Doug Jensen
Carbonite Member
Watch out, Chris, you're skating on some thin ice when you start agreeing with anything I have to say around here.
FS7: Rec-709 (800%) as an Exposure MLUT
- Thread starter karma17
- Start date
starcentral
Veteran
Yep we definitely do and as long as it's working for you, that's all that matters.
From the chart I posted on the first page and looking at where white sits when slog is "properly" exposed you can sort of tell that LC709/LC709A with it's lower white is going to give you a less contrasty image which is definitely helpful in some situations like you mentioned versus with 709(800) it will probably look blown out even when it's actually not. The hi/lo feature is neat but I just never use it when I can just keep an eye on my WFM overlay without having to press any buttons on the camera ,etc. .
I tend to send a LUT'd image out SDI sub with 709(800) for director/top level client which I can also glance at as needed, and then on SDI main I'll monitor slog to my external camera monitor where I can view slog values on the waveform monitor and if I like switch LUT on/off for my own viewing pleasures.
Last edited:
moldcad
Veteran
That Mr. Jensen will always differ in his opinions with Mr. Chapman is something everyone in this (and all others) forum should know by now, I think.
But to me, saying that with the flat log image the basic focus assistance tools (like the peaking) works equally well as with an MLUT-normalized picture in one's viewfinder, is simply and totally not true - after all, to trigger peaking some minimum contrast threshold is simply indispensable. Of course - with a scene which is bright and contrasty by itself, not only will its log representation be above that threshold and trigger peaking; it actually can be more precise (less false positives) than with MLUT on, but an average scene might simply not provide enough contrast anywhere to do so... In the ideal world, it's the peaking algorithm which should be responsible for that, offering wide enough intensity (or responsiveness) range to cover all possible log scene contrast levels, thus warranting peaking triggering exactly where it should while eliminating false positives - but with the FS7's own EVF (and its peaking "intensity" limited range) those really low light/flat contrast scenes - when further flattened by s-log representaion - are more likely to fail showing peaking at all than to get cluttered with too much false positives... Just my 2 cents, anyway.
Piotr
But to me, saying that with the flat log image the basic focus assistance tools (like the peaking) works equally well as with an MLUT-normalized picture in one's viewfinder, is simply and totally not true - after all, to trigger peaking some minimum contrast threshold is simply indispensable. Of course - with a scene which is bright and contrasty by itself, not only will its log representation be above that threshold and trigger peaking; it actually can be more precise (less false positives) than with MLUT on, but an average scene might simply not provide enough contrast anywhere to do so... In the ideal world, it's the peaking algorithm which should be responsible for that, offering wide enough intensity (or responsiveness) range to cover all possible log scene contrast levels, thus warranting peaking triggering exactly where it should while eliminating false positives - but with the FS7's own EVF (and its peaking "intensity" limited range) those really low light/flat contrast scenes - when further flattened by s-log representaion - are more likely to fail showing peaking at all than to get cluttered with too much false positives... Just my 2 cents, anyway.
Piotr
sleekmedia
Well-known member
I agree with the above statement. I have seen evidence of this with my FS5 + Shogun Inferno. With S-log 3 and low-light scenes I have noticed that when adjusting the Atom HDR feature between 400%-1000% values that at 1000% I see almost no peaking and at 400% I see many false positives and even peaking on grain in the image. I have found it hard to focus sometimes in scenarios as such especially when stopped down when everything in the frame is fairly sharp but you want to ensure critical focus exactly on your subject. I have learned how to work around it in probably not the best fashion by finding my focus and or focus marks at a higher ISO to get more accurate peaking there then drop back down to the appropriate level when I am confident I am at where I need to be but this is obviously not great for fast paced environments. But overall I have to agree with the above statement that with S-log and low-light, peaking is not to great and going the opposite direction with an MLUT is not much better sometimes as you get too many false positives which I find even harder to work with. Sometimes I think using False-color rather than focus peaking works better for low-light. Anyone have any other suggestions on how to better manage focus in low-light scenarios.
Last edited:
starcentral
Veteran
Thanks for pointing that out. I fixed the image and one on page one too.
But who's to say that Doug or "everyone" working and monitoring in slog uses peaking?
I know for myself I rarely if ever use peaking focus assist and I've even had full day shoots doing crane work with macro lenses monitoring in slog (slog because we were shooting HFR ) and pulling focus off a 7" monitor. I think it boils down to what you get use to or have experience with, but if you do use peaking for focus assist then I agree that in slog it doesn't work as well versus LUT'd image.
Doug Jensen
Carbonite Member
Dennis, I agree with your point about not making assumptions about how other people work, but just for the record, I would never, ever, under any circumstances not use peaking for focus with any professional video camera. It is on 100% of the time and it is the only focus assist tool I need. I find it works great with or without S-LOG and I totally disagree with the comments above from people compaining that they can't see it or they get false posiitives. That is certainly not my experience.
I only shoot S-LOG or RAW, I never use AF or Focus Mag, and I never use a monitor LUT, yet some how I can nail focus 99.5% of the time shooting with f/2.8 or faster lenses wide open. Imagine that.
moldcad
Veteran
Absolutely true, Dennis - was just discussing one of the (probably) most popular usage scenarios.
Nevertheless: the fact is that out of my 3 cameras (FS7, GH5 and A7Rii), only the latter has no possibility of monitoring log with 709 type of MLUT - and I personally find it the less comfortable to work with
Piotr
starcentral
Veteran
Great clarification Doug.
Also if people are keen on using peaking with slog, be sure to adjust the settings buried deep down in the menu - they can make a huge difference.
NorBro
Major Contributor
Nevertheless: the fact is that out of my 3 cameras (FS7, GH5 and A7Rii), only the latter has no possibility of monitoring log with 709 type of MLUT - and I personally find it the less comfortable to work with
Piotr
The a7RII just missed the "Gamma Display Assist" 709 feature, which is now standard on all Sony mirrorless' with S-Log2/3.
thirdyearff
Active member
Interesting...
I'm also one who never uses a monitor LUT, shoots S2 99.9% of the time and also uses peaking...about 50% of the time. I find I use it (or lean on it) more towards the end of a long day when my eyes get tired and I start second guessing if I'm nailing critical focus or not. It has not failed me yet.
I suppose we all have our process...and comfort zones.
I'm also one who never uses a monitor LUT, shoots S2 99.9% of the time and also uses peaking...about 50% of the time. I find I use it (or lean on it) more towards the end of a long day when my eyes get tired and I start second guessing if I'm nailing critical focus or not. It has not failed me yet.
I suppose we all have our process...and comfort zones.