FS7: Urgent FS7 / 7Q slow raw problem - It's all gone a bit 80s Michael Mann

I believe the main concern people are having is that whilst we have a workaround at the moment thanks to Nick's work, they want to use their 7Q recorded footage without having to jump through hoops at the other end of the production.

Banding in the shadows in raw is one thing, but will .dngs no longer automatically be magenta after the next firmware updates, and will pro res recorded no longer have a ton of added contrast compared to slog3? Those are the questions - not unreasonably I think - that everyone wants to know the answers to.
 
Last edited:
Ben Scott said:
I believe the main concern people are having is that whilst we have a workaround at the moment thanks to Nick's work, they want to use their 7Q recorded footage without having to jump through hoops at the other end of the production.

Banding in the shadows in raw is one thing, but will .dngs no longer automatically be magenta after the next firmware updates, and will pro res recorded no longer have a ton of added contrast compared to slog3? Those are the questions - not unreasonably I think - that everyone wants to know the answers to.
Click to expand...

Yes, that is the question I have not heard a straight answer to from Convergent Design. It is great that LUT's have been created as a workaround but it is really not a real solution. Will FS7 / OD7Q+ be able to capture RAW images properly without the need of this workaround anytime soon?
 
broughtonfilm said:
Yes, that is the question I have not heard a straight answer to from Convergent Design. It is great that LUT's have been created as a workaround but it is really not a real solution. Will FS7 / OD7Q+ be able to capture RAW images properly without the need of this workaround anytime soon?
Click to expand...

Yes, new firmware is out : http://www.dvxuser.com/V6/showthrea...y7Q-and-Odyssey7Q&p=1986528926#post1986528926

Download link : https://convergent-design.com/odyssey7q-firmware.html
 
One must process DNG files no matter what. A LUT has to be applied to do anything with them. We are now giving you a better LUT.

If it walks like a duck & talks like a duck, it's a duck.

As for the contrasty ProRes, I'll look into that but we are moving forward with that. First and most important was to correct the magenta cast. Note that while we are moving forward, the next firmware will have our custom 3D-LUT implementation, and really in the end what we are talking about is a better 3D-LUT application in the machine. So it all goes hand in hand. And Nick has been a great help.
 
Well it's not just a LUT that you can stick on the footage and its done. It's a Resolve-only workflow workaround. What about the actual cause of the issue? Again I don't think it's unreasonable to not want dngs straight out of the recorder to natively show the characteristics in the image I posted in the first post of this thread, and to show those same characteristics in previews.

Is this the thing that has been fixed in this new Odyssey firmware which states it has fixed the magenta cast in the notes?
 
Last edited:
Mitch Gross said:
Gentlemen, What I can tell you is that the LUT correction was for color matrix issues that led to a magenta color cast. What you are seeing appears to be shadow banding issues in the 2K RAW data stream. We are looking into this to see if there is anything we are missing or that Sony is improperly sending to us, but this is a separate issue.
Click to expand...

I can confirm this banding is not a result of the LUT. I took a second look at the footage, and while it isn't as easy to see, the stepping in the green channel is definitely present even without the LUT. Hopefully Sony will fix this in the upcoming firmware update.
 
I do not have an FS7 myself, so I cannot comment on what has or hasn't changed with the new firmware. However there are four separate issues under discussion here.


1. The inappropriate metadata with the 7Q puts in DNGs (in some cases) causing a dramatic magenta shift in the white balance on decode.


My white-balancing Raw LUTs are a work-around for this.


The way other software decodes the DNGs is too far removed from the intended result so a correction is not possible.


If the firmware update removes the metadata issue, these versions of the LUTs will no longer be necessary.


2. No software (AFAIK) offers S-Log3/S-Gamut.Cine decoding of DNG files.


My Raw LUTs take advantage of the BMD Film decode offered by Resolve to provide transforms to standard Sony colour space and gamma. Other software which does not offer a decode to a known log coding makes it difficult (sometimes impossible) to transform accurately to S-Log3.


For S-Log3 based workflows this is necessary, not a work-around.


3. The RAW->ProRes in 7Q firmware v4.10.100 does not match standard S-Log3.


My ProRes LUTs are a work-around for this.


If the new firmware changes the RAW->ProRes processing these LUTs may no longer be necessary. I have not had an opportunity to test if this is the case.


4. Banding in the shadows of 2K HFR Raw recordings.


This appears to be in the Raw data. It is not caused by the LUTs, but they may make it more visible.


As Mitch has said, Convergent Design are investigating the difference between the Raw and AVC-I. I am not involved in this, so not in a position to comment.
 
You must log in or register to reply here.
Back
Top