Zen & The Art of Mastering LOG & RAW Exposure
Вставка
- Опубліковано 7 бер 2023
- LOG got you down? Struggling under the weight of your image's noise floor? Fear no more! We have an action packed and in-depth guide to getting the very best out of your log or raw recorded image. This applies to any log profile, but we also included six of the most commonly used cameras to walk you through the process.
For those who just want to skip ahead (you shouldn't, but we get it), below is a quick-jump guide. All links referenced in this video and others your should visit are found below.
CHAPTER GUIDE:
00:00
00:17 INTRO TO LOG EXPOSURE
02:29 ADAPTING FILM TO DIGITAL
04:14 PUSHING FILM STOCK
05:05 CINEON, THE OG LOG
05:38 CORE EXPOSURE CONCEPT
06:04 MY APPROACH TO EXPOSURE
13:21 ON SET MONITORING LUTs
17:23 CAMERA SAMPLES INTRO
18:56 EVA-1 / V-LOG
19:57 C300mkII / C-LOG2
21:25 EFFECTS OF ISO/EI
22:11 TRIAGING EXPOSURE
22:42 FX9 / S-LOG3
23:24 ALEXA LF / LOG-C3
24:42 KOMODO / R3D & LOG10G-3
27:55 BMPCC / BRAW & BM FILM
28:59 LUT vs COLOR SPACE TRANSFORM
29:15 CAVEATS & CONDITIONS
33:08 RECAP / TL;DW
VALUABLE LINKS FOR FURTHER READING
PUSHING FILM
Should you over-expose your film? • Should you overexpose ...
UNDERSTANDING LOG
Linear vs. Logarithmic dimming (a good article on the difference between the two ways of encoding light data):
www.pathwaylighting.com/produ...
Log vs RAW: artgrid.io/insights/log-vs-raw/
Lots of academic based documents on gamma & color science: poynton.ca/
WAVEFORMS
Guide to using waveforms: download.tek.com/document/Col...
MIDDLE GREY
Where in the world is middle grey? xtremestuff.net/where-in-the-...
More on middle grey: docs.google.com/spreadsheets/...
RAW EXPOSURE
RED Exposure guide (RAW): www.red.com/red-101/exposure-...
BRAW vs ProRes RAW: www.cined.com/blackmagic-raw-...
LUTS & COLOR SPACE TRANSFORMS
LUTs vs Color Space Transforms: blog.frame.io/2020/04/27/luts...
Working with Color Space Transforms: www.4kshooters.net/2022/03/31...
ARRI LUT generator (another way to make offset LUTs) www.arri.com/en/learn-help/le...
Primer on LUTs (if you know next to nothing) academy.wedio.com/lut/
LIGHTING QUALITY
Complete LED lighting database: indiecinemaacademy.com/comple...
#log #raw #tutorial - Наука та технологія
Damn! I'm saving this video to reference again and again. Truly appreciate the effort you put into this! Killer!
Thank you so much for the wonderful comment! I couldn’t be happier to hear this. 🙏🙏
THE MOST informative and productive video of explaining the relation between log exposure and waveform. I really appreciate the video man.
Thanks for the great comment!! Thrilled that you liked the video. ✌🏼
bro this video was amazing, it deadass taught me a ton. ty for making this!!
Thanks for the great comment, tammam!
Excellent approach to explaining this complex topic! I really appreciated the use of visuals and presentation style. I can see myself revisiting this video repeatedly in the future.
Thank you ikestrom!
For me the best way to expose log is to set the contrast ratio on set and ETTR the sensor so that you just need to pull down in post, while also keeping the depth of field at a reasonable level.
Thanks for the comment!
A fantastic video with lots of well-researched information, thank you!
Thanks poppijoni!
After watching this, I think I need a drink. I use a Panasonic BS1H. I shoot in Log, and I over expose. I think I'm doing the right thing.
Maybe? Very situational….
Thank you for making this
Thank you Ty!
Great video!! Thank you, sir!!!
Thank you Gomez Creative Media!! Super appreciate that you took the time to comment.
Thanks for the excellent and interesting explanation!
Thanks so much Edwin!
Great video! Thanks for educating the community!
