FS7: The Great S-Log-2 vs S-Log 3 Debate

As I said before, I prefer Slog2 as it has 8 really solid stops with decent data per stop - I prefer this to the 10 slightly thinner stops of Slog3.

I dont prefer Slog2 because it clips less when used with the same Tstop as slog3.

S
 
morgan_moore said:
Im going to break some myths, and then agree with everone else :)

AC wrote..
"S-Log2 captures more dynamic range than S-Log3, it goes to a higher level on the waveform, S-Log3 clips the highlights sooner."

Wrong! Sooner than what? Any digital capture should be driven by protecting the highlights and then examining what is in the frame and how noisy the shadows are.

Again Karma17s image is dumb.
The correct way to expose is to protect the highlights. The clipped highlights just show 3 to have a higher effective iso than 2 and the operator failed to account for that.

Now this image is far more telling. If you know what to look for and how to expose.. (I cant embed the image here)
http://www.cinematography.com/cine-uploads/monthly_07_2015/post-34233-0-73986000-1437896168.jpeg

WHat you will see is the data per stop of the two curves

Now Slog 3 has more stops that have a decent amount of data while Slog 2 has more data per stop on less stops.

Ergo Slog2 has less dynamic range (less stops with a decent amount of data) but is more maleable in post.. haveing more levels of data per decent stop.

So..
In a lit/controlled low DR situation Slog 2 will be the better choice as you get a more maleable file
In a wide DR situation Slog3 will be better because you get data over more stops.

Neither is better or worse.

I however tend to shoot in lit/low DR situations, or not give a hoot about blowing some bits of the image, maybe a window, and therefore prefer Slog 2.

Realistically Slog2 has about 8stops of DR while Slog 3 has 10,
Click to expand...

Morgan I think you and few others need to read the article again.. AC,s comment you have taken totally wrong.. the comment about DR and clipping is actually a myth he is denouncing !!

"I’ve written about this many times before, but still it comes up again and again. Which is better? Which should I use? I hear all kinds of crazy comments and a lot of incorrect information, so first of all lets dispel a few myths:

S-Log2 captures more dynamic range than S-Log3, it goes to a higher level on the waveform, S-Log3 clips the highlights sooner."

The DR of both is the same too.. Im not sure you have understood the concept of Log footage..especially as you have been exposing it for highlights ! please read the whole article again..
 
Last edited:
Fair enough I see he is debunking that.

But all of you seem to miss the importance of data levels per stop and how this affects the tonal quality/gradability of an image.

Lets go into the shadows, a setting may record a dark area of a scene but if the recording has no fidelity ie is noisy or over compressed or does not have enough tonal data then.. well is it actually usefully recorded?

From the image I linked (and my experience) the data fidelity of Slog2 'dies' about 8 stops from clip while slog3 'dies' about 10 stops from clip -ergo Slog3 has more 'useful' dynamic range. But the 8 stops from clip of Slog2 are more solid (useful).

One is therefore choosing between not two things that are the same, but something with less useful DR and more data fidelity (slog2) something with more useful DR and less data fidelity (slog3).

--

One of the problems with any DR test is the subjective nature of 'fail'.. aka at what point an image has such a poor S/N ratio in the shadows as to be unusable is a personal judgement.
 
Ill try the analogy of an engine and gearbox.

Configuration A has a long gearbox.. it has a high top speed, but the acceleration is poor and the responce sluggish.
Configuation B a short gear box.. it has a poor top speed but is responsive.

Two ways of using the same amount of horse power. One better around town the other better on a long cruise down the highway.

Yes A and B have similar horse power (data bits recorded) but the allocation is different, giving a different response.

And also that is why DJensen getting his raw engine (4Xmore horse power) and using it to critique the two gearboxes is not giving good information. Because with that many horses under the bonnet no one cares about the configuration of gearbox :)

Look at the image again and consider the stops as gears and data bits per stop as torque provided in this gear.
Yes both have 14 gears, but in Slog2 only 8 give enough torque to be a driving pleasure, and pokey they are, while Slog 3 has 10 usable gears..
http://www.cinematography.com/cine-uploads/monthly_07_2015/post-34233-0-73986000-1437896168.jpeg
 
Last edited:
morgan_moore said:
From the image I linked (and my experience) the data fidelity of Slog2 'dies' about 8 stops from clip while slog3 'dies' about 10 stops from clip -ergo Slog3 has more 'useful' dynamic range. But the 8 stops from clip of Slog2 are more solid (useful).

One is therefore choosing between not two things that are the same, but something with less useful DR and more data fidelity (slog2) something with more useful DR and less data fidelity (slog3).
Click to expand...

