Jump to content

Sigma Fp review and interview / Cinema DNG RAW


Recommended Posts

  • Replies 1.2k
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Popular Posts

So i said i'd post some stills, these are basically ungraded. This frame is in a sequence with car lights, i like the tonality of this very subdued moment. Shot 12bit to manage shadow tonality.

Modified Sigma's smaller grip to allow it to hold a T5; haven't printed it yet so not so sure how comfortable it is.

Ah gotcha. I'm not really directing this at you but just to anyone in general who isn't aware or wants to learn more, and I'm probably being a 'stickler for accuracy' here but obviously a sensor/

Posted Images

How effective would 16mm and S16 be on the fp ? Which mode would be best to shoot in with S16 lenses? I thought I read a post somewhere that said this camera had an S16 mode......most likely meaning the crop suited as an S16 fov. Not sure but it would be cool if we could mount S16 PL mount lenses on it

 

Link to post
Share on other sites
4 hours ago, JimJones said:

How effective would 16mm and S16 be on the fp ? Which mode would be best to shoot in with S16 lenses? I thought I read a post somewhere that said this camera had an S16 mode......

You must have mixed that up with the Blackmagic 4K, which has an S16 mode (since the last firmware update). If you were to adapt an S16 lens to the Sigma fp, your maximum usable resolution would be 1760x990, and you'd have a strong vignette/crop on the camera display.

Link to post
Share on other sites

OK thanks. I must be confused then. Done so much reading these last few days on the fp that my head is spinning. I am sure someone somewhere said they could use S16 on the fp but as you say it doesn't quite work out, so........

How would Mamiya 645 lenses converted to hard PL mounts go on the fp? What sort of crop factor would it be? 645 is twice the size of the full frame fp or close enough anyway.

Link to post
Share on other sites
23 hours ago, paulinventome said:

The knock on effect is that, like the shadows which go green, the highlights go magenta. As @Lars Steenhoff mentions i do wonder if this is a bug/issue or just a side effect of the lack of tonality.

If you recall the linearisation mapping that you posted before, there is a steep slope upwards at the end. Highlights reconstruction happens after linearisation, so it would have the clipped red channel curving up stronger than the non-clipped green and blue; it needs to preserve the trend. This hypothesis should be easy to test with a green or blue biased light. If the non-linear curve causes this, you will get respectively green or blue biased reconstructed highlights.

I don't think the post raised shadows tint is related to incorrect black levels. It is more likely a result of limited tonality, although it might be exaggerated a little bit by value truncation (instead of rounding). This can also be checked by underexposing the 12 bit image additional 2 stops compared to the underexposed 10 bit image, and then comparing shadow tints after exposure correction in post.

Link to post
Share on other sites
3 hours ago, cpc said:

I don't think the post raised shadows tint is related to incorrect black levels. It is more likely a result of limited tonality, although it might be exaggerated a little bit by value truncation (instead of rounding). This can also be checked by underexposing the 12 bit image additional 2 stops compared to the underexposed 10 bit image, and then comparing shadow tints after exposure correction in post.

I've done a bunch more digging. I think you're right however i have errors in highlights now with 12 and 10 bit - compared to 14 bit stills and also magenta single pixels. However i am trying to work out whether this is data or resolve. I have a feeling it might be resolve.

I am perhaps going to revise my comments about 14 vs 12 vs 10 bit. Some initial tests suggested to me that the difference was lopping off 2 bits and a couple of stops between 14 and 12. But after some more testing now i simple cannot tell the difference between full range 12 bit and full range 14 bit. My exposure is +1 on sky and shadow. And both 12 and 14 bit (and 10) pull back full detail in clouds despite all showing as fully clipped. As a control i compare to Lightroom and what i've realised is that in lightroom where there is no highlight option actually it is taking full advantage of that ALL the time seamlessly. Resolve and Lightroom match.

This is suggesting to me that perhaps the lopping off of 2 bits is in the shadows. Yet between 14 and 12 i cannot break. Whereas 10 is clearly lacking detail.

I have no method here at the moment to test dynamic range (aka i can't find my wedge chart)

So not 100% sure at the moment and more tests to be done.

But 12 bit is very robust - or 14 bit stills are hampered somehow.

cheers
Paul

Link to post
Share on other sites

Okay, so the 12 to 14 difference is in the lower stops. The DNG sets the point at which is EV0, so in the 12 bit that is set two stops lower.

You can see the raw data in the files. So Resolve will bring them in based around the EV0 point. 

But it is interesting at least in this test (outside a window with clipping clouds) that i can't visually really see a difference - when i push the shadows up, and expand that - either on scopes or visually.

This makes some sense now that the difference between 14 and 12 is at the bottom of the range and it's filled top down. So really you are missing two stops at the shadows but actually this is really only a handful of code values at that range.

What i will do is lock off a shot and shoot 8/10/12/14 and compare.

What amazes me about this scene is the level of recovery. The lower picture shows as shot, it's at least 2 stops over. Then pull it down normally and then turn on highlight. This is recovered from the Red channel which hasn't clipped in a lot of places. If you took the image into Lightroom and brought it down the same thing happens automatically. So using Lightroom to judge native dynamic range of an image is not possible (the degree of recovery depends on the scene)

Cheers
Paul 

 

482164217_14bitskytoclip.thumb.png.1c91b97664c847377113622396e74365.png

121957530_12bitskytoclip.thumb.png.d7a78e58eb3065de6a94db891777b34a.pngIn the i

hr.thumb.png.7ad3fef005ffc094cce95bd44d319e5f.pnghr.thumb.png.7ad3fef005ffc094cce95bd44d319e5f.png

Link to post
Share on other sites
2 hours ago, rawshooter said:

Btw. - has anyone found a way how Resolve (16.1) properly recognizes the Sigma fp raw video files as CinemaDNG files, instead of DNG sequences with separate .wav files?

Oh yeah, I didn't even know that was a possibility- I want an answer to this too, having to manually add then layout the audio separately is a pain in the ass.

Link to post
Share on other sites
11 hours ago, Brian Williams said:

Oh yeah, I didn't even know that was a possibility- I want an answer to this too, having to manually add then layout the audio separately is a pain in the ass.

Is there not timecode both in the audio and frames. Usually i sync everything via timecode.

Ah, i just looked. There's no timecode in the wav files!!! But there is in the video. Surely that's a bug and that's why resolve can't automagically do anything.

If you click on a wav and put the starting TC based on the DND sequence and then Auto Sync via TC that audio will be part of the sequence and you can drag it down.

cheers
Paul

Link to post
Share on other sites
2 hours ago, Brian Williams said:

FPLog, or SigLog? Whatever they want to name it, but something similar to Blackmagic’s BRAW? That’s a workaround, right?

They mentioned having a log format but that's not the same as RAW, or BRAW. BRAW appears to part debayer in camera but i don't really know what goes on. To me that doesn't sound like RAW and i've seen it written that it's basically YUV format - I'm sure people here know better.

The beauty of this camera is the RAW output - i just don't see the point of a log baked compressed format to be honest. It would be nice to be able to have a lossless compressed RAW and being lossless i wouldn't have thought that stood on Reds shoes. SlimRAW in camera. Compressed anything can cause issues, even in the Red ecosystem.

As Lars says, there is a deal with Atomos but i don't think there are any definitive specs on this - is it even RAW?

A little 2TB drive is not much or large and you can record to that for an hour or so. Same price as a high spec 256GB card in fact.

cheers
Paul

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...