Sigma Fp review and interview / Cinema DNG RAW

[paulinventome]

3 minutes ago, Chris Whitten said:

If I push the exposure to the extreme (in my opinion) the blotchy green and magenta is there in both Resolve settings. 

Yup. Good. Now shoot the same scene in 12 bit and compare. Then be amazed at what the camera *can* do in the shadows...

This is why i want sigma to do a  curve so they can squeeze 12 into 10 so we can shoot 12bit at 25p...

cheers
Paul

[Chris Whitten]

My latest clip, correctly exposed in the camera IMO.

Using the Cinema DNG setting in Resolve, then pushing the exposure to extremes, about 5 stops. Green and purple blotching.

[Chris Whitten]

Yes, I'm also using Rec709 2.4 gamma.

 

I have been using Blackmagic Film because 'Crimson Engine' on Youtube said that's how he was getting best results and on trying it myself I found it looked good.

[Chris Whitten]

I will try 10 and 12 bit.

I really wanted to keep the file sizes down if possible. So I would rather use an effective 10 bit going forward.

In addition, it easier to see this 'bug' when pushing the exposure to extremes. I would never want to do that in the real world. The issue is going to be, is the green/purple blotching in everything - well exposed clips, correctly graded, but just very hard to see. That would still be a bug IMO

[Scott_Warren]

No offense to Crimson Engine, but it's a strange suggestion for people to use an incorrect colorspace as a starting point with a camera that has pretty great color in its own right as Paul has mentioned! (Something can look good with weird settings, but you always want to strive for base technical accuracy + aesthetics in my opinion.)

You might also consider using the ACES pipeline since it has a film print like "look" built into the display chain already. It takes the color matrix built into the DNGs and scales them for output for your monitor without weird translations or conversions except to ACES' own massive color space.

You'd be free to add in a Colorspace Transform node and try out different camera color sciences to see if something else is more to your liking (Alexa, RED, Sony, etc), but I've found it's a quick and repeatable way to get excellent color out of the camera quickly. 

The attached shot was achieved just by dropping in the sequence and pushing exposure and saturation a bit to match to what I saw in person. (Metered the window light with an external incident meter as I tend to do.)

[Chris Whitten]

I've used ACES for a while and now stopped using it actually.

There is no specific profile in Resolve 16 for this camera, so I think it's ok to test and see what works.

[Scott_Warren]

1 minute ago, Chris Whitten said:

I've used ACES for a while and now stopped using it actually.

There is no specific profile in Resolve 16 for this camera, so I think it's ok to test and see what works.

Fair enough! The joy of this stuff is there aren't any rules If it looks right, it is right. 

[Chris Whitten]

Here is the latest test.

Both ISO 640

4K 10 bit at 25fps with a HOYA IR Cut filter, exposure pushed to extremes in Resolve.

Then 4K 12 bit at 23.95fps with the IR Cut and exposure pushed in Resolve. To me both of these look a lot cleaner, the 12 bit very clean.

I would never push the exposure this much. To me, the IR Cut filter reduces the purple and green blotch by quite a lot.

I mean the whole image looks rough by simply pushing the exposure by 4 or 5 stops. Properly exposed and graded I think you wouldn't see the purple and green issue?

[Chris Whitten]

The 10 bit clip with just a bit of lift, gamma, gain, a bit of contrast and a little added saturation. 

CinemaDNG default, no LUT, no other adjustments. I think it looks ok. No doubt the 12 bit looks even better.

[paulinventome]

42 minutes ago, Chris Whitten said:

I would never push the exposure this much. To me, the IR Cut filter reduces the purple and green blotch by quite a lot.

My point is that the very lowest shadows going green/magenta shouldn't really be happening.

If we assume that the sensor is in 12 bit mode and the sensor data that the camera is getting is the same as we see with a 12 bit cDNG then we can assume that these 10bit shadows are a software bug because 12bit cDNG looks perfect. That's really my point. And i think if you have a night scene and you are on the bleeding edge of exposure and can't shoot 24p then you are going to hit this problem just trying to expose normally.

