SIGMA FP with ProRes RAW and BRAW !

[Llaasseerr]

Just a note IMO about linear Raw vs Log encoding. Let's keep in mind that log encoding was developed by Kodak for Cineon scanning of film negative, and is basically what was passed on to digital cameras. It's not an inferior image encoding method developed for hybrid 4k mirrorless cameras. The LogC encoding on the Alexa is probably the most prominent log curve for a digital camera to be very close to Cineon, and Red and Sony eventually capitulated to that.

It's not inferior to linear raw, unless the individual log curve implementation by that camera manufacturer is left wanting. It's a more efficient encoding of highlights and shadows because it gives less code values to highlights, rather than half of a linear image for the brightest stop. It also allocates more code values to the shadows than linear raw. I would rather take 12 bit log ProRes RAW from the Sony cams/Ninja V than 12 bit linear DNGs, but obviously that's a much more expensive solution than the fp - like 5x the price.

The reason that log footage is often inferior in the prosumer cams is because it's probably chroma subsampled, denoised and DCT compressed. ProRes 4444 log from the Alexa Classic is a thing of wonder though.

If you apply the inverse transform of the log encoding curve, then your image is back in linear space and if you factor out the ways the image has been decimated, then it is very close to an original raw recording.

 

[Ryan Earl]

8 hours ago, Llaasseerr said:

Yes the most desirable scenario is to use this in as minimalist a fashion as possible like a Leica that can capture Vistavision, exposing for the middle.

I used a Leica CL with a 40mm for a long time for stills and the Sigma FP comes really close to that idea of minimalism. 

9 hours ago, HockeyFan12 said:

The S1 in the field is convenient but I don't love the image as much nor the post workflow.

 

I think the FP is very convenient.  And I haven't actually come across any aliasing in practice with the 4K DNG vs the URSA Mini 4.6K where I have it in the test scene (showing on the URSA on the color checker) when I compared the two cameras.  The FP is very clean and vibrant.

I bought SmallRig accessories, cage, rail adapter, and ssd holder and a top handle when I need to adapt the Sigma FP into a larger setup, but also just using it bare bones with the EF adapter and an Canon IS lens has really been smooth.  

I've remapped the False Color to the 'AEL' button on the back as well as the 'Mode' button to ISO and 'Tone' to Peaking.  Then with the aperture / shutter dials enabled in cinema I can bump around exposure.  I have a few Hoya Pro NDs and a magnetic adapter ring.  I have a Canon 5DS for stills and while the colors are slightly different out of the box I don't have any issues trying to match it if that's what's needed.  

And the post workflow is honestly very quick if you drop into a DaVinci Color Managed or ACES project and not any different really than taking the stills to Lightroom / Photoshop Camera Raw or Capture One. 

I don't expect it to be an Alexa, but I think there is plenty of runway in the files for any kind of imagery you want to make with it.

[TomTheDP]

11 hours ago, HockeyFan12 said:

I guess to me the S1 is the ultimate in on set convenience and the Alexa is the ultimate in convenience in post. 

Increasingly I feel like I should have gone for the middle ground for both (P4K and speed booster XL) and left it at that but the last thing I need is to invest more money and time in another camera system.

Were I starting over I think I'd go for a P4K, speed booster XL, and a LOT of old Nikkors.

But I'm not sure I'm starting over. So I don't really know.

Yeah I like the color coming out of the FP better and CDNG just looks more organic than anything I have seen from the S1. That said otherwise they are very similar cameras. Possibly the same sensor? You get more resolution and frame rate options on the S1.

I agree the S1 is super convenient. pretty nice button layout, IBIS opens up a lot of new shooting styles, solid full hdmi, and small but pretty high quality codecs.

The thing the S1 is still lacking like every other "cheap" camera is poor focus peaking and HDMI lag. It's hard to pull critical focus as the HDMI out introduces noticeable lag. Not a problem for the most part but if I am on set with an AC, its not a great situation. Plus these smaller cameras are often the "gimbal" or odd rig cam where you actually really need remote focus. I don't know if you've ever used ARRI focus peaking or RED but it's night and day better than on these cheap cameras. Another issue, though this is less critical I suppose, is proper REC709 monitoring. On Alexa or RED what you see on the LCD is what you are getting. Accurate color and noise. Monitoring on the S1 or any cheap camera I have seen is not accurate.

