SIGMA FP with ProRes RAW and BRAW !

[OleB]

20 hours ago, Llaasseerr said:

The value proposition with the Ninja V is clearly compressed raw. It has a bit less highlight dynamic range because it doesn't have the highlight recovery of DNGs, but that is mostly fake information at best. It can work well, other times not. 

Not too sure about this one. Sensor clipping appears to be in 108 IRE in the Ninja. So possibly Davinci could be using the info of what is above 100 IRE and use this for their highlight recovery tool? Suppose in FCPX you can shift down this information back into the range manually and have the same result. 

[mercer]

With FCPX, has anyone tried using the "zebra" tool in the View drop down window to see if it corresponds with anything?

Also Color Finale has a raw importer tool called Transcoder which will read the dng files. I'd be curious to see how they compare with the ProRes Raw files.

And finally, regarding internal raw... is it possible that the settings from photo mode are effecting cine mode? For instance... if you have the photo setting at 4:3, perhaps it's effecting the meter readings of the cine mode 16:9 and if they're both set to 16:9 maybe it will click. I'd also check the Director's Viewfinder.

Anyway... just some thoughts. 

[mercer]

6 hours ago, Llaasseerr said:

That LUT is a default example of how well the rolloff can be utilised.

Re: ML Raw, yes it's really the same workflow as with any raw footage. At some point, the raw footage needs to be transformed to log to apply a film print style LUT. In ACES that transform is built into the reference rendering transform/output device transform. If doing it manually, it's the process of transforming to, for example, LogC then to Rec709. Those are the two basic, most common examples. 

 

With ML Raw, the basic idea was to use BM Film but there really isn't a designed post workflow.

I've used everything form the LogC transform in the MLV App, to BM Film in Resolve, to LogC CST and also used Rec709 in the Raw Panel. None of them seem any better than the other, really.

I think this is why I'm intrigued by the FP because it seems similar to ML Raw in some ways, and I've found that some times you just need to wing it and make it work.

[Llaasseerr]

5 hours ago, kye said:

I must admit that I haven't kept up with your discussions on this, but I got the impression that you can't use the False Colour mode on the FP to accurately monitor things across all the ISOs - is that correct?

The way I would use this camera would be either manually exposing or using it in auto-ISO and using exposure compensation, but I would be using the false-colours in either mode to tell me what was clipped and where the middle was.

I'd be happy to adjust levels shot-by-shot in post (unlike professional workflows when working with a team) and know how to do that in Resolve, so I'd be comfortable raising or lowering the exposure based on what was clipping and what I wanted to retain in post.  If the false-colour doesn't tell me those things then it would sort of defeat its entire purpose..

I personally think it's possible to use the v4 firmware false color feature on the camera, but I don't have it yet to verify that. I've just been playing around with test footage.

[Llaasseerr]

1 hour ago, mercer said:

With ML Raw, the basic idea was to use BM Film but there really isn't a designed post workflow.

I've used everything form the LogC transform in the MLV App, to BM Film in Resolve, to LogC CST and also used Rec709 in the Raw Panel. None of them seem any better than the other, really.

I think this is why I'm intrigued by the FP because it seems similar to ML Raw in some ways, and I've found that some times you just need to wing it and make it work.

I had a look at the MLV App a few months ago and I was quite impressed. From memory, I think exporting as ProRes LogC worked out okay and was predictable. 

I'm personally not into winging it because I'm used to a color managed workflow. I've been able to get fairly predictable results with the DNGs from the fp, which is good enough for me. I know I could always fall back on a light meter, but I'm pretty sure some people got the new false colors working too so I'm interested to give that a try if I get the camera.

[Llaasseerr]

1 hour ago, OleB said:

Not too sure about this one. Sensor clipping appears to be in 108 IRE in the Ninja. So possibly Davinci could be using the info of what is above 100 IRE and use this for their highlight recovery tool? Suppose in FCPX you can shift down this information back into the range manually and have the same result. 

I would not use IRE unless I had tight control of the color imaging pipeline, then I might use it in a pinch to check things on a log image since it can only show a 0-1 display referred space.

It seems for now we established we have no idea what the Ninja V is doing as far as display, so to me IRE and really false color too (on the Ninja) is useless until that is figured out.

When viewing in a linear space compositing tool (Nuke) that requires zero image manipulation to see the entire range of the image, it's very easy to check the max linear value of the pixel, and the DNGs with highlight reconstruction have a higher value than the PRR images by about 3/4 of a stop.

In terms of linear floating point values where middle grey is 0.18, I'm reading about 16.5 max RGB clipping value in DNGs and about 9.35 in the ProRes Raw files.

