FS7: What is the actual base sensitivity of the FS7

Sam, thanks for kind of doing what I'm asking for, but I can't tell anything at all about noise from a jpeg photo of the car. Noise evaluation requires moving video.
You also don't say what we are looking at? Is this S-LOG2, S-LOG3, HG, STD? What was your method of exposing? Levels? How it has been processed in post? What codec? All of these things matter.
I appreciate the effort, but this doesn't demonstrate anything at all to me. It may have been useful to you, but I don't get anything out of three frame grabs.
 
Doug Jensen said:
Sam, thanks for kind of doing what I'm asking for, but I can't tell anything at all about noise from a jpeg photo of the car. Noise evaluation requires moving video.
You also don't say what we are looking at? Is this S-LOG2, S-LOG3, HG, STD? What was your method of exposing? Levels? How it has been processed in post? What codec? All of these things matter.
I appreciate the effort, but this doesn't demonstrate anything at all to me. It may have been useful to you, but I don't get anything out of three frame grabs.
Click to expand...

Do you think there's no noise difference between an image lit and exposed to 800EI and one lit and exposed to 2000 EI ? None? At all? Certainly, by exposing to 800EI, one sacrifices dynamic range, no argument. But if your shooting environment is controlled, and you won't clip the highlights with a stop or two lost, does that even matter?

No difference in noise at all between 800EI and 2000EI? None?
 
I never said that, nor do I think such a statement is true. Please do not put words in my mouth.
 
Last edited:
Doug Jensen said:
I never said that. Please do not put words in my mouth.
Click to expand...

I didn't put words in your mouth. I asked a question. Which, by the way, you avoided answering. So does that mean you agree there is a noise difference between 800EI and 2000EI (respectively properly exposed)? If we agree these is *some* difference, is it not valid for colorists and DoPs to prefer one noise profile over the other and expose for it, to their taste?

I'm asking questions because I don't understand your position, not to be rude.
 
Last edited:
Yes, I think there is a difference in noise. But how much difference? Does it matter? Is it worth the trade-offs? Noise is not eveyrthing worth paying attention to. I do not find the FS7 (2000) or my F55 (1250) at all nosiy running at the EI or ISO that Sony recommends.

Hey, I'm just asking questions as well. But if someone cares to answer them, I would like to see some actual proof in a real-world situation to back it up. S-LOG and Hypergammas are curves. They are not linear. If you put highlights or skintones on parts of the curves where they are not designed to go by the designers of the curves, then lower noise or not, the actual overall results may be inferior. And I'm not saying someone who says otherwise is wrong, I'm just saying, prove your point.
 
I always shoot in Cine EI mode and have my camera set to ISO1600 to the middle position of the gain switch. This gives me a slightly lower exposure viewed compared to the actual recorder ISO2000. Noise up to 60fps is never a problem to me. I also think the FS7 gains on pushing exposure so that's how I roll.
 
Doug Jensen said:
Sam, thanks for kind of doing what I'm asking for, but I can't tell anything at all about noise from a jpeg photo of the car. Noise evaluation requires moving video
Click to expand...

What you are looking at is enough evidence to make an observation or to consider yourself or the OP performing their own tests.

They are jpg12 in dropbox so jpg compression does not significantly degrade the image.

Tech and workflow..

The car shot; Slog3 CineEI at noted values LUT on - open the iris until zebras set at 90 illuminated the front car wing as per AChap exposure method. Image processed direct with CatBrowse using meta for LUT and EI values

The selfie.
Slog3, LUT off WFM evaluated and seen similar to the 'input' WFM on each image, consistent lighting.

Image processed in Resolve using LC709a and my correction mainly using the Offset Wheel.

The second image (the selfie) to me is most telling as the 'thin' negative seems not only noisy but soft too.

--

My opinion of the thick neg selfie (bent down in post) is that the image is crisp enough to be on par with top end cameras. The imperfections in my skin are rendered with horrifying accuracy.


--
My last short I felt looked a little soft/dark - I think the LUt on the DP502 led my to underexpose - I should have relied more on the camera WFM


My guess is that XAVC compresses the shadows more heavily as they are (in the sony head) less 'important' to the image (as colour info is less important than luminance).. it is therefore way better to push down in post than lift.

One would only want to lift in extreme DR situations or where one is plain short of light.

S
 
Last edited:
Doug Jensen said:
Yes, I think there is a difference in noise. But how much difference? Does it matter? Is it worth the trade-offs? Noise is not eveyrthing worth paying attention to. I do not find the FS7 (2000) or my F55 (1250) at all nosiy running at the EI or ISO that Sony recommends.

Hey, I'm just asking questions as well. But if someone cares to answer them, I would like to see some actual proof in a real-world situation to back it up. S-LOG and Hypergammas are curves. They are not linear. If you put highlights or skintones on parts of the curves where they are not designed to go by the designers of the curves, then lower noise or not, the actual overall results may be inferior. And I'm not saying someone who says otherwise is wrong, I'm just saying, prove your point.
Click to expand...

