First S-Log Test with Sony F3 on DV Info Net (http://www.dvinfo.net/article/acquisition/sonyxdcam/first-s-log-test-with-sony-f3.html)

My Good friends at Next Level pictures, Khalid Mohtaseb, Jon Bregel and I shot the hell out of some Slog out at NAB!

Enjoy!

Sony F3 S-Log Test - Behind the Scenes on Vimeo

Sony F3 S-Log Test - Behind the Scenes on Vimeo

Visit their site as well! Next Level Pictures - Blog (http://www.nextlevelpictures.com)

Wow... you've got mail,

Tim,
Great work in such a short time. An absolute text book achievement in such a short time.
Terrific gig. Perhaps you can inform the forum as to how using Log is not Latin.

Excellent! Very nicely done.
I also thought the "making of" documentary was perfect. I've rarely seen anything better on the Extras of a DVD or Blu-ray. Gives us all something to strive to achieve.
Thanks.

I think that anyone who was on the fence over whether 3300 upgrade was really worth it could be convinced by this video. pretty freakin' cool.

Really beautiful work Timur. I appreciate you and your team's efforts to do a real-world test with what seems quite limited lighting and time. I just got back from NAB and am staring at my bare bones F3 wondering where to go next, but the S-log upgrade will be at the top of the list.

Cheers,
Peter

Hi Timur. Didn't Andy Shipsides also say that he got the S-log upgrade on his F3 at NAB?

Whose camera do you think we used? ;)

Is my computer monitor off or does all the original XDCam stuff look overexposed compared to the S-Log?
Also what gamma was the XDCAm

Is my computer monitor off or does all the original XDCam stuff look overexposed compared to the S-Log?
Also what gamma was the XDCAm

Technically, it was. It looks like the Rec709 viewing LUTs are set to make the camera one stop slower in an effort to get you to expose down and save some highlights in the S-Log.

So if they shot S-Log to external recorder, Rec709 LUT to SxS, and then were VIEWING S-Log on the external monitor out, that would create the 1 stop difference we see in that clip. They were exposing the S-Log correctly, but exposing the baked-LUT material hot.

Also, frankly, if the S-Log has taken a trip through Apple Color or Resolve, and the XDCAM was cut in "straight" as a comparison, all bets are off. Part of the point of S-Log is you have wiggle room in post to boost or cut exposure, and I'm sure whoever colored this took full advantage to 'kick the tires'.

If all this scares you, it will easily be possible to make a viewing LUT that keeps the exposure similar to S-Log.

Yes, I was wondering about the difference and if the S-log option wasn't available to the internal recorder.

The S-log looks much more impressive than the straight Rec709 recordings I've seen to date from the camera.

It also looks more usable with that rear V/F removed and a top mounting system in place. Sony should make it possible for their dealers to remove the factory top handle & V/F and replace them with suitable top plates, without the camera warranty being infringed. .

Sony should make it possible for their dealers to remove the factory top handle & V/F and replace them with suitable top plates, without the camera warranty being infringed. .

From the mouths of people who would know at NAB, there are folks trying to work this out.

Excellent tests guys, thank you for the wide variety of shooting situations, very informative.

For me the original XDCAM material has a vibrance that appears lost in the S-Log, but that's just personal taste over the S-Log grade which is less saturated. What I would like to see is a comparison done between correctly exposed graded 10 bit Cinegamma material recorded on a decent codec and S-Log, it would be interesting to see how big the difference there is.

I noticed a couple of the night shots of the girl where her cheeks appear to be a teeny bit over exposed or suffering from highlight compression. It's pretty obvious on the XDCAM material but not so pronounced on the S-Log. I wonder if these shots could have done with a little lower exposure. Was this a result of monitoring or something else? Not a criticism of the project as it demonstrates that even with S-Log you've got to watch those skin tone and keep them in the linear part of the gamma curve.

Again... thanks, excellent job.

YEa exposure was the big question of the night.

Having never worked with sLog i wasnt sure of which to expose for. I ultimatly exposed for 709, but gave it a touch of over exposure to see what could be "recovered" in post.

What ISO was used?

I think S-Log will be a steep learning curve, trying to figure out exactly how to expose correctly given the very flat looking unprocessed S-Log image or the reduced latitude LUT applied images.

I think S-Log will be a steep learning curve, trying to figure out exactly how to expose correctly given the very flat looking unprocessed S-Log image or the reduced latitude LUT applied images.

My experience shooting Red since 2008 would be the opposite. I can see why you'd want to be precise about it, after all that's what we all strive for. But my experience, in practice, has been that 10bit log video and RAW (and yes, I've worked with them a bunch) are much more forgiving and in the end, provide means to create happy accidents in post with color correction ('oh, I wonder what it would have looked like if I had shot that 3600K instead of 3200? Ah, nice!')

If you're really concerned, you monitor with a LUT while shooting and then you're always VERY pleasantly surprised in post just how much more shadow and highlight detail there is to work with.

The Red, unless you are monitoring "RAW", always has a LUT applied. You work with that on, and then when you're really concerned about whether that highlight detail is truly holding, you bump into "Raw" monitoring for a sec to see where the highlights really are. It's almost always a nice surprise when you do that.

In other words, you light or expose for the worst case scenario (the LUT applied, latitude trancated), knowing there's a (not small, in my opinion) safety buffer in how the S-Log really looks.

What ISO was used?

the sunlight stuff is 800 ISO, the night stuff is 3200

Thanks! Amazing!

Sony's Facebook page says their S-log tests are the worlds first footage. I wonder who really was first.

It would be a lot more useful as a comparison if the S-Log AND native XDCAM recordings were both properly exposed.

And of course would be most interesting to see a comparison where the SxS recording is graded to match as closely as possible the S-Log footage.

Hi Timur. Didn't Andy Shipsides also say that he got the S-log upgrade on his F3 at NAB?

Great job Timur. Glad I could help on the project.

Andy