 

 

[Llaasseerr]

3 hours ago, OleB said:

Thank you. Highly appreciated! Will take a look on this later. 
 

Maybe the fault is the not so great color management in FCPX. Possibly it would be way easier with Davinci Resolve taking BRAW as base. Do not have such a recording device at hand for testing though. 

The native mode is quite similar to something I would expect the V-log to look like. However both black and white points are weird. Black at about 10 IRE and white about 65 IRE. Should they not be at the full scale, so 0-100? The middle grey though is at its correct position for the respective ISO setting as far as I can tell. 

It is really hard to say, but the display might be V-log since 0% black is at IRE 7.3 and 90% white is at 61 IRE. The rest of the range above that is highlight information, probably up to 85-90 IRE in V-log depending on the sensor.

Middle grey is at 42 IRE which is convenient because this is pretty much the same as what you would expect with a Rec709 LUT.

So it's possible Atomos just made the decision to use V-log to display Sigma raw since there's no Sigma log, which would mean that the V-log metadata tag it adds is not superfluous.

One way to check would be to add the standard V-log to Rec709 LUT on the Atomos and see if it matches V-log to Rec709 in FCPX, assuming that you first interpret the linear raw files as V-log as well.

https://pro-av.panasonic.net/en/cinema_camera_varicam_eva/support/pdf/VARICAM_V-Log_V-Gamut.pdf

[OleB]

8 minutes ago, Llaasseerr said:

One way to check would be to add the standard V-log to Rec709 LUT on the Atomos and see if it matches V-log to Rec709 in FCPX, assuming that you first interpret the linear raw files as V-log as well.

Will check that once I am back from the office. If this is turns out to be true, the solution might be as simple as adding that V-log to Rec709 LUT on the Ninja V…

[OleB]

8 hours ago, Llaasseerr said:

And it's important to know that it's not the Arri Rec709 LUT in itself that is the Mojo.

Have read that big part of the Arri look is that they let color saturate only until middle grey luma. After that saturation stays the same only brightness increases. Does the LUT do the same? Or is there a post production way (curves) to do the same?

[Llaasseerr]

14 minutes ago, OleB said:

Have read that big part of the Arri look is that they let color saturate only until middle grey luma. After that saturation stays the same only brightness increases. Does the LUT do the same? Or is there a post production way (curves) to do the same?

Naively I can say that there is less saturation in highlights in film, so that sort of makes sense. To be honest, I don't think there's anything really magical about the LUT but the classic K1S1 was enduringly popular as a SOOC look.

I'm guessing maybe you could experiment with Sat/Luna curves in Resolve?

[OleB]

We are getting really close. The LUT seems to do the trick. There is a minimal difference in highlight brightness (monitor shows 78 IRE, FCPX shows 75 IRE) when imported into FCPX and applying the same LUT as in the monitor. Suppose this is due to the not 100% matching RAW to V-log conversion which is built into FCPX.

ISO values and middle grey seem to work as intended now. Interestingly the brightest part in the picture is only reaching red in the false colors when using ISO 400 and upwards. Seems the maximum steps from middle grey to white are increasing (that reacts as per Sigma ISO manual).

Will investigate further now.

[Ryan Earl]

2 hours ago, OleB said:

Will investigate further now.

Almost want to get a Ninja V just to see how crazy this is!

[Llaasseerr]

3 hours ago, OleB said:

We are getting really close. The LUT seems to do the trick. There is a minimal difference in highlight brightness (monitor shows 78 IRE, FCPX shows 75 IRE) when imported into FCPX and applying the same LUT as in the monitor. Suppose this is due to the not 100% matching RAW to V-log conversion which is built into FCPX.

ISO values and middle grey seem to work as intended now. Interestingly the brightest part in the picture is only reaching red in the false colors when using ISO 400 and upwards. Seems the maximum steps from middle grey to white are increasing (that reacts as per Sigma ISO manual).

Will investigate further now.

Does the linear to V-log transform (without the log to 709 LUT) in FCPX match the default image on the Ninja V? That would be another way of checking this.

Could the slight mismatch be a video vs data levels display issue?

[Llaasseerr]

Was just checking out the new Fuji X-H2S and its new F-log2 curve that has a lot more range than the original F-log.

The fact you can record ProRes raw open gate on the APS-C, and I would say the Atomos will apply the new F-log2 curve for monitoring, this is awesome and exactly what Sigma needs to do. Simple.

Well, if we figure out that V-log matches between the Ninja and NLE then that will actually be effective enough.

[OleB]

31 minutes ago, Llaasseerr said:

Does the linear to V-log transform (without the log to 709 LUT) in FCPX match the default image on the Ninja V? That would be another way of checking this.

