I believe you slightly misunderstand the differences 12bit and 14bit. It's not about color, it's purely dynamic range at that point. Kinda like the differences between 16bit and 24bit audio. Dynamic range is not about highlights at all. From 8, 10, 12, 14 and 16bit linear, they all have a point at which they clip just like audio. I like to call this 0db. The difference between all of them is how far down can you go. With 16bit audio you can go down to -96db and with 24bit you can go all the way down to -144db. This is exactly how the dynamic range works in camera sensors. The color differences are down to the fact luminance. Due to having lower dynamic range, the image looks more contrasty because the scene is more spread out between the highlights and shadows, but increase the dr by a few stops and suddenly everything looks way less contrasty because more of the scene can occupy the midtones.. There's just more headroom. Also the closer you get to shadows, the more the detail destroyed by the noise floor and noise reduction. The closer it's to "0db" aka clipping the then the more color information you lose. That's why flog2 looks more "filmic" and very close to the Arri Alexa XT/Mini. The XH2S sensor is seriously close in terms of DR, it's just one stop less due to not having dual gain output. I really recommend getting into RAW photos and shooting in 12bit mode and 14bit mode to really appreciate the differences.
I did some testing between the two when I got my h2s in november, but I thought the differences between the colour accuracy was just due to exposure since I used philips hue lamps to do the testing and didn't think this would make a difference in the real world. But now in spring when recording wildlife/nature stuff in direct sunlight it really shows that the old f-log has some issues with very saturated colours, especially pink and purple/violet, those tend to come out blue-ish and look terrible compared to f-log2.
Hey thanks. I’m not sure yet, still figuring it all out but it seems like you can shoot an even exposure or maybe plus one. I don’t think you have to overexpose as much as you might on certain Sony cameras
Did you apply the Flog2 to Rec.709 conversion LUT as a timeline LUT and added clip nodes with exposure adjustments afterwards? I thought resolve applies timeline adjustments first, so all your adjustments you made are based on Rec.709. Have you tried adding a node with the LUT/colorspace transformation to the end?
I use Resolve Color Management: Color science - DaVinci YRGB Color Managed. Color processing mode: HDR DaVinci Wide Gamut Intermediate. Output color space - Rec.709 Gamma 2.4 I convert the log clips at the media pool level if that makes sense. So not using LUTs and not using CST
I believe if you apply the Input Color Space using the context menu on the clip in the media pool, then your log footage is cramped into Rec.709 first and then your adjustment nodes are applied. You should always add a CST node at the end after all other nodes and don’t transform the clip upfront. Otherwise it’s like opening a raw file, exporting it as a jpeg and then start using the sliders on the jpeg.
Not when you set it up as I outlined. The working color space in the project is HDR DaVinci Wide Gamut Intermediate. It doesn't get compressed to Rec.709 until export
@@PhillipRPeck You're right, if you use color processing mode "HDR DaVinci Wide Gamut Intermediate" or "custom" with DWG as timeline color space the transformations will be applied at the end. Sorry for the confusion.
I’d venture a guess that most people shooting Fuji for video are using one of the baked in film simulations. Not having the ability to properly set exposure for middle grey via zebras in camera is a massive miss for Fuji.
"Most people are shooting baked simulations and most people are shooting flog." You have official source for those numbers guys? Cause a guess or a feeling doesn't make it reality.
In flog 2 the base iso is 1250. Is there a second base iso like in the Sony fx3 and a7s3? Do you know anything about it...I couldn't find anything about it from fuji.
I don't think so. I think Gerald Undone has a video where he goes through the ISO range and finds a cleaner high ISO but he thinks it's probably noise reduction as Fujifilm didn't say anything about a second base
second ISO is 3200. If in doubt, just take a clip with your lens cap on, and then gradually step up your iso. Then in post, watch the amount of noise captured for any given iso. Each shooting mode (film sims, flog, flog2) may have different steps. I basically just use flog2 so i don't remember. tip: disable in-camera noise reduction, it softens the image and renders strange artifacts. Most of the noise you'll get with this camera (if exposed properly) will probably disappear anyway just by exporting in h264 for delivery.
Flog2 will be supported in DMR19
Yeah, I mention that in my X-H2S firmware update video. I've been using it in the 19 beta
Forgive my ignorance, what is DMR19?
@@matango1979 I think he's referring to DaVinci Resolve 19 which is currently in beta
@@matango1979 sorry my i am refering to DVR 19
I believe you slightly misunderstand the differences 12bit and 14bit. It's not about color, it's purely dynamic range at that point. Kinda like the differences between 16bit and 24bit audio. Dynamic range is not about highlights at all. From 8, 10, 12, 14 and 16bit linear, they all have a point at which they clip just like audio. I like to call this 0db. The difference between all of them is how far down can you go. With 16bit audio you can go down to -96db and with 24bit you can go all the way down to -144db. This is exactly how the dynamic range works in camera sensors.
