SIGMA FP with ProRes RAW and BRAW !

[Llaasseerr]

45 minutes ago, OleB said:

Actually the result for ISO 100 and ISO 800 settings is exactly the same. False colors are not changing at all so if you expose as per that, all is identical. They first do so if you go one step higher than ISO 800. Only thing you have to consider in FCPX is that if you directly shoot in ISO 100 you can leave everything as it is, if you shoot in ISO 800 and want to bring back the picture, because in FCPX it will look blown of course (being as you said ISO 100 pushed 3 stops), you can either select manual ISO 100 or just dial down the highlights. By now I prefer the later way. 

I decided to take the ISO 800 and upwards approach, because there the false colors are accurate and additionally the screen preview is at least workable. 

Do you mean the exposure looks the same on the Ninja V screen? because I understand that 800 image is shot at ISO 100, but in my mind it's pushed +3 stops right? I guess I would understand if it was in front of me.

False color aside:

My observation about the clipping point of ISO 100 being around 3.5-5 is based on looking at the DNGs, so I was thinking that it would be applicable to the Ninja V recording too - but perhaps not, maybe that is (sadly) apples to oranges. Basically if I pushed an ISO 100 DNG by +3 stops to emulate what you're talking about with ISO 800, the highlights would clip much higher as I mentioned. But this is all theoretical until I have the camera.

[Llaasseerr]

1 minute ago, Owlgreen said:

It really wasn't. I've seen the words "ruin the color" multiple times on multiple forums in reference to using anything other than P3 D60 and a CST. If you didn't say those words then my comment didn't have anything to do with you.

Ok, thanks and I'm sorry for jumping to conclusions.

Basically my point with using the ACES input transform is that it will take into account the native sensor color space and embedded metadata to make a more informed colour transform, and that BMD Film is not important in that context. I think people see a linear image in Resolve and it looks blown out and they freak out, but that's because linear is not designed to be viewable. BMD Film is bringing an image into a viewable log range, but for example in ACES (and probably a managed Resolve workflow) you can place a non-destructive view transform on the viewer as a preview, and retain the full linear data.

Or transform to log, that's perfectly valid too. My issue with BMD FIlm Gen 1, is did BMD ever publish their log curve and colour gamut? I don't think they did. So in my mind it's better to use something like Alexa logC/AWG. If it's a technically accurate transform from linear toa known log curve with a white paper, and the linear input is a technically accurate transform from the raw, then the image integrity has been maintained and you know where you stand.

[OleB]

20 minutes ago, Llaasseerr said:

Do you mean the exposure looks the same on the Ninja V screen? because I understand that 800 image is shot at ISO 100, but in my mind it's pushed +3 stops right? I guess I would understand if it was in front of me.

False color aside:

My observation about the clipping point of ISO 100 being around 3.5-5 is based on looking at the DNGs, so I was thinking that it would be applicable to the Ninja V recording too - but perhaps not, maybe that is (sadly) apples to oranges. Basically if I pushed an ISO 100 DNG by +3 stops to emulate what you're talking about with ISO 800, the highlights would clip much higher as I mentioned. But this is all theoretical until I have the camera.

If you ask me the best explanation would be that for true RAW, via Ninja V, the ISO values of the camera are incorrectly transferred/named. The best hint we have is that manual from Sigma. The real ISO values of both channels are 100 and 640. So it seems for the cine mode they have decided to name them differently. 

So basically you have to dial in ISO 800 and get ISO 100 and so on, as in my table which I have mentioned way above. Comparing those "real" ISO values with an incident meter gives a match between false colors and the meter.

Do not ask my why they have done this...

[Llaasseerr]

11 minutes ago, OleB said:

If you ask me the best explanation would be that for true RAW, via Ninja V, the ISO values of the camera are incorrectly transferred/named. The best hint we have is that manual from Sigma. The real ISO values of both channels are 100 and 640. So it seems for the cine mode they have decided to name them differently. 

So basically you have to dial in ISO 800 and get ISO 100 and so on, as in my table which I have mentioned way above. Comparing those "real" ISO values with an incident meter gives a match between false colors and the meter.

Do not ask my why they have done this...

That makes some sense, thanks. I saw this dpreview post recently (possibly this is old info here, not sure) and I wasn't sure if it was just applicable to photo mode. I'm referring to the spreadsheet in the first post:

https://www.dpreview.com/forums/thread/4556012

I haven't cross checked it with what you posted before, but this seems like a legit source from Photons to Photos.

