Odyssey7Q February 2014 Firmware Update

paulcurtis said:
I shot RAW properly over the weekend, fairly gruelling shoot and moving fast. Here are my observations:

1. The elevated blacks in the current display/recording of RAW reduce contrast when nailing focus, i don't know whether they affect peaking or not but visually it makes a little more difficult to 'see' focus. I ended up messing around with the peaking settings to find a workable solution. Same with the exposure tools. I believe this is sorted in the next firmware

2. I did have some freezes and crashes. I managed to get the O7Q to hang between a lens change where i turned the 700 off before the monitor which required a pull the power on the monitor and reconnect (because nothing worked on the screen, i held the shut down button for a while in case that would also reset but not). Also when in some menu options on the 700 the raw signal seems to output vertical noise patterns and that seemed to confuse the O7Q. Not a massive issue because resetting is easy. I don't have exact steps to reproduce as i didn't have time

3. The one LUT that's in there is of limited use, i think this is being addressed as well but i don't know when. Whilst i may understand the log look a lot of others don't. We were dealing with some quite saturated light and so understanding how that is being recorded with a log view is very, very difficult. That single LUT helped a little but it wasn't ideal. We were also matching light from previous shoots and that made it doubly difficult to see what the camera is seeing. The fact that the WFM changes with a LUT applied reduces the use a little as well. Ideally the WFM and exposure tools show the RAW levels and the LUT is just for a look on the monitor.

4. I've not totally sorted the post out. I'd done some tests that indicated i could use clip house as a quick way to transcode the RAW files into something more realistic but due to the bright colours and the default tint issues so far that's proved a complete no go. The looks are totally out of whack with what was on set in terms of colour. I will try the resolve route. I'm fairly confident that the data is in there and nothing clipped but the balance is very difficult to monitor and test for at the moment. This issue of accurate colour monitoring off it is quite an achilleas heal right now.

5. The issue of what colourspace is coming out of the camera is confusing still for me. The camera is set to Slog yet the cDNGs are linear RAW. The monitoring is log too. I see in other threads that people have been taking that into resolve -> BMD film -> LUT but to me that BMD film is perhaps logging the linear and therefore is redundant (i'm assuming the LUT at the end is linear log - so the route is Linear RAW -> Log -> Linear to Log with the middle step not needed?). The Adobe route at the moment is pointless AFAIK because you just get to select a 'look' from the RAW not all the data. So i'm looking for the best way of turning that RAW into a Log based file for post.

6. I'm still not sure what the deal is with WB and recording. Changing WB on the camera does change the look on the monitor (in a counter intuitive way) but is this balance baked in or not to the recording? I think it is because it is changing the signal being output. No biggie, just a case of knowing.

Broadly speaking the O7Q performed very well, it's not so heavy so hand holding was possible. The screen is a joy to work off. I think there's just some UI and workflow issues along with some teething problems - some of which might well be a lack of knowledge on my part!

The post debayer methods do make a massive difference. On clip house there are 3 methods of debayer and each produce quite different edge results. Some with increased magenta/green artefacting and some smoother. I've not been able to side by side and compare with Resolve or Adobe yet, but that is a comparison that's very worthwhile to try.

cheers
Paul
Click to expand...

1 - This is being addressed in the Feb. firmware update.

2 - The FS700 can sometimes send odd signal bursts over SDI when switching modes or on shut down. We will look into this to see about preventing such issues. It is true that these signals can sometimes prompt the Odyssey7Q to record when it is set to Camera Trigger.

3 - This is being addressed in the Feb. firmware update.

4 - I'm not to sure about Cliphouse, but Resolve should be solid workflow and Adobe will be available with the Feb. firmware update.

5 - This has to do with the way the FS700 encodes its RAW. It can be in Slog2 or in Rec709. We recommend Slog2, which is the "assumption" placed upon the material when accessing it via Resolve.

6 - On FS700 RAW, the white balance is baked into the recording.
 
Mitch Gross said:
4 - I'm not to sure about Cliphouse, but Resolve should be solid workflow and Adobe will be available with the Feb. firmware update.
5 - This has to do with the way the FS700 encodes its RAW. It can be in Slog2 or in Rec709. We recommend Slog2, which is the "assumption" placed upon the material when accessing it via Resolve.
Click to expand...

