Attila Bakos

It's likely that I see this differently than most people. I don't have a Panasonic camera but I build IDT's, among other stuff, so small differences matter to me a lot.

That's a fact, Panasonic did not release an IDT, they say the GH5 is not compatible with ACES. Bullshit.

To anyone using RCM: just checked the latest Resolve and while Fuji files are displayed correctly in Davinci YRGB since v16, RCM is still buggy, and you can't even use my matrix fix there. It's interesting because even ACES is fine. I'm talking about this issue:

It's the compression, I never release LUTs with such problems

Would anyone be interested in LUTs that could be applied on S-Log2/S-Gamut3.cine clips on an ACEScct timeline? I extracted color information from Fujifilm profiles, and built special LUTs for ACEScct. The effect is subtle because the gamma is unchanged, only color is changed: Normal S-Log2/S-Gamut3.cine clip on an ACEScct timeline, with only exposure adjustment: Provia LUT applied: Classic Chrome LUT applied: I have all the other profiles as well.

I sell a package that contains this fix and a bunch of others: http://colorizer.net/index.php?op=technical It's not the flagging that's fixed, but the bad YUV->RGB conversion when the flag is ignored.

Or not ? https://www.fujirumors.com/fujifilm-x-t3-successor-not-to-be-announced-in-january-but-in-february-march/

@androidlad Do you happen to know anything about this? https://www.fujirumors.com/breaking-fujifilm-x-t3-successor-shipping-in-march-early-spring-2020-trusted-source/

Btw which lightbulb do you use to test extreme gamut?

I think he posted some code screenshot in the past where he fixed hue twists. I'm 99% sure he fixes this issue but I'll let the man talk

3 stops.

That sounds like P in the HSP color model. To me it reprents luma better than L in HSL, or V in HSV.

Be careful with these cheap recorders. I wanted to do the same, monitoring on a laptop, with some scopes. I wanted to do it as cheap as possible, so I ordered this device: https://www.fasttech.com/p/9677213 I also prepared some pretty complex ffplay scripts that capture the stream and add a scopes realtime, like rgb parade, vectorscope, because I couldn't afford scopebox. The device worked but the quality was pretty bad. It has an mjpeg compressed output and an uncompressed output. Even the uncompressed output suffered from resolution loss, it modified the gamma of the source (not a video range/full range issue), and there was a very noticable color shift as well. And it took me 3 months of discussion (in which I had to make videos too) with the support to get a refund! I even got a response from the manufacturer and they told me my results are fine, the captured image is good. It was unbelievable really. My point is that these cheap capture devices might be fine for gamers, but they are absolutely useless when it comes to monitoring. Of course my opinion is based on this device alone.

Why is it too high if it is not clipping? The values for the top 6 stops seem to be distributed equally in S-Log2.

I found a clip where I tested the stabilisation of the GoPro, it has my face in it, so you can see how the DCTL corrects the Protune clip. Protune Flat/Native: DCTL: Normal contrast/saturation adjustment in Davinci YRGB mode: As you can see the DCTL corrects for the yellow skintones, also the jacket of the lady in the background is supposed to be red, which is not what you get with normal contrast/saturation adjustments. I believe my DCTL/LUTs can really make your life easier when shooting GoPro Protune Flat/Native.

The value of Y (luminance) in the lift/gamma/gain bars is calculated from R, G, B (each channel has a different weight in the calculation, green has the most), it's an additional controller that helps you to change luminance without changing chroma, so yes, if all you need is a luminance adjustment, try and see if it works better than using the wheel. The wheel adjusts all three color channels equally.

I actually know @Deadcode, we help each other where we can ACES won't force you to change the way you expose, it will actually be easier to correct an overexposed image. The offset wheel's effect is very close to a true exposure controller in ACES.

The IDT is the entry point for ACES, it linearizes the clips and converts to the color space to ACES AP0. Then ACES will convert this to a logarithmic gamma (either ACEScc or ACEScct in Resolve, whichever you choose), and to AP1 color space. So yes, you can use the IDT with both.

You might want to check this tutorial on mixinglight: https://mixinglight.com/color-tutorial/getting-know-aces/

Unfortunately I don't have a GoPro (got the clips from a friend), but if you send me a clip with people, I can show you how it would look. Remember that the color mode must be flat, and white balance must be native.

Two new packages added to my collection. Details here: http://colorizer.net/index.php?op=aces

Colorizer.net is my stuff, thanks @Deadcode for mentioning it. I also have Fuji film simulation LUTs specially designed for S-Log2/S-Gamut3.cine, if you're interested. (I used an A6300 for the conversions.)

http://www.dvxuser.com/V6/showthread.php?303559-Measuring-rolling-shutter-put-a-number-on-this-issue! XT3 4k 24fps ------ 20.6 ms (20.9-20.4-20.4) Z6 4k FF ---------- 23.0 ms (slashcam.de)

Well, it's certainly enough to steal a size 64 LUT Or do you use higher resolution?

Oh I'd like to send a still as well! It's an 512x512 PNG and it has 262144 unique colors, is that a problem? I'll show myself out... ?