F5: S-log3 noise at 2000 iso

S

Sami2

Active member
i am planning on buying the f5 and shot a test clip with it in cine mode with s-log3 at 2000 iso. As i take the mxf file to premiere pro cc i noticed that there is a lot of noise in the shadows. The footage is a little under exposed.

Is this normal hence for some reason most of te reviews i have looked through on youtube seems to have the camera set at 1250 iso shooting with s-log2. So is the s-log3 or the 2000 iso giving me more noise than i should be getting on regular use with this camera?
 
There is inherently more gain on the lower end of the slog3 curve (shadows) but you can reduce it once you push the value back down in post. You would get the same noise in slog2 if you matched the shadow level of slog3 in post.

If you need to boost your shadow gain I suggest investing in Neat Video NR plugin. I did.
 
The reviews were done with F5 at 1250, not with f55.

Has anyone found a side by side raw comparison of f55 and f5 online, or is there a site from which i could download some raw samples of both cameras?

cheers :)
 
If you expose SLog3 as proposed by the Sony white paper your camera is rated ISO 2000.
SLog3 is lifting the shadows by approx. 1 1/2 stop compared to SLog2.

But SLog is only a gamma curve that doesn't affect the noise floor of the sensor at all.
SLog is meant to be be delogged with the appropriate LUT in post to bring levels back to
"normal". The noise floor will be down again, too.
Basically SLog is all about tonality.

With SLog2 you gain highlight tonality, with SLog3 you gain tonality in the shadows.

The difference after delogging is very subtle!

Some people shoot SLog rated at lower ISOs which means they overexpose to a certain point
to bring down noise even more.
 
"The reviews were done with F5 at 1250, not with f55.

Has anyone found a side by side raw comparison of f55 and f5 online, or is there a site from which i could download some raw samples of both cameras?"

Still the question remains :)

Thanks Erik for going over the basics
 
Sami2 said:
The reviews were done with F5 at 1250, not with f55.

Has anyone found a side by side raw comparison of f55 and f5 online, or is there a site from which i could download some raw samples of both cameras?

cheers :)
Click to expand...


The reviews of the F5 could not have been done at ISO 1250 because the sensor of the F5 is native ISO 2000. Now it is possible to set a different EI but that does not affect the image-- only what is seen in the viewfinder. If a reviewer claims to be shooting the F5 in CineEI mode at 1250, then that review is wrong.

Ned
 
Sami2 said:
Has anyone found a side by side raw comparison of f55 and f5 online
Click to expand...

http://www.dvxuser.com/V6/showthread.php?322147-F55-or-F5-RAW-vs-XAVC-comparison-Where/page2

But in any case where there is no extensive grading done, the difference is subtle, usually if there is no grading with RAW you notice slightly more detail and contrast in highly detailed areas. Sometimes the shadows have more details, halos, or other natural color renditions otherwise not seen once put through the XAVC codec.

Since these cameras can record with SR codec, it ought to be included in any comparison as well. It's much better of a codec than XAVC.
 
If you take S-log2 and S-log3 and bring down the black level to zero (using just a fast colour corrector or similar) with both gamma curves do you end up with the same amount of visible noise?
 
Theoretically it should. Personally I have never bothered to try any tests. I just jumped to slog3 because of its compatibility with cineon gamma based workflows. It makes life simple for most people so for me at least, it made sense.

I do like boosted shadows sometimes, it was much like Cine4 gamma on the F3 which lifted shadows somewhat which in effect it you were exposing by "eye" or monitor caused you to slightly underexpose all together. Of course monitoring in slog exposing by "eye" can be difficult so that's another story.

In general because I do feel that slog requires more meticulous exposure than saw RAW, I swallowed my pride (and ego) and invested in a good post production noise reduction software plugin because sometimes the unexpected can happen and I'd rather have a way to deal with it if needed.

Having said that, I have never really seen any visible noise in slog3 footage in either my own or other clips online. Maybe people are better at exposure and post than I am, but either way, slog3 seems the way to go for at least the reasons I mentioned.
 
http://www.youtube.com/watch?v=VvMh5KKvoPk&index=4&list=UUH8br3MZz9ONIQaw6KAQYBA here is the test for nsoltz in which they got the iso "wrong"

I have been shooting raw with fs700 and i really hate the amount of work i have to put into making the shadows look natural and good, so it was super refreshing to use slog3 and get really nice looking shadows that have some nice form really easily :)

The problem i face is convincing myself that the switch to f5 has solid ground and i will not be dissapointed later on that i did not spend extra 10k for f55.
 
I don't think he is using CineEI mode which is what everyone was talking about which uses the cameras native 2000ISO. The C300 in this test was not shot with native ISO 850 either and was shot at 400 ISO. In Creative mode you can adjust the ISO as normal.
 
Definitely was shooting Custom mode, meaning he could rate the camera at 1250. Original implication was that the reviewer was shooting in CineEI mode, meaning he couldn't shoot the F5 at ISO 1250.

Ned
 
You must log in or register to reply here.
Back
Top