Yes, i've managed to get it working in Resolve. I couldn't set the cDNG settings to project and BMD film. Instead i had to set to clip and then the white balance As Shot and then those other options become available (they are greyed in the Project setting). Pop the LUT on top and good to go. Getting some nice colour out of it (finally!)

thanks
Paul
 
The 4K RAW recording is a great news !

On another matter : do you you think the FS700 would benefit in any ways of the new S-Log 3 curve ? Or is it pointless ?
 
Why not ?

I mean with a firmware update (without being forced to send back the camera please) IF that was possible, would there be any benefits ?

I understand that it's best paired with S-Gamut 3 but since the F5 doesn't have the wide gamut sensor of the F55 and it's sensor is "said to be" really close to the FS700 one, shouldn't the FS700 benefits also from having S-Log 3 ?

S-Gamut 3/S-Log 3 is presented as being closer to film stock in the way it's sensitive to light and being the closer one to the sensor native colorspace, dixit Sony's whitepaper.
 
S.Blanchardon said:
The 4K RAW recording is a great news !

On another matter : do you you think the FS700 would benefit in any ways of the new S-Log 3 curve ? Or is it pointless ?
Click to expand...
Well, it's a rather theoretical question, isn't it? There is so much more to a camera, even a RAW one, than the sensor. Without being a Sony engineer knowing a lot of info of what's going on under the hood it really is impossible for me to know.

That said, there's a reason that Slog2 is not being replaced by Slog3, but instead remains available concurrently. They are different ways of driving the sensor.

In a nutshell, I would say that Slog2 is a classic Sony technical way of getting the most out of the sensor. It takes the device to it's extreme ranges and gets the most signal reasonably possible from it. It is technically the best result from the camera. On the other hand, Slog3 derives an image with a different curve on the toe and especially the knee slope, rolling off the highlights. While this may not be the technical outer range of the sensor, it more closely emulates how film (remember film) reacts to light and extreme under and over exposure. Therefore the result is a more "filmic" response and some (many) would argue a more pleasing one. In many ways, this is far more Alexa-like in its look.

So would the FS700 benefit from this? Sure, why not? It's nice to have choices. But understand that Slog2 is already an accurate mapping of your camera's sensor to pull the most possible useful information from it.
 
ShermanUS said:
Wow !!!! I take my hat off to dan & mitch
Click to expand...
Thanks guys, but I want to take a moment to point out that while Dan & I are perhaps the most public faces of Conergent Design, there are a lot of other people doing athe lion's share of the heavy lifting. Dan and I are not crunching numbers all day, working out code. Dan & I are not assembling Odyssey7Q's and putting them through QC testing. Dan & I are not designing circuit boards, or dealing with board stuffers, parts suppliers, or accounting. We do our part, but there are a bunch of others who do their's.
 
Sorry if this has been asked a thousand times, but just to clarify, is 4K compressed in a popular codec recording from the FS700 likely to be added to the 7Q firmware in the next couple of months? I know 4K raw has just been announced, which is great for many no doubt, but for me the file sizes are unmanageable in the majority of cases (perhaps for small shots I could use it). But 4K at a 200-300mbps compressed codec (ProRes ideally) would be great. I know it's been discussed, just a bit confused whether its definitely coming and the timing of its release. Thank you.
 
Just got our O7Q, lovely unit!

Mitch- presumably if you have enough bit depth, Slog2 vs Slog3 difference is nothing you can't easily replicate in resolve, right? Or is there some intrinsic value to applying the corrections in the camera space which is difficult to replicate afterwards?

Kris
 
pythonman said:
Sorry if this has been asked a thousand times, but just to clarify, is 4K compressed in a popular codec recording from the FS700 likely to be added to the 7Q firmware in the next couple of months? I know 4K raw has just been announced, which is great for many no doubt, but for me the file sizes are unmanageable in the majority of cases (perhaps for small shots I could use it). But 4K at a 200-300mbps compressed codec (ProRes ideally) would be great. I know it's been discussed, just a bit confused whether its definitely coming and the timing of its release. Thank you.
Click to expand...

