A7s sample footage

squig said:
Did you try S-log with cinema colour? It's a lot more Canon like. S-Gamut is a bit weird. My biggest mistake was trying to use S-Log LUTs. Now I just get things looking right in R11 and then apply the default Filmconvert profile and everything looks how it should. The F55 S-log profile colours are all wrong.
Click to expand...
They are. But if you add magenta back in it's all equalized. I am sure there are better ways though... I should try a different color profile with S-Log...
 
Yeah, the F55 profile works great with all of the profiles, I find. It's just the Sgamut that takes some finesse to get it back in line. It's the best I've found until we get a proper profile but hey, whatever gives you the look that works, that's the one to use!
 
There is a line somewhere in the A7S documentation stating that "S-Gamut" is not really S-Gamut and only partially covers it. So it is not a surprise that S-Gamut tuned LUTs or software or whatever aren't working exactly as expected.
For anyone finding footage too green I'd suggest the Movie color mode. It is a bit offset towards magenta on the G-M axis compared to the other modes.
 
Cine4 and Slog2 might seem similar but the contrast you can get while still living within the same exposure space is def not the same. I've went at it multiple times. Similar? Sure but there's no way I would shoot cine4 thinking I was getting the benefits of Slog2's range. Again, whatever works and whatever saves you from more grey hairs but I'm a much bigger fan of slog2's image post grade. That extra latitude just creates a more photographic image to me. Whether you need the stops or not, you're going to have a more film like range to work with and I like that personally.


Hey Squiq, I was looking at all footage too and not just mines. Even the Phillip Bloom one above this post, the colors are just off. I don't man, it's driving me nuts. I wanted so much to like it but I guess I'll wait for something else to pop up.
Click to expand...

People are matching their A7s to F55 footage and it's looks fantastic. For $2500....I think Movie color and +2 color phase would have given you what you were expecting to see. Oh well, it's not going anywhere and I'm sure Sony will be launching some other cams soon enough.

<font color="#FFFFFF">
 
Last edited:
macgregor said:
I've finished my tests with the A7s and they'll be published soon at DSLRmagazine.
11 stops of usable dynamic range in Slog.
11 stops of usable dynamic range in Cine4.
4K test charts will be provided. :D
Click to expand...

Thanks. I'm sure you're aware Cinema 5D concluded there were 14.1 stops of usable dynamic range. Do you have any thoughts on why you guys reached a different conclusion? Not trying to put you on the spot or start any drama. I sincerely appreciate you folks who do the tests.
 
Last edited:
Color etc seems to be quite good here. They noted that this was the first time they'd used the camera, so were just using it as a test...
 
cyork said:
People are matching their A7s to F55 footage and it's looks fantastic. For $2500....I think Movie color and +2 color phase would have given you what you were expecting to see. Oh well, it's not going anywhere and I'm sure Sony will be launching some other cams soon enough.

<font color="#FFFFFF">
Click to expand...

I think the video looks great.

You're right it's not going anywhere for sure, just not everyone's going to like it.
 
Samuel H said:
Great: that lines up perfectly with the other test :)
I guess the Cine4 thing is a matter of playing with the other settings: did you tweak the knees and slopes? Or maybe cine4 works better at lower ISOs (this other test was going up to 50K ISO IIRC)
Looking forward to your detailed results, and, even more, to your detailed opinion about the camera.
Click to expand...

Samuel, does 11 stops sound more credible to you than the 14.1 that have been mentioned? Can you explain the difference? Also, this is off topic but I'm curious, if the RX10 Mkii has SLOG2, how many stops of usable dynamic range do you think it will be able to produce? Just asking you to make a guess. :)
 
I don't too much beef with the noise at 3200 in Slog2 but I sure wouldn't argue if Sony brings back the option to shoot it at ISO 1600. One thing I noticed on a recent shoot, and I'm not sure it had anything to do with it being 100 degrees outside, but shooting slog2 in crop mode gave me some crazy blocking, enlarged noise and sparkles (?). Strange purple and green spots coming from reflections off the water. Like when stage lighting hits somebody in the face and it overloads that exposure and color. (blues/purples are the worst). Once I graded the footage, the noise and blocking was hardly a problem but it was a reminder that FF mode is the better mode for noise and artifacts.
 
I also have a question for Samuel :D Is the method used by McGregor to measure DR similar (in essence) to what you used to come out with the DR table on your site (where BMPCC DR = 9.5)?
 
hgf4 said:
Thanks. I'm sure you're aware Cinema 5D concluded there were 14.1 stops of usable dynamic range. Do you have any thoughts on why you guys reached a different conclusion? Not trying to put you on the spot or start any drama. I sincerely appreciate you folks who do the tests.
Click to expand...

"Usable" is very subjective. It all depends on how you are using the range and for what... Cinema 5D's tests I think may have been more about absolute DR.
 
hgf4 said:
Samuel, does 11 stops sound more credible to you than the 14.1 that have been mentioned? Can you explain the difference? Also, this is off topic but I'm curious, if the RX10 Mkii has SLOG2, how many stops of usable dynamic range do you think it will be able to produce? Just asking you to make a guess. :)
Click to expand...

11 usable stops out of 14 total stops is A LOT of usable stops. On the 5D2 I measured 11+ total stops but just 6 usable stops: the bad codec eats up all the detail in the shadows.
(BTW I do see 14 total stops in the chart that C5D posted)
 
BlackLegSanji said:
I also have a question for Samuel :D Is the method used by McGregor to measure DR similar (in essence) to what you used to come out with the DR table on your site (where BMPCC DR = 9.5)?
Click to expand...
I have no idea what method macgregor used, but I can say my method for determining what's usable and what's not was VERY demanding. I only considered it usable if I could still see very fine detail in the shadows. Bigger shapes were still discernible, but I considered them unusable. I may have put the bar a bit too high on that test, but the guy I was doing that for, I think using such a high bar was best for his needs.
 
Bought two DK-19s in Tokyo yesterday. Makes it much easier to see in daylight, and I think a bit more comfortable to hold up to the eye.
 
Went away for a weekend and shot some stuff. All SLOG2 with Kholi's PP. Probably should've switched to a cine profile for the lowlight shots, was shooting up to 50k ISO and the noise was all over the place.

Theres a mix of 25p and 50/60p on crop mode. Decided to use this video to learn some more resolve, so graded it there with ImpulZ LUTs and some curves. Used gradients to bring back some of the skies, which mostly worked well. I went for a colorful summer look so its pretty saturated.

Also used the time-lapse app, which was 10€ well spent. I love how it takes the hassle out of calculating the time and builds the video file for you too.

http://vimeo.com/103018448
 
Last edited:
Hi all could yal tell me which one you like best? Thanks

Slog2, SGamut, +9 Sat, +2EV, 6400ISO.

View 1
ScreenCap01_1.17.1.jpg
View 2
ScreenCap02_1.17.2.jpg
 
You must log in or register to reply here.
Back
Top