The color differences are down to the fact luminance. Due to having lower dynamic range, the image looks more contrasty because the scene is more spread out between the highlights and shadows, but increase the dr by a few stops and suddenly everything looks way less contrasty because more of the scene can occupy the midtones.. There's just more headroom.
Also the closer you get to shadows, the more the detail destroyed by the noise floor and noise reduction. The closer it's to "0db" aka clipping the then the more color information you lose. That's why flog2 looks more "filmic" and very close to the Arri Alexa XT/Mini. The XH2S sensor is seriously close in terms of DR, it's just one stop less due to not having dual gain output.
I really recommend getting into RAW photos and shooting in 12bit mode and 14bit mode to really appreciate the differences.
I did some testing between the two when I got my h2s in november, but I thought the differences between the colour accuracy was just due to exposure since I used philips hue lamps to do the testing and didn't think this would make a difference in the real world. But now in spring when recording wildlife/nature stuff in direct sunlight it really shows that the old f-log has some issues with very saturated colours, especially pink and purple/violet, those tend to come out blue-ish and look terrible compared to f-log2.
I bought an XH2s recently, and I am keen to use FLOG and FLOG2 so this has been a very useful video. Thank you very much.
Glad it was helpful!
Thank you Phillip, very informative! What would you recommend for the meter settings when shooting F-log2?
Hey thanks. I’m not sure yet, still figuring it all out but it seems like you can shoot an even exposure or maybe plus one. I don’t think you have to overexpose as much as you might on certain Sony cameras
Did you apply the Flog2 to Rec.709 conversion LUT as a timeline LUT and added clip nodes with exposure adjustments afterwards? I thought resolve applies timeline adjustments first, so all your adjustments you made are based on Rec.709. Have you tried adding a node with the LUT/colorspace transformation to the end?
I use Resolve Color Management: Color science - DaVinci YRGB Color Managed. Color processing mode: HDR DaVinci Wide Gamut Intermediate. Output color space - Rec.709 Gamma 2.4
I convert the log clips at the media pool level if that makes sense. So not using LUTs and not using CST
I believe if you apply the Input Color Space using the context menu on the clip in the media pool, then your log footage is cramped into Rec.709 first and then your adjustment nodes are applied. You should always add a CST node at the end after all other nodes and don’t transform the clip upfront. Otherwise it’s like opening a raw file, exporting it as a jpeg and then start using the sliders on the jpeg.
Not when you set it up as I outlined. The working color space in the project is HDR DaVinci Wide Gamut Intermediate. It doesn't get compressed to Rec.709 until export
@@PhillipRPeck You're right, if you use color processing mode "HDR DaVinci Wide Gamut Intermediate" or "custom" with DWG as timeline color space the transformations will be applied at the end. Sorry for the confusion.
No worries. It's a complicated topic with a lot of incomplete and erroneous information out there
I’d venture a guess that most people shooting Fuji for video are using one of the baked in film simulations. Not having the ability to properly set exposure for middle grey via zebras in camera is a massive miss for Fuji.
Most people are shooting flog 2 on xh2s
Yeah, I was pretty annoyed that you can't set them lower than 50. WTF, Fuji?
"Most people are shooting baked simulations and most people are shooting flog."
You have official source for those numbers guys? Cause a guess or a feeling doesn't make it reality.
@@REMY.C. Fuji Fanboy Police 👮🚨 incoming!! 🤣
@@RandumbTech 😂😂😂 not at all but to say such a thing you must have numbers not just a gut feeling ;-)
In flog 2 the base iso is 1250. Is there a second base iso like in the Sony fx3 and a7s3? Do you know anything about it...I couldn't find anything about it from fuji.
I don't think so. I think Gerald Undone has a video where he goes through the ISO range and finds a cleaner high ISO but he thinks it's probably noise reduction as Fujifilm didn't say anything about a second base
second ISO is 3200.
If in doubt, just take a clip with your lens cap on, and then gradually step up your iso. Then in post, watch the amount of noise captured for any given iso.
Each shooting mode (film sims, flog, flog2) may have different steps. I basically just use flog2 so i don't remember.
tip: disable in-camera noise reduction, it softens the image and renders strange artifacts.
Most of the noise you'll get with this camera (if exposed properly) will probably disappear anyway just by exporting in h264 for delivery.
I've tested both and can't see a huge difference between both, actually so little I just keep to Flog...
Me too
i farted
Congrats
Stank butt