F55: S-SLOG3 vs. S-LOG2 -- can you tell the difference?

Doug Jensen

Doug Jensen

Carbonite Member
i decided to shoot and grade some test footage today. See the Vimeo description for an explanation of what I did. Please note that the only grading done to this footage was the application of a LUT -- this is not supposed to be "finished" footage. Can you tell which is which?

https://vimeo.com/226519193

BTW, it's interesting that the S-LOG2 version rendered from Resolve 11% faster.
 
Last edited:
Very pretty effect Doug. Obviously they look identical and that does tell us some things, but is it an accurate idea of what would have happened had you shot the footage not as RAW but as XAVC in the camera where the amount of highlight or shadow attitude would have differed. Obviously the color is identical in your tests, but again would that also be the same if you didn't start from the same RAW file? BTW what LUT's did you use for each?
 
Maybe it's a left-brain / right-brain / no-brain thing, but in a few of the shots I did prefer the look of the A side over the B side.
Not that I have any idea which side is S-log 2 vs. which is S-log 3
 
Speaking of LUTs - which version of Slog2 and Slog3 LUTs are you using? My version of Davinci came with a set of LUTs for Slog3 but no Slog2 LUTs (or maybe I can't find them). Therefore I downloaded them from Sony forum and Convergent design website and they are quite different. I then found a post on Sony forum comparing various versions of the LUTs, differences are quite significant. Which version seems to be the most recent and where is the official Sony source of them? Thanks! This is a nice looking test!
 
broughtonfilm said:
Speaking of LUTs - which version of Slog2 and Slog3 LUTs are you using? My version of Davinci came with a set of LUTs for Slog3 but no Slog2 LUTs (or maybe I can't find them). Therefore I downloaded them from Sony forum and Convergent design website and they are quite different. I then found a post on Sony forum comparing various versions of the LUTs, differences are quite significant. Which version seems to be the most recent and where is the official Sony source of them? Thanks! This is a nice looking test!
Click to expand...

By far the best looking results of S-Log2/3 normalization (out of all "Sony" LUTs I tried) can be obtained with those implemented in their Catalyst suite, IMHO. I only use Browse so it might be different in Prepare or Edit, though. But for grading with Resolve - especially now that Jacob has made his test PQ HDR LUTs available to me - I find Logarist a superb workflow indeed; used RCM or ACEScc almost exclusively for my HDR10 grading but am switching to his great LUTs - highly recommended!

Piotr
 
Last edited:
A. Slog3
B. Slog2

Viewed on my phone, but looked like more saturated greens and reds in A. If you did the same thing with Xavc-i the differenece would be more noticible.
 
moldcad said:
But for grading with Resolve - especially now that Jacob has made his test PQ HDR LUTs available to me - I find Logarist a superb workflow indeed; used RCM or ACEScc almost exclusively for my HDR10 grading but am switching to his great LUTs - highly recommended!
Click to expand...

I also have only good things to say about Logarist. And Jacob is the nicest guy!
 
Thanks for sharing this Doug. Viewed on my iMac set to sRGB display color profile, some of the shots on A appear ever-so-slightly more saturated & some skin tones are a tad warmer... but strangely I don't notice it on every single shot and would be merely guessing which is which.
Curious - since I've never shot raw with 55 w/ deck... and since there's a very minor difference on Sony's recommendation on exposing the two Slog's (61% vs 59% IRE for 90% white ) - assume you rated camera at least 1 stop lower and overexposed to roughly 73% for 90% white while shooting raw?
 
markfpv said:
Curious - since I've never shot raw with 55 w/ deck... and since there's a very minor difference on Sony's recommendation on exposing the two Slog's (61% vs 59% IRE for 90% white ) - assume you rated camera at least 1 stop lower and overexposed to roughly 73% for 90% white while shooting raw?
Click to expand...

Don't those recommended IRE values lead to the same exact exposure when it comes to RAW signal? Meaning, if you apply Slog2 and Slog3 curves on the RAW signal, wouldn't the IRE values of (correctly exposed) 90% white naturally map to 61 and 59 IRE respectively without correction?
 
^ That's my assumption... but then wouldn't 18% grey also map to 41 & 32 respectively - and we would see a difference in those curves?
 
markfpv said:
^ That's my assumption... but then wouldn't 18% grey also map to 41 & 32 respectively - and we would see a difference in those curves?
Click to expand...

Of course you'd see it... if you were looking at a bare log signal - Slog2 appears darker. But you are looking at a LUTed image, LUTs are there exactly to take care of that.
 
Doug Jensen said:
S-LOG3: SGamutCineSlog3_to_LC-709TypeA
S-LOG2: From_Slog2SGamut_to_LC-709TypeA
Click to expand...

Thanks but maybe I didn't make it clear. Over the period of couple of years Sony has released multiple versions of each conversion LUT. There have been multiple SGamutCineSlog3_to_LC-709TypeA LUT's for example. I was curious what was the source of this specific LUT, if it was Sony website (and when was it downloaded), Davinci, Convergent design, etc. While theoretically they should be the same I found them to all be slightly different.
 
The S-LOG3 LUT is the one that comes within Resolve 14.
The S-LOG2 LUT is part of a package of LUTs that were posted in February 2016 and are available for download from the Sony.eu website. I don't recall the exact URL off the top of my head (it is quite long) but I provide it in my Resolve tutorial video.
 
Doug Jensen said:
The S-LOG3 LUT is the one that comes within Resolve 14.
The S-LOG2 LUT is part of a package of LUTs that were posted in February 2016 and are available for download from the Sony.eu website. I don't recall the exact URL off the top of my head (it is quite long) but I provide it in my Resolve tutorial video.
Click to expand...

Thank you Doug!
 
You must log in or register to reply here.
Back
Top