[OleB]

13 minutes ago, Llaasseerr said:

That makes some sense, thanks. I saw this dpreview post recently (possibly this is old info here, not sure) and I wasn't sure if it was just applicable to photo mode. I'm referring to the spreadsheet in the first post:

https://www.dpreview.com/forums/thread/4556012

I haven't cross checked it with what you posted before, but this seems like a legit source from Photons to Photos.

I came to this because one year ago I was wondering why the false colors of the Ninja V are not working correctly above ISO 100. My expectation was that once you dial up the ISO (let us say in full steps) that the clipping should occur one stop earlier simultaneously. But I saw that false colors did not change at all until ISO 800 and only afterwards the behavior started which I was expecting.

Next thing I checked was if there is headroom (pulled the files in FCPX) over the clipping point when doing that, there is nearly none if you dial in the exposure to the border of clipping as per false colors. So far so good. 

Then I built a lighting setup where I exposed for the practicals just so that they do not clip. Think that was on ISO 3200 setting in that first trial. I took my incident meter to dial in my key (of course setting that thing to ISO 3200) and guess what...key totally under. Confusion started and I was wondering if I am too stupid to understand the camera somewhat. Should at that stage better used the false colors to check skin color...

But instead giving up I was investigating further and in the end figured out what in the end led to the conversion table. Now I can use my incident meter again for setting the lights, contrast rations etc. and last but not least, the false color screen is now confirming that I am ending in the correct exposure range doing so.

[Ryan Earl]

Sigma FP Color Mode "ON" Portrait, Standard, etc - UHD DNG - WB - Daylight

Sigma FP Color Mode "OFF" Portrait, Standard, etc - UHD DNG - WB - Daylight

It's a shift that could be meaningful for some, the violet spray bottle on the left side of the image stays true with color mode set to anything but 'off.'  The reds also get orange and the greens go a little brown.  It's not just less saturated, I don't know, it seems shifted.  

Then 'ON' and 'OFF' will match BMD color and gamma.  Where you will see both or a difference is in ACES or DaVinci Color Managed.

It's a strange behavior, that's why I think they went in reverse from BMD film to the color mode "off" which I think was added later.

[Llaasseerr]

15 minutes ago, Ryan Earl said:

Sigma FP Color Mode "ON" Portrait, Standard, etc - UHD DNG - WB - Daylight

Sigma FP Color Mode "OFF" Portrait, Standard, etc - UHD DNG - WB - Daylight

It's a shift that could be meaningful for some, the violet spray bottle on the left side of the image stays true with color mode set to anything but 'off.'  The reds also get orange and the greens go a little brown.  It's not just less saturated, I don't know, it seems shifted.  

Then OFF will match BMD color and gamma.  Where you will see both or a difference is in ACES or DaVinci Color Managed.  They will look as they do in 'off' if developed through in BMD Film regardless if 'off' was set in a DaVinci YRGB project where you can decode to it.

It's a strange behavior, that's why I think they went in reverse from BMD film to the color mode "off" which I think was added later.

Thanks for showing that shift, it looks like a gamut mismatch. I mean, it seems best to rely on the "none" mode.

Re: transforming to BMD/BMD Film in the RAW tab, actually it's so long since I looked at it and I see it's not even a log curve any more. It seems to be putting the image into a display space immediately, which is not the result I would expect.

 

 

[Llaasseerr]

38 minutes ago, OleB said:

I came to this because one year ago I was wondering why the false colors of the Ninja V are not working correctly above ISO 100. My expectation was that once you dial up the ISO (let us say in full steps) that the clipping should occur one stop earlier simultaneously. But I saw that false colors did not change at all until ISO 800 and only afterwards the behavior started which I was expecting.

Next thing I checked was if there is headroom (pulled the files in FCPX) over the clipping point when doing that, there is nearly none if you dial in the exposure to the border of clipping as per false colors. So far so good. 

Then I built a lighting setup where I exposed for the practicals just so that they do not clip. Think that was on ISO 3200 setting in that first trial. I took my incident meter to dial in my key (of course setting that thing to ISO 3200) and guess what...key totally under. Confusion started and I was wondering if I am too stupid to understand the camera somewhat. Should at that stage better used the false colors to check skin color...