In terms of usability the FP has to be rigged up for handheld shooting for sure. The micro HDMI is very concerning but I'll do my best to keep it intact.

The Pocket 4k is hard to beat. It is so cheap to start with. Blackmagic menus are the best in the game. The non tilt screen is annoying as well as the relatively short battery life. Overall though I have never loved the BM image, which is why I've not bought any of their newer cameras. Overall though the P4K is a great value, 1k for something that is close to full frame with a speedbooster and has internal compressed "raw" up to 4k 60p. It does suffer from the same hdmi lag and shit focus peaking. It does have full hdmi which is a blessing.

10 hours ago, Llaasseerr said:

Just a note IMO about linear Raw vs Log encoding. Let's keep in mind that log encoding was developed by Kodak for Cineon scanning of film negative, and is basically what was passed on to digital cameras. It's not an inferior image encoding method developed for hybrid 4k mirrorless cameras. The LogC encoding on the Alexa is probably the most prominent log curve for a digital camera to be very close to Cineon, and Red and Sony eventually capitulated to that.

It's not inferior to linear raw, unless the individual log curve implementation by that camera manufacturer is left wanting. It's a more efficient encoding of highlights and shadows because it gives less code values to highlights, rather than half of a linear image for the brightest stop. It also allocates more code values to the shadows than linear raw. I would rather take 12 bit log ProRes RAW from the Sony cams/Ninja V than 12 bit linear DNGs, but obviously that's a much more expensive solution than the fp - like 5x the price.

The reason that log footage is often inferior in the prosumer cams is because it's probably chroma subsampled, denoised and DCT compressed. ProRes 4444 log from the Alexa Classic is a thing of wonder though.

If you apply the inverse transform of the log encoding curve, then your image is back in linear space and if you factor out the ways the image has been decimated, then it is very close to an original raw recording.

 

Yeah at the end of the day it's recording 12 bit linear RAW, when the sensor is capable of 14. So data is being thrown away. I am personally not big on the Sony look which is why I wouldn't go the A7S3 route. I honestly feel the FP image is better despite capturing less color information. The thing about the Alexa is even if you just record in 422 it looks fantastic. I feel the color you can quickly get off the Sigma is really really nice. It's almost making me want to sell my Alexa, but the FP is just not suited for A cam usage on a "real" set.

 

[Llaasseerr]

2 hours ago, Ryan Earl said:

I used a Leica CL with a 40mm for a long time for stills and the Sigma FP comes really close to that idea of minimalism. 

I think the FP is very convenient.  And I haven't actually come across any aliasing in practice with the 4K DNG vs the URSA Mini 4.6K where I have it in the test scene (showing on the URSA on the color checker) when I compared the two cameras.  The FP is very clean and vibrant.

I bought SmallRig accessories, cage, rail adapter, and ssd holder and a top handle when I need to adapt the Sigma FP into a larger setup, but also just using it bare bones with the EF adapter and an Canon IS lens has really been smooth.  

I've remapped the False Color to the 'AEL' button on the back as well as the 'Mode' button to ISO and 'Tone' to Peaking.  Then with the aperture / shutter dials enabled in cinema I can bump around exposure.  I have a few Hoya Pro NDs and a magnetic adapter ring.  I have a Canon 5DS for stills and while the colors are slightly different out of the box I don't have any issues trying to match it if that's what's needed.  

And the post workflow is honestly very quick if you drop into a DaVinci Color Managed or ACES project and not any different really than taking the stills to Lightroom / Photoshop Camera Raw or Capture One. 

I don't expect it to be an Alexa, but I think there is plenty of runway in the files for any kind of imagery you want to make with it.

What do you think about the idea of adding an after-market OLPF? Or is it best just to blur the image a tad 🙃. Not that there is presently an OLFP for this camera that I know of.

I ask because I know it can experience moire in some cases, and Sigma decided not to add an OLPF to retain sharpness. But tbh 2k resolving power is enough for me if the pixels are good pixels.

