3%? source: http://www.cineticstudios.com/blog/2015/7/slimraw-an-easy-method-to-save-space-with-cinemadng-footage.html According to my math you get 134 minutes of 4:1 into a 500GB SSD which is more space in disk than .braw 5:1 offers... Not bad if you have a capable machine to handle post, I guess, even though, slimraw promises a way faster workflow.

Zach, think of this as being like a martial art. you wanna be a black belt, but theres lots of other colors of belts that you have to get first, and thats gonna take a long time! you currently have whatever belt u get for just showing up on the first day of class - the ones 5 year olds have so, u have to like......... do stuff??? no ones just gonna give you a black belt, you have to earn it, and there are no shortcuts in evolution

It is better than nothing! A minute long film that is made is better than a zero second long film! Let the AI write the script for you! Just make it:

Hell, even just reading out select sentences from this thread would be funny. It would sound like crazy people all standing around blabbering about kind of the same stuff but kind of not. It would even have quite a crescendo when you get to this post here and it reveals what you did and what they've just watched.

Very cool, they will be very pleased. Do you have any work online? Jamy Hang just shared these, one still from his film, and the others from fashion:

Anyone selling anytime soon? I want to buy one for my pocket 4k.

It’s not even FF ...

@Mattias Burling Sent you an email ?

Didn't know you'd been to my in-laws place.

I'd keep my eye on it but wait. Every new model seems to get better by quite a lot, and Insta seems to be the DJI of 360 cameras so they're the ones to watch. You're right that 4K isn't enough for 360 degrees, and they're not going to up the bitrate much beyond the 100mbps, at least for the two-lens consumer versions. That's why I say to wait for the 8K versions. It's highly promising new tech, but it's just not quite there yet for the kind of quality that people on this forum demand as a minimum standard.

I thought you got good at running by watching those "exercise motivation" videos on YT? Tell me you're not tired just watching that...

Yeah. The more 3D stuff I watch, the more I think you have to be reeeeeealy careful if you move the camera! Can you put a USB into it when it's mounted? If so, you'd likely be using it with a pole mounted in its blind spot so you could potentially just coil a USB cable up the pole? Log on 8-bit low bitrate consumer camera doesn't sound that promising... Great! Another product that will only work in countries that are so cold all you do in winter is sit inside wanting to kill yourself and then when it's barely above freezing everyone strips off naked in parks and old people and babies die because the temperature got all the way up to the temperature we run our refrigerators at.

Oh man...that sucks ! The small Panasonic 14mm and 20mm pancakes are pretty good little lenses. For versatility, particularly as it has OIS, the yacht club special is cheap, cheerful and definitely worth having in your pocket. Although judging by the size of your pocket in the picture above you could probably fit a large tele-zoom in it.

@androidlad Upon further inspection I found out that the Rec709toRec601.cube LUT is not 100% accurate, sometimes it will be slightly off. This problem might be something that can't be fixed with a single LUT, or matrix. For a really accurate conversion transcoding seems to be the only way: ffmpeg -i INPUT.MOV -vf colorspace=all=bt709,scale=in_range=full:out_range=full -c:v prores_ks -profile:v 2 -c:a copy OUTPUT.MOV Just remember to switch ProRes to full range once it's imported.

Having looked at the spec, it seems likely to be based around this Sony display. Based on AR/VR glasses that also use this display, I'm suspecting it will be between $800-1000? https://www.framos.com/en/ecx335b-20618

He looks like he's struggling to work out where to plug that unicorn tail in.

Messaged portkeys yesterday; the price is slightly higher than I expected, though still considerably cheaper than Zacuto's offerings... (price is officially annnounced next week) Zacuto have just announced their rebranded JTZ Khameleon EVF and it has a retail of 1950 USD.