Thank you Phil! Super appreciate it!
@@vistek Nice video I really appreciate this. But why is everyone always leaving out Fujifilm xh2s F-log 2 in most comparison videos 😢
Hi Shridex Digi. Thanks for the comment. I don't think there is a conspiracy or anything. For me, it was just that I had to draw a line somewhere. The video is already twice as long as I would have liked. This is a comment I get often when doing any sort of "comparison" - so many people wonder why such and such camera wasn't included. The reasons are always varied. Sometimes is accessibility. UA-cam videos are often made based simply on what the creator has at their disposal. Other times its about clarity of message; adding too many elements can dilute the over all message and understanding. Then I think it's about what I said earlier - just time. Not just run time, but time for everything else (testing, shooting, editing, etc). The only other explinition I can offer is that F-log/F-log2 is rather late to the game and Fujifilm is still not seen as a major player in the video world, despite the beautiful image out of the X-H2 series cameras. I suspect that will change soon, but it's still very much under the radar for many people. Same can be said for N-log (Nikon), etc. Any brand that doesn't also have a notable dedicated video camera in their line will often be over looked. But fortunately the science is all very much the same. So if you know what your white paper levels are, and you test your own camera extensively - then it shouldn't really matter if your particular camera isn't represented. That's the beauty of camera color science.
@@vistek I understand bro. And as I said I truly appreciate 🙏🏽. Just wanted to highlight I love Fujifilm. love from Nigeria. Hopefully someday I will be able to afford a FUJIFILM XH2S and make my own comparison
I love dale with all my heart! - You’ve helped me with this video and in person! - river
Very kind! Thanks, Pal. Happy to help any time!
Thank you!
Thanks for the comment Spicyshow!
Small mistake I noticed was using the HDR Palette in Resolve while not having the Timeline Color Space set properly.
In the project settings it was set to “sRGB” but it should set to whatever Color Space/Gamma you’re working with. Unless you manually change it in the drop-down menu of the palette, it is set to “Use Timeline”.
When correctly setting up the HDR Palette, you should be able to only use the global exposure wheel (it works in stops) to match between your images and do the exposure compensation.
Thanks for the insight & comment Ismael.
Just super
Thank you!
Just wanted to say thanks and the new studio setup looks good (maybe a bit dark for me).
Thanks for the comment MREP! I’m a vampire so it’s perfect for me. 🤣
I think Joel from Phantom LUTs touches on the over exposing caveat. If I remember correctly, color rendering can be affected particulary. Awesome info mate!!
In a video, or in the Phantom LUT manual?
Hi,
Have you seen what Ed Lachlan has created ? The EL zone system.
About lights, Prolycht ( and now Dedo ) brought us the RGBACL color accuracy system which I find quite interesting.
Can't you just use something like LUTCalc to make an offset LUT? That's what I've been doing and now I'm questioning whether I've been doing it wrong.
LUTCalc is totally fine, as is Arri’s LUT tool. If you’re just new to the concept, I think Resolve is a better tool for two reasons: you can immediately see your results (aka there is a satisfying visual component) and you can make small changes if you like or need (such as highlight roll off etc). I meant to put a link to LUTCalc in the comments. Thanks for the reminder to do that.
So if you have a pocket camera...When your outdoors, you should set your ISO to 1000 and ND just below clipping but at night set it to 1250 and add as much lights as needed?
Thanks for the comment, Volak. In my opinion, it's not as simple as that. Every scene whether inside or out, has a lot of complexity. When looking at your scene, you have to decide what elements are more important. Can you live with shadows having less information in favour of more information in the highlights, or vice versa. The subject your shooting and the story you're trying to tell should guide this decision. Sometimes in a very bright environment, I may drive my ISO down so that anything in the shadows will have more gradients and detail - so long as I can keep my highlights from clipping. It take a bit of trial and error. testing, and experience to start to get an instinct for it and to be able to make those decisions quickly when you're pressed for time.
My best advice is to take your camera outside (or inside) and shoot the same scene in every ISO value. Take a look at how the image is rendered. You'll start to get a sense of what 100 ISO looks like vs 1000.
@@vistek MAkes sense, thats for the reply!