But instead giving up I was investigating further and in the end figured out what in the end led to the conversion table. Now I can use my incident meter again for setting the lights, contrast rations etc. and last but not least, the false color screen is now confirming that I am ending in the correct exposure range doing so.

Okay, that is really weird. So visually, as you dialed up the ISO the image got +1 stop brighter on the Ninja V but the false color was still monitoring the underlying base ISO?

In my mind, as you increase ISO you increase the highlight clipping point and simultaneously get noisier shadows, assuming ISO invariance. The test you sent me bears that out, but the max highlight value did not increase at a full stop per ISO increase - so that seemed strange.

But it's great that you got that consistency and repeatability now, by doing tests with your light meter to sanity check. I will need to come back and read your posts if/when I get the camera so I can fully appreciate it. I mentioned that my existing method for sanity checking has involved a basic light meter reading to set a general exposure for middle grey, and verifying with my grey card that it hits 0.18 when developed to a linear image. And also ideally that the green of the false color hits the grey card when monitoring.

[Ryan Earl]

4 minutes ago, Llaasseerr said:

Thanks for showing that shift, it looks like a gamut mismatch. I mean, it seems best to rely on the "none" mode.

It's confusing, I might have written it out in a confusing way, but I don't think the "off" mode or decoding to BMD color and gamma are as accurate, so personally I'm not using them.

I set the profile in the camera to "Portrait" then record in 12 Bit DNG then use DaVinci Color Managed because I'm often using it for reproducing objects that require accurate color. 

I would consider the "off" mode to be a shifted 'cinema' mode like a Kodak film stock, where green isn't green, it's a little brownish green.  It could be better for skin tones for example and pleasing, but I don't see it as 'accurate.'

[Llaasseerr]

18 minutes ago, Ryan Earl said:

It's confusing, I might have written it out in a confusing way, but I don't think the "off" mode or decoding to BMD color and gamma are as accurate, so personally I'm not using them.

I set the profile in the camera to "Portrait" then record in 12 Bit DNG then use DaVinci Color Managed because I'm often using it for reproducing objects that require accurate color. 

I would consider the "off" mode to be a shifted 'cinema' mode like a Kodak film stock, where green isn't green, it's a little brownish green.  It could be better for skin tones for example and pleasing, but I don't see it as 'accurate.'

Interesting, because Sigma touted "off" mode as the closest profile to the raw DNGs, but maybe you're on to something. I'll admit I'm set in my ways with developing DNGs via ACES and I find DaVinci Color Managed unintuitive and confusing. But I'm used to using ACES on commercial projects and using it with Nuke plus other tools. For a whole film color pipeline, no-one uses DaVinci Color Managed, but for projects contained in Resolve I see how it could be valid.

Can you upload these images and I can put them through ACES and post them here?

[OleB]

30 minutes ago, Llaasseerr said:

Okay, that is really weird. So visually, as you dialed up the ISO the image got +1 stop brighter on the Ninja V but the false color was still monitoring the underlying base ISO?

Exactly this! When you are increasing ISO from 100 upwards (no matter what monitor mode you use, even Rec709) the preview picture increases exposure, but the false colors remain the same until ISO 800, only from that you see a change in the clipping point and the exposure of the preview. 😄 

[Ryan Earl]

10 minutes ago, Llaasseerr said:

Can you upload these images and I can put them through ACES and post them here?

Sure, here is a link to 2 DNG Frames.  Let me know what you think.  Anybody can download until the link expires.

https://we.tl/t-uRbyLstW0Y

[Llaasseerr]

9 minutes ago, OleB said:

Exactly this! When you are increasing ISO from 100 upwards (no matter what monitor mode you use, even Rec709) the preview picture increases exposure, but the false colors remain the same until ISO 800, only from that you see a change in the clipping point and the exposure of the preview. 😄 

Okay, you probably stated this already that if it measures the value directly off the sensor then it's not factoring in the ISO push. I guess what is needed is for the false color to work in a CineEI mode where it reflects the preview. Like CineEI on Sony pro cams, it's just the base ISO but the preview can show an adjusted ISO value that is just written as metadata and reflects what you would do in post.

It could be done predictably with a false color LUT, probably. Since that would be applied after the preview picture exposure change.