The spec sheet is the same as the 360 ONE X so yes that will be 18mp total. 360 ONE X also shoots LOG but I only briefly experimented with it as I wanted something I could share fast out of the camera on holiday so didn't fancy the extra steps in colour correcting it. Should be lots of examples on YouTube of LOG footage for you to have a look at though. The ONE X had a spare battery in the kit as a nod I guess to acknowledging battery life isn't great but I'm not sure if this one does? What I find is that because its capturing everything then I don't need to use a phone as a preview monitor so that improves battery life but also I just had a small usb power bank in my bag to keep it topped anyway. What I would say after using it up a mountain is that if you do extended takes you won't need to pack a hand warmer!

Max did RP vs A6400, colour wise RP is much better, the A6400 got funky ass colour and zombie skintone at 4min mark, 1080P RP is much less jaggy than A6400 and more detailed, but 4K A6400 is much better for detail and low iso and AF. And DR is better on Sony thanks to slog/hlg profile. Everyday Dad also did A6400 & RP

I don't care about PR. I just think one could do some cool things with the 3d180. The depth felt really convincing even in the static birthday demo shot, so I figure making a high speed chase or a Hardcore Henry-like fight scene with that could produce some pretty interesting results. Of course it could also produce just motion sickness ?

I mostly used to record raw for one specific purpose: Denoising in post in scenes where the camera could not keep up with a very dark stage that contained well lit elements. This was absolutely needed on the BMPCC with ISO1600 and is still needed on the BMPC4K when not blowing our the well lit scene elements. For the simple reason that it would not compress the noise, so that NeatVideo (3, later 4) could do a great job on it.

if I have 1 camera with a tiny sensor in my arsenal (except my phone) then it better be a 6-8K 360 one. They do almost all the classic stuff, plus a myriad other things that classical concepts can't. 4K is too little for 360, the first - cheap now - 4K cameras had unacceptable quality for even home videos. 5.7-6" is the starting point I believe, but 360/VR will be truly benefited with 8K video.

These are very nice. One of the main challenges I see with this LUT pack is in the variation of skin tones. Many images posted have skin tones where there are areas of skin that are obviously yellow and others that are obviously pink, or where the skin tone is completely pink and there is nothing towards the orange side of that spectrum. I will be the first to admit that I'm not a keen judge of colour, but to me good skin tones seem to sit in the ambiguous area between the "why is that person yellow? are they sick?" and "why is that person pink? are they sick?" tonal ranges. Some examples show colour in both of these areas and I suspect that its to do with excessive saturation in the final grade. I've messed around with these LUTs on my own footage across a number of different shoots and it does push the skin tones quite far towards the magenta, but when I look at Alexa footage it doesn't seem to be graded with that level of push, so I think controlling that aspect of the colour is an important part of grading this material (and I suspect also when grading Alexa files, simple because of how well the LUT pack matches the Alexa, unfortunately I haven't had the pleasure of playing with Alexa footage myself). To this end I find a lot of benefit in using hue adjustments to ensure that the hues in skin tones are contained within that sweet spot between yellow and pink. I'm curious to hear what other people's views on this are, maybe it's a personal taste thing?

I just operate under the concept that you choose your base frame rate and carefully match it to shooting normal speed clips, but when shooting slow-motion you just conform it properly and it kind of doesn't matter in a way. For example, if you're on a 25fps timeline and you shoot 50p slow-motion then it will conform to 50% of real speed, but if you shoot 60p it will conform to 40% of real speed. I'm not really sure there would be that many situations where a 50% speed shot is required and a 40% speed shot wouldn't also be acceptable. In that sense I kind of view 50p and 60p as being the same,.