I agree with that.

I understood your points, and it's not just about how far away from the clipping point is death for useful fidelity, but the fidelity of each stop along the way. There's more data and fidelity in each stop above mid gray with SLog2 and more below it with SLog3, which incidentally can extend into rendering the finer details of sensor noise. The total range is the same with either. I don't have a strong preference for either, but situationally I might pick SLog3 for its fatter mid range for most people and places and things and SLog2 if I wanted to capture the widest range of highlights with color and depth.
 
Well that is a legit interpretation. But I dont really buy mid grey as a defined point. I think (the classic and known) open up a stop when using log and pull it down in post technique would circumvent your concept of 3 having a fatter midrange?
 
Well, I never have. I always thought that was more of a 8 bit thing on A7 series. I don't stress too much over this, I've always had good exposure with either. When I say mid gray, I'm just referring to Sony spec, 32% on SLog2 and 40% SLog3. The F55 is already getting a bit more light on the sensor with EI1250 as a base.

Conceptually, opening up a stop from would be moving into a range with fewer shades, less fidelity, unless it were raw. I remember Alister in the early days of the F5/F55 when SLog2 was all there was saying to be careful putting faces too high as there was less data there, and that you didn't need to. He seemed to change positions on that (years) later, but I never did. And I still think he too was playing around with the Sony A7 where that might have influenced his decision.

Also, when I said SLog2 has more data above mid gray I meant than SLog3, not more data its own mid gray.
 
Having used Slog/Slog2/Slog3 I prefer Slog3 as it puts skin at about the right level so that I don't have to push it around in post. On the other hand I really liked the texture Slog gave me when raising skin tone in post so in reality i think either work just fine. To me slog worked fine in 8-bit 4.2.0 but both slog2/3 falls a bit short when not used in 10-bit 4.2.2. I think all of these have their pros and cons but in the end one finds a workflow that works and stick with that.

For instance on my FS7 I always use slog3 but on my FS5 and A7RIII I do not use slog. Instead I've set the FS5/A7RIII with a baked in profile that looks just like slog3 from the FS7 with LUT applied. I also often use Arris LUT with FS7 and it looks great.
 
As an aside, I don't use luts. Resolve has gotten so good with it's IDT's, ODT's and tone mapping controls, I freely mix S-Log2, S-Log3 and VLog/L on the same timeline with no adverse consequences. The Resolve tone mapping has gotten so good, I dare say I don't need my Decklink card to do HDR. Well, I still do because that's how I fine tune exactly the look I'm after but it is so very close to that now that you can grade on a 2.4 gamma 709 monitor and just change the ODT at the time of rendering to ST2084/2020 1000 nit after turning off tone and gamut mapping, and the HDR output will be close, probably just fine.

Where the caution comes in with mixed SLog2/SLog3 is remembering which shots were which because Resolve needs to know that.
 
Tom, totally agree about LUTs. I find with doc stuff, where you don't have the luxury of getting everything just how you want it, a LUT will often need more grading than just doing it yourself from scratch. And they are destructive. And when I have exposure, etc., where I want it at the shooting stage, then color transforms, which are not destructive, work best for me. And if I'm mixing my FS7 with other cameras then CTs are great. Sorry, bit off topic.
 
Sony's VENICE color science on the VENICE looks really good-
Click to expand...

I tried out AC's new VENICE v3 S-LOG2 and 3 LUTs and I do not mind them at all. The S-LOG2 LUT is one of the nicest I've used on S-LOG2. They are a complete set for camera, grading, over and under ones, and for various monitors. I've only used them in grading so far because basically that's all I require at the moment. The greenish cast that the original LUTs had has been eliminated. Worth a $5 Latte donation.

Chris Young

http://www.xdcam-user.com/2018/08/n...des-minus-green-luts-for-fs5-fs7-f55-a7s-a7r/
 
Colourists can get good results from Sgamut3.cine/SLOG3, however something about SLOG2 always caused them to **** the bed.

Regardless of any technical merits, I will only shoot Sgamut3.cine/SLOG3 with Sony cameras these days - as (unless you have control of your images all the way through post) it's the safest way to ensure that people don't mangle the pretty pictures you've worked so hard to get.
 
JPNola said:
This thread has reminded my why I'm glad, and I think wise, to not do post. :laugh:
Click to expand...

I think that is very very wrong.

The reason I often disagree with 'experts' AC and DJ is that the dont do enough post. Dont to it in initial camera testing with a scientifically clean mind (only accepting results not accepted assumptions)

OK lets break 'doing post' into to bits - first aquiring a look and then second, munching through hundres of clips. Sure avoid the second bit. Maybe sit in on a pro when doing the first - but you must at least be there when doing the first.