[Llaasseerr]

1 hour ago, Ryan Earl said:

Sure, here is a link to 2 DNG Frames.  Let me know what you think.  Anybody can download until the link expires.

https://we.tl/t-uRbyLstW0Y

I think you're on to something with this. I think "ON" is better. I think the "OFF" image is displaying a native wide gamut image in "Sigma gamut" and we don't know what the transform is. It doesn't seem like it's accounted for in the DNG metadata when in "OFF" mode. 

Having said that, the ACES "OFF" version looks to me like the mov "OFF" version that I've seen before (in the toy Ferrari images) with the desaturated color space. So in that sense it's consistent? But not what we wold want.

Here are the two images exported as srgb with the default ACES look and output transform. The "ON" version is very close to your "ON" version. There is an issue where the look transform at the end is flattening the highlight detail on the piglet doll's face, but when the file is viewed in linear space, the detail is there. So that could be fixed.

 

 

 

 

OFF

ON

[TomTheDP]

So my first impressions were instant regret. I took some video to find terrible banding in the deep shadows. After updating to the current firmware and playing around a bit more I think I am satisfied. You definitely got to follow an image workflow.

@Ryan Earl suggestion seems to work well.
Set the RAW to the P3 D60 colorspace and gamma to linear than apply the color space transform and set the output to REC709 and gamma 2.4. Adjust the max input on tone mapping to 1000 and you are set.

Setting it to the blackmagic colorspace is a bit quicker to do and looks pretty good. But the colors are vastly different and it doesn't retain information quite as well. Might be good for matching to other BM cameras though.

What I have found is that the shadows fall apart way quicker in both 10 bit and 8 bit RAW. 12 bit seems the way to go. To be fair I was pushing the shadows really hard.

Overall the image is nice and organic. I think I am sold. I really hope sigma can fix the monitoring for 3200 iso, would really make it quite the camera.

Its no Arri Alexa but there is something nice about CDNG. I wish Premiere supported DNG better.

[Llaasseerr]

57 minutes ago, TomTheDP said:

So my first impressions were instant regret. I took some video to find terrible banding in the deep shadows. After updating to the current firmware and playing around a bit more I think I am satisfied. You definitely got to follow an image workflow.

@Ryan Earl suggestion seems to work well.
Set the RAW to the P3 D60 colorspace and gamma to linear than apply the color space transform and set the output to REC709 and gamma 2.4. Adjust the max input on tone mapping to 1000 and you are set.

Setting it to the blackmagic colorspace is a bit quicker to do and looks pretty good. But the colors are vastly different and it doesn't retain information quite as well. Might be good for matching to other BM cameras though.

What I have found is that the shadows fall apart way quicker in both 10 bit and 8 bit RAW. 12 bit seems the way to go. To be fair I was pushing the shadows really hard.

Overall the image is nice and organic. I think I am sold. I really hope sigma can fix the monitoring for 3200 iso, would really make it quite the camera.

Its no Arri Alexa but there is something nice about CDNG. I wish Premiere supported DNG better.

Setting the RAW to Linear/P3D60 and setting the output transform to Rec709/gamma 2.4 is very close to what ACES is doing when viewing/rendering for a Rec709 intent.

The main difference with the underlying linear images is that the ACES gamut is larger than P3. I've attached two screengrabs of the linear image stopped down by -1 stop to see the highlight detail and it looks the same. The P3D60 is the more saturated one. The ACES one looks less saturated without the native gamut transformed to the output device.

One thing I've noticed in the "linearP3D60 to Rec709" version vs the "ACES linear to Rec709 ODT" version is that the highlights appear to be clipping because the CST does not provide any highlight rolloff except the tone mapping, which is not quite getting the highest values - whereas the ACES rrt/odt includes an aggressive highlight rolloff.

A non-ACES method that preserves the highlights would be to CST transform to Alexa logC/AWG and disable tone mapping, then apply the Alexa 709 LUT.

 

ACES rrt/odt highlight rolloff

CST tone map

 

ACES linear -1 stop (for visualisation only)

P3D60 linear -1 stop (for visualisation only)

[Ryan Earl]

ACES is looking like the way to go, I’ve found that especially at IS0 800 the shadows are not so crunchy in some of my footage compared to the CST method.

[Llaasseerr]

The methods are quite similar, but ACES is probably more simple to manage if its used in its own project. If I was going to output to Alexa log, I would also start in ACES to get there. It does manage things like the gamut and the highlights more seamlessly.

I've been looking at the CIE chromaticity chart a lot more to check what the gamut is doing between operations. It's useful in unmanaged projects especially, to check there's no clipping to a smaller gamut like Rec709 before final output.