+1!
once we can record either 4Kraw AND 4K10bit (something like ProRes) for type of shooting session, OD7Q is going to be the best gear for supporting our FS700 couple more years at least!

freiheit
 
Mitch Gross said:
Well, it's a rather theoretical question, isn't it? There is so much more to a camera, even a RAW one, than the sensor. Without being a Sony engineer knowing a lot of info of what's going on under the hood it really is impossible for me to know.

That said, there's a reason that Slog2 is not being replaced by Slog3, but instead remains available concurrently. They are different ways of driving the sensor.

In a nutshell, I would say that Slog2 is a classic Sony technical way of getting the most out of the sensor. It takes the device to it's extreme ranges and gets the most signal reasonably possible from it. It is technically the best result from the camera. On the other hand, Slog3 derives an image with a different curve on the toe and especially the knee slope, rolling off the highlights. While this may not be the technical outer range of the sensor, it more closely emulates how film (remember film) reacts to light and extreme under and over exposure. Therefore the result is a more "filmic" response and some (many) would argue a more pleasing one. In many ways, this is far more Alexa-like in its look.

So would the FS700 benefit from this? Sure, why not? It's nice to have choices. But understand that Slog2 is already an accurate mapping of your camera's sensor to pull the most possible useful information from it.
Click to expand...

Mitch, thanks for taking time to answer my question, your explanation is crystal clear !

So if S-Log 2 is already getting out the most of the sensor, the next best thing in line for us greedy FS700 users is to have some nice LUTs to get close to "film look" ? (yes, I do remember film ! :grin:)

What is the "level of interoperability", if I can say so, between the LUTs that are gonna be implemented in the OD7Q and the one we can have inside Resolve ?

O-T: it's fun to notice that even if the technology in audio/video is moving at a tremendous pace, one always want to emulate as close as possible analog technology, be it film stock for moving pictures or analog tape for audio !
 
SLog2 is baked in like the white balance and ISO settings into the RAW signal. Weird but yeah that is how Sony does it on this camera. To alleviate this the RAW signal would have to come more "upstream" of this processing.

And I agree about SLog2 vs SLog3. 3 ends up looking nicer. I wish we could have 3. Just have to try to replicate it in post.
 
slipstrea808 said:
SLog2 is baked in like the white balance and ISO settings into the RAW signal. Weird but yeah that is how Sony does it on this camera. To alleviate this the RAW signal would have to come more "upstream" of this processing.

And I agree about SLog2 vs SLog3. 3 ends up looking nicer. I wish we could have 3. Just have to try to replicate it in post.
Click to expand...

Nicer? Neither look nice and pretty much absolutely have to have nicer grading added :P

I wouldn't want to have to light or monitor in Slog2 or 3, but they are ok for storing your image in a known container that is easy to apply prebuilt LUTs to. Frankly I'd probably prefer to just always shoot with a gamma/LUT curve that isn't log but looks pretty and keeps all the sensor data unclipped and without the annoying problems that all the non-slog2 gammas have on the fs700.
 
pythonman said:
Sorry if this has been asked a thousand times, but just to clarify, is 4K compressed in a popular codec recording from the FS700 likely to be added to the 7Q firmware in the next couple of months? I know 4K raw has just been announced, which is great for many no doubt, but for me the file sizes are unmanageable in the majority of cases (perhaps for small shots I could use it). But 4K at a 200-300mbps compressed codec (ProRes ideally) would be great. I know it's been discussed, just a bit confused whether its definitely coming and the timing of its release. Thank you.
Click to expand...
It's next up on the list. So yes, we do expect to have it reasonably soon. Can't be more specific as I don't want to disappoint.
 
RTMotion said:
Just got our O7Q, lovely unit!

Mitch- presumably if you have enough bit depth, Slog2 vs Slog3 difference is nothing you can't easily replicate in resolve, right? Or is there some intrinsic value to applying the corrections in the camera space which is difficult to replicate afterwards?

Kris
Click to expand...
There is value to doing it first as there is only so much information carried over, but the difference is fairly minor.
 
You must log in or register to reply here.
Back
Top