But dont really trust a 'pro' for the first bit either - you are better to use your own hands. Pros are cool but often are pro at fixing work that has been shot by DPs with exposing accepted assumptions*.

Any DP needs to 'do' post - because without doing post you have no idea what your camera does, or how to expose it. You have no idea how it performs when the chips are down and you are scraping for light, or presented with a scene of vast dynamic range.

You cant swing your look from gritty to studio clean. Punchy to romantic wash.

If you dont 'do post' you have no idea what to do on set with your lighting ratios or your Tstop.

As I said above I reject the concept of mid grey.

The reason for this is every (10bit+) camera is 3 cameras depending on.. your grey point, your curve in post.
Harsh and gritty with huge highlight roll - no problem
Studio clean - sure but your wont get much headroom

Before your take a camera on set you must shoot a bunch of brackets, do post on them, and learn the cameras response. You need to do post before you even venture on set.


While you dont do post you have no idea. You are a candybox DP who reads the camera manual or a blog and exposes to that - working with a box - not a camera.

Ps - I started doing post when post was done in the dark - with chemicals and only one go.

---

*One bugbear is the amount of web chat about noise reduction in post. To me that imedialty flags DP error - under exposure - Ive used NR about three times when scraping up footage of some news event shot in near blackness.
 
Last edited:
I dont think that's really true.. I mean the top Dp,s in the world dont do their own post.. of course they and any camera person should have some knowledge of post..that goes without saying.. but you dont have to do it.. in fact there is a lot of room for DPs ,who think they know how to post to actually learn bad habits and not really understanding quite fundamental stuff but thinking they do.. which can be way more damaging ... there is reason someone is just an editor, or just a camera person.. or just a grader .. they are always better at that one job .. whilst have knowledge of the other fields ..
 
Donny 123 said:
I dont think that's really true.. I mean the top Dp,s in the world dont do their own post.. of course they and any camera person should have some knowledge of post..that goes without saying.. but you dont have to do it.. in fact there is a lot of room for DPs ,who think they know how to post to actually learn bad habits and not really understanding quite fundamental stuff but thinking they do.. which can be way more damaging ... there is reason someone is just an editor, or just a camera person.. or just a grader .. they are always better at that one job .. whilst have knowledge of the other fields ..
Click to expand...

Well I dont think 'the top DPs' are always getting the best from thier camera systems.. they tend to have access to enough light quantity and control that they could just shoot with a striaght 709 camera and get it right.

Most top DP dont come from a background of working with flexible digital files.. you would need to look to the top stills people for that.

I also think that top DPs do sit in on camera tests in pre-production and have a large input on developing a look.. even if a A/C and colourist are twiddling the knobs (iris, gain and post curves) for them.

Once a set lut is created and a grey point agreed, then they are basically good to go having nailed their look which can then smoothly transition through post without thier input.

As for specalised pros being better.. well of course they are quicker, and may know how to acheive things, but ultimately the DP is the one who is navigating what to acheive. That is what directing photography is.

I consider directing the photography to be a two part process, lighting and exposure, followed by 'developing' the 'negative'.

Did you ever hear a (chemical) film DP not go bananas if the post house decided to pull or push thier film stock without consulting them?
 
Last edited:
morgan_moore said:
I dont know if Steve Yeldlin is a top DP (in your eyes) but he certainly an example of one who is deeply involved in how digital negatives interact with post

http://www.yedlin.net/OnColorScience/index.html
https://www.imdb.com/name/nm0006911/
http://www.yedlin.net/
Click to expand...

Yes he is , but it also explains my point.. these guys doing feature films and /or high end episodic TV shows.. are the dp,s only..some operate but they are not doing the post production.. sure they may or might not actually present for it..decide with the director before the shoot, what camera, lenses the look\, work with art dept etc.. but they are not doing the post ..the designated colorist is.. he is known for particularly having this interest and everyone has seen his comparison videos.. but you do know they don't actually do the post work on their paid work.. ??? or the editing or actually putting the lights themselves ..thats my point.. no one would argue its not good to know about post but you are saying you have to do it yourself to be a good camera person.. thats crap.. you dont .. and the evidence is there.. every member of the BSC /ASC and all the other great dp,s around who dont actually do post work themselves .. .. actually there are very good Dp,s who dont even have a good knowledge of post.. but their camera work and lighting is of a very high level..
 
Last edited:
You must log in or register to reply here.
Back
Top