Understood. Such is your opinion. I leave you with a quick quote from Alister Chapman, who has spent considerable time with CINE EI on the the F5/F55:
http://www.alisterchapman.com/2013/12/28/cine-ei-mode-when-recording-s-log23-and-raw-on-the-f5-and-f55/

Alister Chapman said:
When I shoot with my F5 I normally rate the camera as 800EI. I find that 5 stops of over exposure range is plenty for most situations and I prefer the decrease in noise in the final pictures. I rate the F55 similarly at 640EI. But please, test and experiment for yourself.
Click to expand...

And yes, he's speaking about the F5, not the FS7, but we're talking about the same imager, with the same 2000ISO native.
 
Last edited:
I've decided to post up my ISO sensitivity test results from the F5 (which theoretically should be pretty much identical to the FS7, assuming they do share the same sensor), hopefully they can clear things up for people, as I've seen nothing but anecdotal evidence on the matter to date.

You can check out the tests on my blog: http://dreamsmiths.com.au/blog/
 
Doug Jensen said:
Ebrahim, I'd be very interested to hear what you discovered in your detailed testing of the FS7. Do you disagree with Sony's numbers for Cine EI, Hypergamma, or Standard?
Click to expand...

No I haven't touched an FS7 yet, merely giving advice to people who seem to be conflicted on what ISO value to use on a camera. Test, that's the only way to be 100% sure what works for you.

I've worked with a lot of cameras where I found their rated base ISO doesn't produce the best images (most notably C100, C300, Sony A7s, Red MX) especially under different non-ideal shooting situations, and by testing ISO values and how they affect dynamic range, noise and colour reproduction I was able to choose the one that works for me. I advice not taking manufacturers words for it and just stick to a certain gain level as one might be missing a lot and making his life harder. It's a trade off between noise, headroom, shadow noise floor, and colour fidelity, choose your best balance.
 
Grug said:
I've decided to post up my ISO sensitivity test results from the F5 (which theoretically should be pretty much identical to the FS7, assuming they do share the same sensor), hopefully they can clear things up for people, as I've seen nothing but anecdotal evidence on the matter to date.

You can check out the tests on my blog:http://dreamsmiths.com.au/blog/
Click to expand...

That's like saying in F1, a Lotus or a Williams should be as fast as a Mercedes since they all have the same engine.

The FS7 may have the same sensor as the F5, but it is designed by a different team, in a different factory and has different colour science.

Finally, your tests may well be accurate, but you really need a hardware waveform, and a proper 18% grey target that is properly lit.
 
Ebrahim Saadawi said:
No I haven't touched an FS7 yet, merely giving advice to people who seem to be conflicted on what ISO value to use on a camera. Test, that's the only way to be 100% sure what works for you.

I've worked with a lot of cameras where I found their rated base ISO doesn't produce the best images (most notably C100, C300, Sony A7s, Red MX) especially under different non-ideal shooting situations, and by testing ISO values and how they affect dynamic range, noise and colour reproduction I was able to choose the one that works for me. I advice not taking manufacturers words for it and just stick to a certain gain level as one might be missing a lot and making his life harder. It's a trade off between noise, headroom, shadow noise floor, and colour fidelity, choose your best balance.
Click to expand...

I won't comment on custom mode because I never use it, but the fact is that if you are shooting in Cine-EI mode, you are shooting at EI/ISO 2000 no matter what you do. Which is why it can sometimes be dangerous giving blanket advice when you haven't used and lived with a certain camera. While there is a 'proper' mapping of values within the S-Gamut and the SLog3.cine colour space that you might want to expose for, or you might take a creative decision based on the scene to over expose what will eventually be the shadows to make them cleaner and/or to ND and under expose highlights or whatever, it is all done at EI/ISO2000 and everything else you're doing is pushing and pulling the image to taste like one normally would.

I don't think it shooting XVAC-i in s-log is even vaguely comparable to shooting on Red and using redcode raw.
 
Zeroth law of Base ISO. 50% of Base ISO made by Engineering the other 50% made in the marketing dept.

This is not as flippant a point as it may first appear.
 
My point was more that Cine-EI give you a number it calls EI200/ISO2000. It doesn't matter if that number is accurate because it can't be changed and one has all the tools and readings to expose accurately for that number and in addition can calibrate a light meter (shock, horror!) to be accurate to that specific camera too.
 
Viddovation said:
The ISO that the FS7 considers to be "zero gain" or "base" varies depending on the gamma mode you're using. For instance, set to Custom/SLog2, it offers ISO2000 as base. Set to Custom/Standard, it offers ISO1000 as base. Set to CineEI/Slog2, uses 800EI as base.

I would suspect your camera is in CineEI after the factory reset?
Click to expand...

No. Me thinks it goes to custom, which is another point of confusion to me.
But then I get confused by mirror focus.
 
Guys owning/having access to F5. Could you test and post some short samples (video) of footage in CineEI - no NR, low NR, mid and high NR?
 
You must log in or register to reply here.
Back
Top