Could the slight mismatch be a video vs data levels display issue?

Will check that tomorrow in detail. But at first glance I would say they are extremely similar (V-Log transform and native monitoring). 

What I don't get is that with ISO 100 & 200 the false colors with LUT applied are not going to full 100 IRE, they stay well below. I was under the impression that the LUT should stretch that out. 

Middle grey is shifting correctly with the ISO steps and display is as well doing so.

All in all at least another really good approach which might turn out to be the best choice instead of my cumbersome curve editing.

 

[Llaasseerr]

29 minutes ago, OleB said:

Will check that tomorrow in detail. But at first glance I would say they are extremely similar (V-Log transform and native monitoring). 

What I don't get is that with ISO 100 & 200 the false colors with LUT applied are not going to full 100 IRE, they stay well below. I was under the impression that the LUT should stretch that out. 

Middle grey is shifting correctly with the ISO steps and display is as well doing so.

All in all at least another really good approach which might turn out to be the best choice instead of my cumbersome curve editing.

 

I'm a bit confused because I thought you said the Ninja recording starts at 800.

What about on a waveform instead of false colors? What is the max IRE value according to that? I assume you mean you're just applying the standard V-log to Rec709 LUT on the naked image on the Ninja.

There is a chance that this doesn't work and you're actually stuck with some random flat image from the camera.

[OleB]

1 hour ago, Llaasseerr said:

I'm a bit confused because I thought you said the Ninja recording starts at 800.

What about on a waveform instead of false colors? What is the max IRE value according to that? I assume you mean you're just applying the standard V-log to Rec709 LUT on the naked image on the Ninja.

There is a chance that this doesn't work and you're actually stuck with some random flat image from the camera.

For Rec709 and PQ it is true that ISO 800 = ISO 100 in terms of sensor clipping and the metering starts to change only from ISO 800 and upwards.

Native mode and now with the standard V-Log to Rec709 LUT the Ninja measures as from ISO 100 (both waveform and false colors change). However I do not know if this really showing something in regards to real sensor saturation or if it is just showing applied gain. Actually I think it is the later because of the Sigma ISO guide. ISO 100-800 is the same but with different distribution and gain levels.

Will take the measures tomorrow, at least for some key ISO values like 100, 800, 3200. Update to follow.

 

[OleB]

So here we go, took the measurements and uploaded the files for your own tests.

https://www.dropbox.com/sh/cgm80q9w682oc19/AADWHieOS-Wka9JuD-GnXerca?dl=0

Have used the LUT which is also included in the dropbox for monitoring and as LOG to Rec709 conversion in the FCPX info tab. RAW to LOG was the build in Panasonic V-log. You cannot load custom conversions into that tab by the way.

Good things first, the light meter is matching for all ISO settings with the middle grey representation on the false colors screen of the Ninja V. Visually the preview picture is also a pretty good match. So that part ist out of the way.

ISO 100 representation on the Ninja V:

How the greyscale looks after import and conversion settings applied in FCPX:

That seems to be really close, with a little bit of less highlight brightness. Middle grey is still where it belongs though. 

All ISO settings will result into the exact same picture of the luma values. Also positive.

Now we are coming to the clipping.

 

ISO 100 clipping point as per waveform in the monitor:

After import:

That is a little bit weird already. Would have expected that the LUT is extending the values further. So basically clipping will be at over 100 IRE in FCPX, but that is only the case roughly for ISO 800.

Waveform ISO 800:

Waveform ISO 3200:

 

 

Looking forward to your own investigation results. For my part I will try to create some more footage to see how this works with real pictures. The colors need work as from what I see, too little saturation for my taste.

[webrunner5]

Speaking of False Color and such check out this article and go about halfway down and see how different each brand handles color in false color. Might be why the FP is so wonky.

https://www.videomaker.com/how-to/shooting/how-to-use-false-color-in-your-next-project/?utm_campaign=VM eNews&utm_medium=email&utm_content=214916445&utm_source=hs_email

[kye]

7 hours ago, webrunner5 said:

Speaking of False Color and such check out this article and go about halfway down and see how different each brand handles color in false color. Might be why the FP is so wonky.

https://www.videomaker.com/how-to/shooting/how-to-use-false-color-in-your-next-project/?utm_campaign=VM eNews&utm_medium=email&utm_content=214916445&utm_source=hs_email

Are you talking about this one?

I figured that I'd have to learn the false colour standard at some point, but seeing that and realising there is no standard, and that some of these are really very unintuitive, maybe I'd just make my own false-colour LUT and get what I need from that!