[Llaasseerr]

16 minutes ago, TomTheDP said:

 

Yeah at the end of the day it's recording 12 bit linear RAW, when the sensor is capable of 14. So data is being thrown away. I am personally not big on the Sony look which is why I wouldn't go the A7S3 route. I honestly feel the FP image is better despite capturing less color information. The thing about the Alexa is even if you just record in 422 it looks fantastic. I feel the color you can quickly get off the Sigma is really really nice. It's almost making me want to sell my Alexa, but the FP is just not suited for A cam usage on a "real" set.

 

I know people take issue with a Sony "look", but I've found a very neutral image when imported to ACES just with the default S-log3/Sgamut3 input transform, displayed through the standard rrt/odt. I think a lot of people just don't do that one basic thing and then have issue with skin tones and weird tints.

I had been planning to buy an FX6, but they're really hard to come by due to the supply chain issues. But a friend has one and it's been interesting to do a ton of tests with it. In my mind, when shooting raw it's close to an Alexa, except as I stated the lower DR.

[TomTheDP]

9 minutes ago, Llaasseerr said:

I know people take issue with a Sony "look", but I've found a very neutral image when imported to ACES just with the default S-log3/Sgamut3 input transform, displayed through the standard rrt/odt. I think a lot of people just don't do that one basic thing and then have issue with skin tones and weird tints.

I had been planning to buy an FX6, but they're really hard to come by due to the supply chain issues. But a friend has one and it's been interesting to do a ton of tests with it. In my mind, when shooting raw it's close to an Alexa, except as I stated the lower DR.

Has he tested it directly against an Alexa?

Cameras are definitely getting better and better.

[Llaasseerr]

17 minutes ago, TomTheDP said:

Has he tested it directly against an Alexa?

Cameras are definitely getting better and better.

No he hasn't, but I've worked on a lot of movies shot on Alexa where generally the footage is ingested to EXR as either linear gamma/Alexa Wide Gamut, or as ACES. I have about 10 TB of footage shot on Alexa with extreme dynamic range in the scene, so I got a good sense of how it performs. We are really pixel peeping the plates and tearing them apart, so we know what's what.

[TomTheDP]

14 minutes ago, Llaasseerr said:

No he hasn't, but I've worked on a lot of movies shot on Alexa where generally the footage is ingested to EXR as either linear gamma/Alexa Wide Gamut, or as ACES. I have about 10 TB of footage shot on Alexa with extreme dynamic range in the scene, so I got a good sense of how it performs. We are really pixel peeping the plates and tearing them apart, so we know what's what.

The rendering on newer Sony cameras is nice even just shooting 10 bit with their 709 luts. The problem I have encountered not particular to Sony is just more subtle or small color differences that are hard to correct for. Would definitely be interesting to see the Sony in mixed lighting against the Alexa. Maybe just shooting in RAW and going through ACES gives a good match.

[Ryan Earl]

3 hours ago, Llaasseerr said:

What do you think about the idea of adding an after-market OLPF? Or is it best just to blur the image a tad 🙃. Not that there is presently an OLFP for this camera that I know of.

I ask because I know it can experience moire in some cases, and Sigma decided not to add an OLPF to retain sharpness. But tbh 2k resolving power is enough for me if the pixels are good pixels.

The HD capture in camera in 12 BIT DNG is where I've seen moire and aliasing in more normal shooting conditions, like hair and details like eyebrows.  Though the rolling shutter is much better.  It's supposed to be half of UHD I think, about 10 ms?

With the UHD, I've haven't seen it where it ruins a shot, but have had it ruin shots on the Blackmagic cameras, so I have bought the OLPF and IR Cut where I could.  Like a fashion / clothing shoot on the URSA 12K when I was initially trying that camera on its v1 firmware.  Huge parts of a jacket were unusable, so I actually cropped in around it to a head shot later.  The RAWlite helps a lot adding extra IR Cut too.  The original URSA 4.6K that I still have is without one, there I've used a Schneider 1/4 Classic Soft or softer lenses and a Hoya IR Cut. In ProRes 4444 vs DNG the artifacts are blurred enough so I don't worry about it.