Now if sigma are unable to output 12bit at 25p then the solution that works for everyone is to do 10bit but to use a curve to redistribute values so that the last stop of recorded light does not take 2048 integer values to describe it and then use the saved values to ensure the shadows get every last drop of sensor data they can. Then we'd have 12 bit quality in a 10 bit container at all these frame rates and EVERYONE is happy. There is literally no downside to doing this. This is what 8 bit is doing and that can produce great images in a fraction of the space. The code is already in the camera!

If you see earlier on in this thread i even show what that curve looks like because each DNG file has a copy of it in 8bit so Resolve can debayer properly. It's standard practice in cDNG.

We need a wave of people to say this to sigma because sometimes i think it's only me....

cheers
Paul

 

[paulinventome]

1 hour ago, Scott_Warren said:

You might also consider using the ACES pipeline since it has a film print like "look" built into the display chain already. It takes the color matrix built into the DNGs and scales them for output for your monitor without weird translations or conversions except to ACES' own massive color space.

Yes, ACES works well once set up correctly.

I just use IPP2 because i'm mostly Red based and find it nice than ACES.

cheers
Paul

[Scott_Warren]

Quick aside about dialing in a better starting point about color, short of having an ACES IDT any time soon:

The OneShot chart is one of the cheaper but fantastic options to get to a good starting point to balance shots together, or to make sure everything in a cut is consistent so that a global look can be authored and consistent without shot-to-shot tweaking. The ink colors are calibrated to match 50% of the saturation of the standard video colors in the vectorscope for alignment. It's also glossy, when helps you get true white and black instead of a matte finish which would only give you lifted black values.
http://dsclabs.com/product/oneshot-plus-2/

For the clip above, I used the auto-match tool in Resolve followed by a quick refinement of the vector directions after that. Nothing else. ACEScc pipeline with a Rec709 output. 

[Chris Whitten]

I watched a video comparing Aces to non-Aces workflow in Resolve and ACES did look more unwanted yellow to me, no matter what you did in the grade.

And yes, as I said earlier, if this purple/green effect is happening in normally exposed, well graded footage but can't be seen it IS still a bug.

I do think the IR Cut filter helps.

Finally, I think people can take a bit less judgmental and lecturing tone, as I have shot a lot of stuff over the last two days on demand "can you try this, can you try that", I have uploaded clips which take 4 hours to upload and cripple my system so I can't do anything else online while I'm waiting for it to finish uploading.

Yes, I understand ACES, yes I understand Blackmagic Film isn't designed for Sigma DNG, yes I understand well lit and well graded footage may still have an issue in the shadows.  Let me know if you see any flickering in the ISO 640 clip I uploaded.

[Scott_Warren]

4 minutes ago, Chris Whitten said:

I watched a video comparing Aces to non-Aces workflow in Resolve and ACES did look more unwanted yellow to me, no matter what you did in the grade.

And yes, as I said earlier, if this purple/green effect is happening in normally exposed, well graded footage but can't be seen it IS still a bug.

I do think the IR Cut filter helps.

Finally, I think people can take a bit less judgmental and lecturing tone, as I have shot a lot of stuff over the last two days on demand "can you try this, can you try that", I have uploaded clips which take 4 hours to upload and cripple my system so I can't do anything else online while I'm waiting for it to finish uploading.

Yes, I understand ACES, yes I understand Blackmagic Film isn't designed for Sigma DNG, yes I understand well lit and well graded footage may still have an issue in the shadows.  Let me know if you see any flickering in the ISO 640 clip I uploaded.

I'd be careful to assume any of us are being judgemental or lecturing, Chris. Online forums, particularly technical ones, tend to have a lot of cut-and-dry information for information's sake. Suggestions from people with different experiences aren't necessarily judgemental. We're all just sharing what we know! I don't have a dog in this fight. I just like making images and sharing what I've learned

[Chris Whitten]

Well I'm reading less comments on my clips, what is going on in them, and more commentary on what I should be using 12 bit and ACES.

I'm not making a movie, I'm exploring the camera myself, I'm also doing a lot of test shooting so we can ALL figure out what is going on.