Great. Really great. I've been studying this for sometime now. I just go a Shinobi to use the false color. My comment/question it this. I've heard before about over exposing 1 to 5 stops. That seems so extreme. How do one know when he should go for a 5 stop over exposer. It seems like the wild west of exposing. I guess it's trial and error. Now I use the GH6 and the Sony Z90 both shoot log.
Hi SUTV, yes you need to test your specific camera for dynamic range. If you don't test your camera it is absolutely a gamble. Since I've shot on just about every camera imaginable, I now have a bit of a 6th sense regarding how much I can push certain cameras - plus the exposure tools help greatly. What the tools don't tell you is how the image rendition may change on the extreme ends. This is something only testing will tell you. I'm always surprised at how few shooters actually take the time to do over and under and gamma type tests with their cameras. In my mind, it's an absolute necessity before you start shooting with it. TEST TEST TEST!
@@vistek Wow, that is a great answer. I get it. have watched this twice and I'm going to watch it again. I've watched ever video on youtube. Dozens and I took the GH6 DSLR SHOOTER GUIDE course and that was good but nobody explained the impossible like you. Thanks and I'm looking forward. I enjoy testing. :)
Thank you, SUTV! Really appreciate those kind words! Makes me happy to know the effort was worth it. :). It's also worth adding that it is worth the investment to learn Davinci Resolve. It is the best program for shooters to understand their footage - in terms of what it can or can't do. When you understand the shooting side, and the post side - your worth becomes infinitely greater. This way, if you're working with clients where you aren't the editor, you'll be comforted knowing you're handing off the best footage possible to their post dept.
Happy Shooting (& testing!)
Great video. It has been on my plate for a long time to share some of this information. There is so much misinformation or half guesses out there.
One area where I want to add some input is that sometimes the base exposure is desired and it’s ok to use noise reduction in post to manage it. All cameras have used some form of noise reduction. Either in camera or in post. It’s a hassle to use noise reduction in post and it can take a long time but sometimes highlight detail is desired. One of the beauty’s of the Pocket cameras is their ability to shrink the middle gray point to map the stops to prioritize shadow or highlight detail. Sometimes that is desired for some shooting scenarios. Sometimes the shadows have nothing of interest and all the desired detail is in the highlights. That’s why it’s important to have the flexibility to exposure as needed. Some scenarios should be shot over. Some should be shot at base. It’s important for users to understand that need.
I shoot with a Canon R6 and it’s amazing how much this camera that typically gets panned in reviews can actually do. Canon chose a native ISO of 800 which is rather high and clearly prioritizes highlights over shadows. BMD uses ISO 400 and at the end of the day the R6 and clog3 which typically get a lot of criticism for noise is about the same as the Pocket cameras at ISO 800. Shoot Canon and clog3 just one stop over and they are again a pretty good match in terms of slightly reduced highlight detail but much better shadow detail. Unfortunately a lot of people don’t realize this and just think of shooting over as a kind of a hack when it’s not. It’s a way we can remap the stops just like BMD does for us automatically. Canon R6 and clog3 look stunning for rec709 when shot 2 stops over.
Other thing to factor in is when reviewers measure dynamic range they are doing so shooting at base exposure. A lot measure dynamic range based on shadows and noise. Base ISO 800 is going to have a lot more noise and skew the dynamic range reviews but that’s not telling us the full potential of the Canon sensors. Sure Canon could make better sensors for their non cinema cameras but alas they don’t. We still have the ability to use those sensors to greater potential. I wish more reviewers would test dynamic range of cameras across different exposure levels. I really think Canon already starting very high at ISO 800 could surprise a lot of people when shot two stops over and measuring dynamic range. There is zero noise in the image. Sure the highlights clip a lot sooner but for rec709 that is likely not a big deal for many situations.
One last point is I work with a lot of HDR and I’m shifting my priorities from shadows to highlights. That’s another reason I’m starting to become a fan of shooting normal exposure and using NeatVideo in post. If most of the noise can clean up which it does since it’s minor then this allows us to get the full 13.3 stops of clog3 while still giving us a lot of brightness in HDR or more potentially for smoother rolloff for rec709.
Thanks for the insights, DD!