With the Sigma FP I've been surprised I haven't seen it in normal shooting, and it's much cleaner in terms of IR.  It doesn't need the extra IR cut.  Though it can have a digital sharpness where I will still cut it with soft filter.  In a test shot, with a lot of detail and packaging with tiny print the Sigma FP does better than the URSA 4.6K.  The URSA 4.6K will have more artifacts spread through the objects and print details where it isn't there on the Sigma FP.  Though in the color checker and charts it is there in ultra fine details but looks better again than the URSA 4.6K. 

So I'm not personally wishing I had an OLPF if it existed or thinking it necessarily should have been added. 

 

[Ryan Earl]

16 hours ago, TomTheDP said:

By the way I tested the ARRI against the Sigma in a mixed lighting condition. Lit myself with a blue light on one side and and orange light over head. I was really pleased how close it looked to the Alexa after adjusting WB, tint, and contrast.

Its really too bad all the cheaper cameras have crap focus peaking and terrible laggy HDMI out. The image quality is there but the practical usability on set isn't. But hey still pretty cool to have something like this for a little over a thousand bucks.

Is the comparison with the Alexa in 2K or 4K? 

[TomTheDP]

35 minutes ago, Ryan Earl said:

Is the comparison with the Alexa in 2K or 4K? 

I compared it with my Classic shooting prores 2k.

[kye]

18 hours ago, Llaasseerr said:

Just a note IMO about linear Raw vs Log encoding. Let's keep in mind that log encoding was developed by Kodak for Cineon scanning of film negative, and is basically what was passed on to digital cameras. It's not an inferior image encoding method developed for hybrid 4k mirrorless cameras. The LogC encoding on the Alexa is probably the most prominent log curve for a digital camera to be very close to Cineon, and Red and Sony eventually capitulated to that.

It's not inferior to linear raw, unless the individual log curve implementation by that camera manufacturer is left wanting. It's a more efficient encoding of highlights and shadows because it gives less code values to highlights, rather than half of a linear image for the brightest stop. It also allocates more code values to the shadows than linear raw. I would rather take 12 bit log ProRes RAW from the Sony cams/Ninja V than 12 bit linear DNGs, but obviously that's a much more expensive solution than the fp - like 5x the price.

The reason that log footage is often inferior in the prosumer cams is because it's probably chroma subsampled, denoised and DCT compressed. ProRes 4444 log from the Alexa Classic is a thing of wonder though.

If you apply the inverse transform of the log encoding curve, then your image is back in linear space and if you factor out the ways the image has been decimated, then it is very close to an original raw recording.

Good post.

To clarify I was comparing RAW Linear with compressed Log in the context of ETTR vs exposing for middle grey.  ETTR is a great strategy if you're using Linear RAW and don't mind compensating for the level changes in post.

The Alexa is a bit special in this way (I believe) as it can output Log in very high-bitrate 12-bit 444, compared to (say) the GH5 200-400Mbps 10-bit 422 log, so the Alexa has an enormous amount more latitude just because of the codec.  It's also special in the sense the readout from the sensor is a higher bit depth than the Log codec.  On a camera with a 12-bit sensor output, if that was transformed into a 12-bit Log profile then the highlights would lose the effective bit-depth they had when in Linear, but the other stops wouldn't really benefit much because although the 12-bit log has better bit-depth for the lower stops, the limitation would always be the bit-depth from the Linear readout.

Regardless, having a 12-bit sensor readout converted to a 10-bit log profile in-camera is probably good enough for almost all applications - I've tried to break the 10-bit HLG on the GH5 by applying curves that were severe beyond reason and the image held up.  

Of course, the thing missing from most consumer cameras is a decent codec, like Prores HQ or 4444.  I'm really looking forward to the GH6 update to see how the 1080p Prores looks.  Unfortunately most cameras have really weak internal h264/5 implementations and the in-camera downscaling is often very poor.  I suspect the push towards RAW in this segment is partly a way to deliver better image quality without having to design cameras that can scale and compress properly or having to buy a Prores license.  Unfortunately it then means they run into the challenges around internal compressed RAW and so we end up with tiny cameras needing an SSD or external recorder strapped to them.  Hardly ideal.