 

[Chris Whitten]

16 minutes ago, Scott_Warren said:

Suggestions from people with different experiences aren't necessarily judgemental. 

Sure, I'm a professional creative. I'm used to suggestions from other more experienced people.

I don't think ACES changes anything about the flickering, or the green/purple bug in shadows.

I often use ACES, I am now going back to REC709 Gamma2.4 for a bit.

I thought it was a reasonable suggestion from someone else to try the Blackmagic Film drop down, and to think of the native ISO as 640. Both of which I've tried myself and had good results with.

[Chris Whitten]

In the ACES colour space and Cinema DNG Default settings I am not getting anything like the same pronounced purple/green blotchy pattern in the blacks after shooting with the IR Cut Filter. This is the 10bit clip with the exposure increased a lot using the wheels, and pulling up the 'shadows' value.

[BTM_Pix]

I'm not sure how much this is going to add to the insight regarding the purple/green issue really as its not a particularly controlled test but here are some 8bit and 10bit (8bit on top) frames of mine that I've chosen that I would expect to show the issue as they contain a lot of dark fabric etc. 

Both of these have been pushed 3 stops and don't show the issue to my eye to any dramatic degree and seem pretty consistent between the two file types.

For what its worth, all of my other Sigma cameras (including both the Merrill and Quattro variants of the Foveon sensor) exhibit this problem far, far more dramatically so I was very concerned when I saw it pop up in this thread that it might be carried over to their CMOS cameras !

However, the bottom frame is 8bit, again pushed 3 stops, and it is visible in the top right hand corner but as with the other two frames the whole image is obviously destroyed for all practical purposes anyway at that point so I can live with it.

[Scott_Warren]

4 minutes ago, Chris Whitten said:

In the ACES colour space and Cinema DNG Default settings I am not getting anything like the same pronounced purple/green blotchy pattern in the blacks after shooting with the IR Cut Filter. This is the 10bit clip with the exposure increased a lot using the wheels, and pulling up the 'shadows' value.

Now this is interesting. I wonder why a colorspace change would have such a dramatic effect on shadow colors? I can't imagine the colorimetry of the BM sensors is THAT different from the fp's when it comes down to numbers. Of course I'm also not an engineer!

[paulinventome]

1 minute ago, Chris Whitten said:

In the ACES colour space and Cinema DNG Default settings I am not getting anything like the same pronounced purple/green blotchy pattern in the blacks after shooting with the IR Cut Filter. This is the 10bit clip with the exposure increased a lot using the wheels, and pulling up the 'shadows' value.

Your 640ISO doesn't flicker. That's correct right? I need to redo mine to double check that 320 was doing it all the time but 400 was the trigger factor on my camera.

I think there are ways you can get around purple/green and most of them, like BMD Film, involve crushing the shadows and possibly desaturating them,  so they're not so obvious. @Lars Steenhoff suggested incorrect black levels but it's not that. I suspect ACES is crushing a bit and an IR Cut filter may leech some colour out overall, you could desaturate the shadows in Resolve.

The blotchiness is a factor of 10 bit shadows, each one of those blotches is a single shadow value. In most codecs and compressed formats you would never see this because noise is breaking it all up! But uncompressed shows you what's really there and of course you can add noise to break up the blotches should you want too.

I don't want to keep repeating because i will sound like a broken record but if you take a 12 bit file and a 10 bit file that has the same grey point then what you will see in terms of actual finite values are:

Stop 0 : 1 value
Sopt 1 : 2 values
Stop 2 : 4 values
Stop 3 : 8 values
Stop 4 : 16 values

Okay, if 10 bit and 12 bit have the same 18% exposure brightness level (they do, in the DNG) what the 12 bit is giving you are 2 more stops of shadows. So a value in 10 bit which is at stop 2 is 4 values but that same stop in 12 is 16 values. This is why, when you compare 10 bit to 12 bit (look at my lens images) the shadows in the 12 bit are not blotchy, but a lot more tonality. Now if i pushed 12 bit a further two stops then i would see similar blotchiness but the colours are more accurate (not green/magenta)

I hope that makes sense to anyone still wrapping their head around things....

cheers
Paul