F3 Scene Files from AbelCine

We'll see what we can do, but Andy does have a day job! ;-)

There will always be a lot more available out of the camera in a LOG mode.
 
jcskeeter said:
http://vimeo.com/22540499

Don't know if you've watched this already, but I thought it was pretty convincing. Really good side by side comparisons of XDCam vs. S-Log.
Click to expand...


This is a perfect example of the discussion about the F3, Alexa, 35mm, Red etc test on the danger of comparison tests.
No - its a terrible comparison of SLog to to XDCam because the XDCam is obviously overexposed. Anyone used to XDCam knows it loooks much better than those tests.
 
Yes that test is a useless demo because it isnt a test at all. The key issue being the word GRADED with the Slog. Grade the xdcam as well and then we are talking. You have to shoot a chart as well just to keep a control, but both types of testing are needed.
 
Lenilenapi said:
This is a perfect example of the discussion about the F3, Alexa, 35mm, Red etc test on the danger of comparison tests.
No - its a terrible comparison of SLog to to XDCam because the XDCam is obviously overexposed. Anyone used to XDCam knows it loooks much better than those tests.
Click to expand...


This is true. I agree with what your saying too. After my post I thought about that.
 
I hate to say it because I know a few of the guys who were out there and shot this, but I wouldn't really call it a test persay maybe test run? but its not really a test to show you anything. The S-log was graded, which means we are seeing nothing to compare to with ungraded xdcam footage. I have to say though I'm noticing A LOT of people are over exposing their xdcam footage. Making it very hard to judge anything xdcam lately.

Also I think recording 10bit 4:2:2 uncompressed or at very low compression would also give you much more to work with over the xdcam footage.

Yesterday I attended a meeting with Sony and one of their engineers, we mostly talked about the gamma curves with the F35 and F35 2.0 software, but the F3 was talked about too. I'll post some info shortly that might clear up some misunderstandings about gamma curves.

You still need to expose properly with a waveform. There is no way to expose properly with a light meter, its just never going to be right. You need a waveform hooked up at all times. Only then can you do any kind of test at all that would represent the differences.
 
Giuseppe Pugliese said:
I
You still need to expose properly with a waveform. There is no way to expose properly with a light meter, its just never going to be right. You need a waveform hooked up at all times. Only then can you do any kind of test at all that would represent the differences.
Click to expand...

Amen.

In general a waveform is the only way to be accurate about video exposure. Histogram is helpful as well as zebras but only helpful. I wish Sony would offer an in camera waveform like Panasonic. How hard can it be?

I'm curious how people expose for cinegammas. Normally at rec 709 I keep zebras at 100 IRE to check for whites and 60 for caucasian flesh, but the 100 IRE zebra is pretty useless for the cinegammas because you've already reached extreme roll off by then. Likewise caucasian faces need to be lower than they are for 709.

What do you guys do when you don't have a waveform especially outside running around.
 
This is starting to get off topic of the OP but oh well... It seems kinda interesting to me about the difference in the graded s-log and the non graded XDCam. I don't know what the s-log looks like before grading but in a way it does show you a difference in the latitude of the s-log when compared in the manner they are compared in the "Test/Demo". I'm curious though? Would one be exposing differently if just shooting XDCam? Or could the XDCam footage in the video be match similar to the s-log? Obviously without as much latitude.

"Tests" are "Test", I get that. But it would be interesting to see the graded s-log next to a properly exposed and graded XDCam. Just a thought...
 
Wait wait now, isn't Slog just a gamma setting anyway? This would be recorded on the sxs as well. The best way to do it, is to do a test with a KiPro, shooting with Cine2 or 1, and then Slog, both graded, then showing the same shots, graded, that came off the sxs. Im concerned about the upgrade costs compared to just shooting with Cine1 or 2. One of the Sony guys said that with Slog it is a 12 stop camera. Isnt this the same thing Abel saw without Slog? What is it actually giving me for $3500? I dont care about 444, btw.
 
Cinesin,

S-Log gives you a full stop extra,. So if we had 11 (nearly 12) stops before we have 12 (nearly 13) with S-Log. Plus you have the advantage of a flat mapping of the sensor which allows for the most correction in post.

Andy
 
I was sure I saw a 12 stop test at Abel's site at some point, pre Slog.

So its 11.5 and we get around 12.5 with Slog, ok.

I think to sell this upgrade Sony should sponsor a real test. Im holding onto my money until I see $3500 worth of proof. Maybe someone should let one of us early buyers do a street test!!

Btw, what does a true 13 stop range compete with? F35? RED (and Epic in normal mode?), SI2K? F950?
 
You must log in or register to reply here.
Back
Top