[TomTheDP]

15 hours ago, kye said:

Regardless, having a 12-bit sensor readout converted to a 10-bit log profile in-camera is probably good enough for almost all applications - I've tried to break the 10-bit HLG on the GH5 by applying curves that were severe beyond reason and the image held up.  

Of course, the thing missing from most consumer cameras is a decent codec, like Prores HQ or 4444.  I'm really looking forward to the GH6 update to see how the 1080p Prores looks.  Unfortunately most cameras have really weak internal h264/5 implementations and the in-camera downscaling is often very poor.  I suspect the push towards RAW in this segment is partly a way to deliver better image quality without having to design cameras that can scale and compress properly or having to buy a Prores license.  Unfortunately it then means they run into the challenges around internal compressed RAW and so we end up with tiny cameras needing an SSD or external recorder strapped to them.  Hardly ideal.

I've had the Emotive Color lut break the Panasonic s1 4k 10 bit files. What happens is you get like weird looking noise. Hard to explain I'd have to post some photos. I suppose neat video would probably fix the issue. I would also imagine if you shoot RAW this probably wouldn't happen but that is a guess.

[HockeyFan12]

44 minutes ago, TomTheDP said:

I've had the Emotive Color lut break the Panasonic s1 4k 10 bit files. What happens is you get like weird looking noise. Hard to explain I'd have to post some photos. I suppose neat video would probably fix the issue. I would also imagine if you shoot RAW this probably wouldn't happen but that is a guess.

In Adobe Suite? Adobe's HEVC implementation seems to be broken and there can be banding but I find the 6K HEVC files super robust if I transcode them in OSX (or Resolve should work) instead. Not sure about 4K.

How is the fp's 4K center crop mode btw? Feels weird putting a speed booster on a FF camera but that could fix the aliasing issues I've seen (which, to be fair, are limited to fabrics that might cause even the Alexa to have aliasing, except on the fp it can be crazy bad).

 

[TomTheDP]

17 minutes ago, HockeyFan12 said:

In Adobe Suite? Adobe's HEVC implementation seems to be broken and there can be banding but I find the 6K HEVC files super robust if I transcode them in OSX (or Resolve should work) instead. Not sure about 4K.

How is the fp's 4K center crop mode btw? Feels weird putting a speed booster on a FF camera but that could fix the aliasing issues I've seen (which, to be fair, are limited to fabrics that might cause even the Alexa to have aliasing, except on the fp it can be crazy bad).

 

I have yet to test it out. I imagine the rolling shutter is less in crop mode, so that could actually be a good idea.

I always shot in regular VLOG 422 H254, not HEVC. Maybe it would work better in Resolve though.

[Llaasseerr]

On 5/10/2022 at 2:47 PM, Ryan Earl said:

The HD capture in camera in 12 BIT DNG is where I've seen moire and aliasing in more normal shooting conditions, like hair and details like eyebrows.  Though the rolling shutter is much better.  It's supposed to be half of UHD I think, about 10 ms?

With the UHD, I've haven't seen it where it ruins a shot, but have had it ruin shots on the Blackmagic cameras, so I have bought the OLPF and IR Cut where I could.  Like a fashion / clothing shoot on the URSA 12K when I was initially trying that camera on its v1 firmware.  Huge parts of a jacket were unusable, so I actually cropped in around it to a head shot later.  The RAWlite helps a lot adding extra IR Cut too.  The original URSA 4.6K that I still have is without one, there I've used a Schneider 1/4 Classic Soft or softer lenses and a Hoya IR Cut. In ProRes 4444 vs DNG the artifacts are blurred enough so I don't worry about it.

With the Sigma FP I've been surprised I haven't seen it in normal shooting, and it's much cleaner in terms of IR.  It doesn't need the extra IR cut.  Though it can have a digital sharpness where I will still cut it with soft filter.  In a test shot, with a lot of detail and packaging with tiny print the Sigma FP does better than the URSA 4.6K.  The URSA 4.6K will have more artifacts spread through the objects and print details where it isn't there on the Sigma FP.  Though in the color checker and charts it is there in ultra fine details but looks better again than the URSA 4.6K. 