 

[hyalinejim]

8 hours ago, Ryan Earl said:

Sigma FP Color Mode "ON" Portrait, Standard, etc - UHD DNG - WB - Daylight

Sigma FP Color Mode "OFF" Portrait, Standard, etc - UHD DNG - WB - Daylight

I haven't been following this topic very closely and don't have an FP so disregard this if it's not relevant or has been covered already. I can tell you from much experience of looking at colour charts that the colours in "OFF" are far more accurate and true to life than in "ON". The blue patch is a giveaway here as it often is when a colour gamut or matrix is being interpreted incorrectly: in the "ON" version that patch is too purple.

[kye]

17 hours ago, Anaconda_ said:

'Don't use color wheel adjustments(Lift, Gamma, Gain and Offset) - so curves only? or it's ok to use the LOG wheels - seems like a weird thing to rule out.

Pretty sure that the Lift/Gamma/Gain controls are designed for use on 709 footage, and the Contrast/Pivot/Offset are designed for use on Log footage.  This might (I'm guessing here) have implications for how the out-of-bounds values are handled (eg, the curves might do predictable things and the LGG might not).

To make a more general comment in the context of the bewildering complexity of the various colour spaces and gammas that are being discussed in this thread, Resolve is now getting to be a very sophisticated engine when it is put into a "colour managed" mode (RCM or ACES) and in the most recent versions some of the controls are now "Colour Space Aware" so will act differently depending on what colour space you have told Resolve that the project is in.

To expand on this somewhat, you might have footage that's shot in one LOG format, and if you tell Resolve to manage the colour space in the same log format that matches the footage, and you adjust the Offset wheel +1 then you will get the same results as if you shot the footage exposing one stop brighter.  However, if you do the same but don't match the log format of the project to the footage the same +1 adjustment will do strange things because Resolve is doing complex things underneath the surface to tailor its behaviour to the specific colour space you've told it to work in.
I've confirmed this to be the case with the HLG files from my GH5, which (almost!) line up perfectly with rec2100.  In previous versions of Resolve without colour management (I never tried ACES) I shot two clips one stop different and tried to duplicate one shot by modifying the other shot, but I failed to find a combination of colour space and tool that would do the correction.  Now in Resolve 17 in Resolve Colour Managed mode it's dead easy to do it and the +1 control on the Offset wheel does an almost perfect job.  (On footage that is directly aligned to a colour space in Resolve, the transformation is perfect - HLG on the GH5 isn't).

I suspect that this means that some tools effectively have a CST prior to their adjustment and then one after it to put the footage back into the mode that your project is in.  This isn't the only place that Resolve does "hidden" CSTs in - you can program individual nodes to do that too I think.  

I had a long and rather frustrating conversation with a pro colourist when I asked about configuring projects to use RCM and they essentially advised me to do everything manually, despite using the RCM modes (of CSTs) being the proper way to do things.  I think the advice was well intended and was simply coming from a place of concern for the depth of complexity actually involved in the tool, and their (probably frequent) experience of non-post-pros going down the rabbit hole and essentially getting lost, so putting up a fence to protect people from the journey makes sense considering how many of them never make it to the other side.  (I was also trying to do something a bit horrible that complicated the workflow beyond how that mode is designed to be used, so my example was even worse than the standard situation).

To make an even broader comment about this whole thread I am reminded of a long conversation / debate about if there was any magic in the Alexa ARRIRAW files, and it was pretty clearly showed from some under/over tests that the RAW from the Alexa was a completely neutral linear capture - just like how sensors are designed.  This should be true of the RAW output from any sensor because it's literally how sensors work.  The challenge is how to process it in a way that gives you the results you want.  It's true there are potentially slight differences between the frequency response of various R G B filters on the sensor, but they're likely to be relatively similar and can easily be adjusted with an RGB matrix if you're really keen on matching cameras (that's the basis of the BM RAW conversions from Juan Melara IIRC).  Otherwise, you can simply take the RAW Linear files from any RAW camera and apply whatever conversions you like - if you like the ARRI colour science then apply their LUTs or transforms, etc.

It's also worth stating that many heavy colour grades will completely obliterate any sense of colour accuracy so I would imagine the pursuit of it is really only relevant in situations where a natural or hyper-natural (eg, commercial look where everything has to be cleaner than real life) look is required.