Interesting discussion, but we need to be careful we don't deviate from what actually matters, which is how the image looks to people. @amanieux is correct that there is lossless and lossy, but there is no real 'visually lossless' category, unless they saved space by removing the metadata or something, but that's basically pointless. It's more accurate to talk about how much visual loss is caused by how much compression. For example, mp3s are very highly compressed (down to single digit percentages) but the perceived sound quality remains a lot higher than the file sizes because they're very smart about what information they're removing, and have chosen to remove the things that humans are least sensitive to. In visual terms there are equivalents to this and certain changes do a lot more perceived 'damage' to an image than others. For example, if you took a video signal and made every pixel 2% brighter then we basically wouldn't notice and no-one would complain. But if you made every pixel 2% brighter, but only on every second frame, it would flicker like mad and would be a total shit-show, despite the video quality actually having half as much total error. Obviously this is a ridiculous example, but when we compress video we are essentially doing two things, one is that we are choosing how much total deviation there will be (all else being equal, more compression = more deviation from the source material) and we are choosing where and how that deviation will be allocated, both in the frame and across frames. When you heavily compress an image you often find horrific blocking artefacts in plain shadow areas, and also on flat surfaces next to edges. This is due to how the image is compressed and which algorithms are used to do so. It's less obvious today since the compression in JPG and MPG has kind of taken over, but those of us who recall comparing images from a GIF file that only had a small number of colours but could do flat surfaces and edges perfectly with a JPG image that could do many more colours but flat surfaces and edges abysmally will know what I mean. It's kind of like when we talk about ISO performance of a new camera - one reviewer says "ISO 1000 is the limit" and the next person says "ISO 6400 is totally usable" and we're really none the wiser because all we know is that they have different tolerances for noise in their images. In video we do have some standard measurements, like Signal-to-noise, but unfortunately that's difficult to calculate, and due to perception not being purely mathematical it's also not a reliable predictor of how good something looks. In a sense, what we should have is a standard test of what percentage of people could tell the difference between the source and a compressed version over a range of delays between the images. That would give us something like "compression example 1 was visibly different to 80% of people over a 1 second gap and only 30% of people over a 40s gap, whereas compression example 2 was visibly different to 60% of people over a 1 second gap and only 10% of people over a 30s gap" and that would be useful when compared with other things like file sizes etc. Unfortunately, I don't know about such a standard test, and even if there is, no-one seems to be doing it.

In addition to @Towd's good advice, I'd suggest bookmarking the demo videos from the big cameras as colour reference standards. eg, the video from ARRI for the LF, the Canon C300, some of the Sony ones, Panasonic EVO, etc. These videos are demoing the camera and to show off the camera they spend a truckload on a grade to make it look great, but they're not trying to show 'a look', they're trying to show the camera in as neutral a way as possible. If you swap back and forth from your grade to these demo videos then you'll have a good reliable reference to make sure your eyes aren't going too far from a neutral position. You can even use the demo videos as references to find images from nice films (film trailers for example) that you can collect to for a 'look book' of reference styles to reference. All these references will help you to not run away with yourself, provide inspiration from reliable sources, but also partly make up for a lack of a proper calibrated colour grading setup. I don't know about the various copyright infringements involved, but (in theory!!!) you could download those reference videos from YT with a video downloader, then pull still images from them in your NLE, and save them into a folder that you've bookmarked in your NLE so you can just pull them up at any point during grading for a quick reference. You know, in theory...

brb kms

Compression isn't even the issue here. Was using 4:1 cDNG all the time with fantastic results while Braw at Q0 still smudges texture, fine details, creates fringing and color bleeding issues as well as contrast halos around edges. It's just an internal processing issue and I really wish BMD could add a low or no internal processing option.