So I'm not personally wishing I had an OLPF if it existed or thinking it necessarily should have been added. 

 

Thanks for the detailed answer!

I've been testing a lot more raw files in the last day that I found linked off youtube recently (see linked video).

I mentioned that I would lean towards re-pegging the DR by at least +3 stops since the camera holds shadow detail very well. Assuming base ISO 100, it seems you can cleanly push +4 stops with EI 100-800 in daylight and then maybe +3 with ISO 3200 in a darker scene if managed. Not sure about ISO 1600, it apparently has a bit less range that 800 and 3200? 

My latest question is around bad posterizing in the shadows with a green tint. I'm attaching a crop of a +4 stop push in post on one of the test clips from the upload I linked to.

I could be imagining this, but I think I've read here on eoshd before that people are seeing a green issue in shadows sometimes. I don't know if this is the issue here, though, or if it's to do with the fact that these files were passed through slimRAW with lossy DNG compression, where generally lossy DCT compression can unduly affect shadows because there's an assumption that it won't be seen.

What I'm hoping is that with lossless or losslessly compressed DNGs, this would not be an issue. But additionally, I've noticed that very high contrast lenses (Sigma) and night time scenes make the blacks fall to zero. This isn't necessarily an issue, but I would rather have them just off the zero floor. Not just because any compression algorithm can cause problems all the way down there, but also that if shooting to push 2-4 stops in post then too much of the image is clipped on the zero end and it looks excessively contrasty.

So it seems a good solution regardless of any compression issues is to shoot with something like a Tiffen Ultra Con 1 and/or a lower contrast vintage lens, assuming you can't just add a bit of bounce fill in a controlled lighting situation. Obviously this appeals for more of a run and gun setup.

If the green/banding issue is because of the sensor, that obviously sucks!

 

 

[Llaasseerr]

Just following up on that last post, I realise I was simultaneously saying the shadows hold up really well, then saying that they had banding and weird tinting issues.

What I should have said is that generally the shadow detail seems very low noise and very good, but I did run into the artifact issue in the case of this hair crop in the post above. So hopefully that kind of thing is not native to the sensor.

[Llaasseerr]

Alternatively it could just be that I'm taking the bits buried right at the left side of the histogram, and expecting there to be enough detail when expanded out by +4 stops into the midrange. Perhaps 12 linear bit just isn't enough for this kind of image manipulation, hence banding and posterization!

Maybe these numbers will reveal something:

Assuming 4096 code values in a 12 bit integer linear raw image converted to linear floating point EXR where middle grey is pegged at 0.18:

So if I underexpose by an extreme -4 stops, then that pushes middle grey down to .0112. With 4096 code values, middle grey at "correct" exposure would be at around 737. At -4 stops below, it would be at 46.

So that's probably my answer - LOL. 46 code values for everything from middle grey and below, and 4050 for everything above middle grey!

Assuming my maths is right. If so, -3 stops would be about 92 code values for middle grey in linear light encoding.

 

 

[TomTheDP]

Shooting at 100 iso I have found there to be an incredible amount of detail in the shadows. I did find banding and posterization to show up in 10 bit and 8 bit recordings.

The shadow control in the RAW tab on davinci can bring up banding posterization sometimes so I found it's best to avoid that. I found using the exposure control in the raw tab to make major adjustments and then using curves to make minor adjustments works well.

In terms of resolution and crop modes I found full frame 4k is noticeably cleaner than all the other modes. 4k crop mode, HD full frame and HD cropped all had more noise. HD is very noticeably softer. It isn't unusable but it's definitely not the golden standard for HD.

[OleB]

Took some time today to create a few clips to test noise of the different ISO settings. Found me a clear light bulb...@Llaasseerr so you can have a look in regards to the highlight clipping. 

You can download the files here: https://www.dropbox.com/sh/cgm80q9w682oc19/AADWHieOS-Wka9JuD-GnXerca?dl=0 

All ProRes RAW. Starting with ISO 100 up to ISO 12800 (keep in mind the metadata will tell you it is ISO 800 upwards)

Have fun playing around 🙂