Thread Davinci Resolve 17

[rand thompson]

Thanks Jake, I didn't know any of the info you provided above!

[rand thompson]

Resolve 17 - New Fusion Features


By
VFXstudy

[Mike P.]

Thanks Rand, that helps a bit (still grappling with what's non-DV-compliant HDR).

Also, a quick note, there is no such thing as "grading for 300 nits". SDR grading is always done to 100 nits regardless of what monitor is capable of displaying, period.

That's what made me ask; SDR is always 100nits, but is there an equivalent value for HDR?

[jake blackstone]

That's what made me ask; SDR is always 100nits, but is there an equivalent value for HDR?

Yes, 600, 1000, 2000, 4000 nits

[rand thompson]

Mike,


You have HLG. HDR10 and HDR10+(Samsung)


Brief Explanation of these three


HLG

Backwards compatibility with older SDR TV's but not as good as DV.


HDR10

Most all HDR TV's support HDR10. Requires no License fees to use like you have with Dolby Vision, However, you can't really modify Metadata for individual clips to optimize each for the best results like over-exposed and under-exposed clips, it's one size fits all


HDR10+

Mainly by Samsung and not widely supported. No license fees to use. Unlike HDR10, HDR10+ allows for metadata modification per clip.

[rand thompson]

RESOLVE 17 - First Impressions & Cool New Features in DaVinci Resolve 17 Beta


By
Casey Faris






Cool Spider Web...What Does It Do Again? DaVinci Resolve 17


By Learn Color Grading

[rand thompson]

Davinci Wide Gamut HDR workflow


I noticed when using both Red footage in the RWG/LOG3G10 format and ARRIRAW that the ARRIRAW footage looks correct Tone-Mapped to 100nits BT709.BT1886 Full and the Red footage doesn't




Red






ARRI

[rand thompson]

UPDATE


When I increased the ISO to 1600 and ran a Auto Analysis in Dolby Vision again it looked normal. The ARRIRAW footage has a 800 ISO in its METADATA so it wasn't high so why does the Red Footage which also had an ISO of 800 look so much darker.











Here is the ACESHDR tone-mapped Fullscreen Screengrab at 800ISO vs the Davinci Wide Gamut with a 1600ISO modification. I still prefer ACES.


ACES




Davinci Wide Gamut

[Mike P.]

That's probably because those test shots were done with ISOCal1, which was almost exactly 1 stop darker than metered ISO800. It got fixed with DSMC2 Unified/ISOCal2, but was a little late given that users have complained about it since MX. Makes sense that DR17 would go with ISOCal2, as it's the new normal (and you can only have one "ISO800" setting), but if what your showing is accurate, footage shot before 2017~2018 or pre-DSMC2 (which never got the recalibration for some reason), will have to bump it to ISO1600 to get everything to fall in line accordingly.

Of course, there's no reason for someone in post to know about that, but better to figure it out in post than if cam-dept/dop didn't know about it on set/metered for "ISO800", as you'd end up hard clipping highlights a stop earlier than necessary (and HL Lattitude was already less than the competitions)... But I digress.

[rand thompson]

Mike,

I had the same thought about the recalibration as well. But since the image I used has gone through the latest iteration of IPP2 in Redcine-X, that should already be incorporated into the sample image. I just thought Blackmagic Design already would have taken that into account with all of their experience with Red's SDKs so I was thinking it had to be something else.