Damn, I was hoping you were going to say it was awful so I could save some money

I once worked with a really good compositor at a large VFX house who admitted to me once that he was totally colorblind. His trick was that he just matched everything by reading the code values from his color picker tool and matching the parts of his composite purely from the values he sampled. I've always remembered that when I feel I can't trust my eyes, or something is not working for me. You can color grade just by making sure everything is neutral and balanced. Later, as you become more comfortable with the process and gain more experience you can start creating looks or an affected grade. Generally to start you want to get your white balance correct. Find something in your shot that you know should be neutral or white. A wall, a t-shirt, a piece of paper, or anything else that should be white or gray. After that, check your blacks and make sure they are neutral, then double check your whites. Finally, check your skin tones and make sure they are correct. You can do this by using the vectorscope and just getting them on the skin tone line. Somewhere in this process you'll want to set your exposure. I generally just make a rough exposure adjustment at the beginning so I can see everything, then dial it in once my balance is set. One thing I do a lot when studying how a film I like is graded is to take screen captures from a Netflix stream or other source and pull them into a project to compare the color values on it. Then you'll have a roadmap for for what you are trying to match.

Speaking of which.... I did tests at f1.4 and at f5.6 The top row is the Samyang 85mm at f1.4 and f5.6 and the bottom row is the Contax Zeiss 85mm at f1.4 and f5.6 I deliberately underexposed slightly as I would do normally with the SL but have added no corrections hence the slightly darker image. First up, the chart stuff. The Contax Zeiss has a glow at f1.4 but its a tiny focus error on my part being a blind buffoon. I could have stopped at this point as its pretty obvious that the difference between them is marginal but I like to give value for money so I did a bit more of a contextual test. It might not seem an obvious test (it looks more like a cry for help with all that booze) but its quite informative to illustrate the out of focus rendering, handling of colour and tonality, detail etc etc. Same lineup, so the top row is the Samyang 85mm at f1.4 and f5.6 and the bottom row is the Contax Zeiss 85mm at f1.4 and f5.6 and focus is set on the Cointreau bottle. Just to add a bit more context to what you are getting separation wise for your f1.4 dollar, I brought a couple of slower manual lenses of a similar focal length to do the same test with with both being shot at their maximum aperture. The top row is the Contax Zeiss 28-85mm at 85mm f3.4 and f5.6 and the bottom row is the Leica 90mm Summicron M at f2 and f5.6 CONCLUSION ? My own view is its not really worth pixel peeping between the Samyang 85mm f1.4 and Contax Zeiss 85mm f1.4 when they are as close as this. The Samyang though.....bargain.

That lens is often mentioned in reviews of other 28mm f/2 lenses. It seems the general consensus is it’s a pretty decent lens but for only a little more you can get the Nikkor ai-s which is phenomenal. Either way, it’s a great option for a budget lens. I have the ZF version and not the traditional Contax “Hollywood.” I got it for a good price but it wasn’t cheap by any means. And the only reason I bought it was because it was a Duclos mod. By just using it for a few minutes in the house, I couldn’t say it was leaps and bounds better than the Nikkor 28 or the Zeiss QBM 25mm. But I have a feeling that the Hollywood is a lens that requires a little more than a couple minutes around the house test. Comment of the day! Thanks for proving my suspicions that I am wasting time and money worrying about lenses.

You should do "Procrastination - The Movie" You've already wrote most of the script.

Roll two dice and then open a dictionary to a random page, go to the line number matching the first dice, and the word number on that line of the second dice. Do that a dozen times and write them all down. Pick the stupidest / worst / most bizarre three and make a script inspired by or including those words. You can choose to apply a three-act structure or conflict or drama or tension or whatever the hell you want, but the vast majority of "cucumber buzzsaw hysteria" or "spontaneous curry nightingale" films will be fun to make and fun to watch..

We will introduce a 0.7” 1080p #oled EVF#, portkeys OEYE-3G in NAB, with 4KHDMI in and out, 3GSDi in and out, brightness waveform, 3DLUT..., all-aluminum body, diopter. NAB also publishes price . ?

I have been doing quite a bit of raw testing and so far so good. Has anyone worked out a good braw workflow yet? Some still grab from my test yesterday (SOOC all metadata, no sharpening added):

Sure, glad I could help! That's unfortunate. @CaptainHook It really needs fixin I have an idea for an in-cam ProRes fix once in hand, but it depends on how things are implemented therein