Will we ever see it happening? 6k S35 - 48p on Komodo

[Cristian Mihai]

Coming from a Panasonic world and already sold my 2 cameras and literally prospecting to buy a RAW camera and maybe a second camera with AF for more of the run and gun kind of shoots. BUT what is holding me to pull the trigger on Komodo is the lack of the 48p on the S35 6K mode. Will we ever see it happening? I'am in a moment where i dont spend my money on another camera but i need a camera to work ...

 

[Christoffer Glans]

Coming from a Panasonic world and already sold my 2 cameras and literally prospecting to buy a RAW camera and maybe a second camera with AF for more of the run and gun kind of shoots. BUT what is holding me to pull the trigger on Komodo is the lack of the 48p on the S35 6K mode. Will we ever see it happening? I'am in a moment where i dont spend my money on another camera but i need a camera to work ...

No

And even if it was done in some miraculous way, there is far more important stuff to fix through firmware that should be focused on in order to increase reliability and stability of the current features and necessary upcoming features. Like, if there was one major disruptive firmware change that is much more interesting for this camera it would be a face/eye-tracking feature for its autofocus. That would be a lot more important than 8 more frames in full frame. You can still shoot 5K at 48fps, so I don't see the importance of it compared to something like a fully functional facetracking auto-focus.

 

[Cristian Mihai]

I dont know much about RED in general but interested in Komodo as it seems to be the first reachable camera from them to my budget but Canon C70 in the 48p mode does NOT HAVE AF for example so just a
S35 senzor with ND. The only camera with RF mount that has very good AF seems to be R5. For me the 48p is a sweet spot and i want to keep the s35 image circle on my fujinon mk zooms. Regarding eye tracking, that;s the Sony teritory for the moment.

 

[Michael Hastings]

Seems like Christopher has a good point in that 5K ought to be "good enough" vs. other potential firmware goodies - 5K downsampled should provide a true 4K resolution. And I would think that given a global shutter vs the inherent "skew" of any rolling shutter sensor, Komodo 5K should rival or exceed the actual resolution of other 6K rolling shutter cameras.

I'm not talking about the huge, very noticeable difference with fast moving subjects like cars, athletics, propellers etc - I'm talking about the at least a few pixels of skew inherent in any shot with any movement at all that would not be visually noticeable but would at least slightly degrade resolution of the final image.

As far as autofocus, face/eyetracking would seem to be mostly about the algorithms RED can develop and whether the camera processor can handle it in realtime. The limited frame rates may indicate that the processor is already nearly maxed out.

While Phase detect AF is a big improvement over the contrast type in previous RED cameras, I'm not sure it can compete with the thousands of Dual Pixel AF detectors on the Canon R5. Also a question of whether the RED processor can match the Canon. Face/eye tracking is very processor intensive. Canon has been at it for a long time and has some amazing low power ASICs (10 generations or more) to handle those functions.

As mentioned in another thread, I'm on the fence with Komodo. I was initially all in (especially with the $1K Hydrogen "discount") and even sold my Epic-W because of the size advantages with Komodo in an underwater housing. However, we Hydrogens were left out of the "Beta" program and it was simply dragging on too long so bought an R5. Having experienced the AF on the R5 it is going to be hard to decide to go backwards. I mostly shoot the 4K mode that is downsampled from the 8K sensor so haven't experienced much of a problem with overheating and it's nice to have the 8K for extraordinary subjects that I can capture at the highest possible resolution. (If you're also into stills - R5 may be the best still camera in the world right now.)

There are some potential issues with the R5 underwater, so I'm still considering adding the Komodo - problem is we have no idea when we can get one - seems like 3 to 4 months (you can order an R5 and get it the next day.) Three to four months isn't just frustrating - who knows what else will come out in the mean time. The summer Olympics are often used to introduce the latest greatest models.

We could very possibly see DSMC3 announced fairly soon. DSMC2 is getting pretty old in "camera tech years". The timeline was hinted several years ago that "DSMC2 would be around until at least 2020" and we are well past that. Seems like given the market conditions there will probably be global shutter cameras in more normal configurations and frame rates that might suit most of us better at low, medium, and premium price points. I've thought about just placing an order for Komodo and canceling if it doesn't seem right when allocated, but I hate when people do that to me so dislike doing it to other companies. Will probably wait until order to ship times are within a few weeks and evaluate then.

 

[Andrew Reese]

As much as I was originally expecting DSCM3 in 2021, I am starting to doubt that. Maybe announcement of DSCM3 could be a possibility in Fall/Winter of 2021, but I highly doubt DSCM3 cameras would actually be be shipping anytime soon.

Despite being an awesome camera in its current form, Komodo is still basically in beta firmware and still lacking many of the features originally expected at full capacity and functionality (i.e. AF w/ tracking, wireless stability, etc.). Then there is also the fact that RED can't even keep them on the shelves and supply the current demand. I fully believe DSCM3 and DSCM4 is already developed sitting on a bench somewhere at RED HQ, but I am just not sure it makes sense for them to release DSCM3 in the current financial landscape with COVID still looming around and an industry which has taken a great beating over the last year. I have seen more used RED equipment this year for sale at low prices than any before albeit due to financial losses and production agencies closing up shop due to covid. Prices are at an all time low for used cinema gear. So until Komodo is really finished in terms of fully functional OS and has met a year or two of full demand sales I just don't know if releasing the next series of DSCM3 cameras makes sense for RED and would't end up hurting prospective Komodo sales.

Furthermore, all new releases from Canon, Kinefinity, Black Magic and Sony are now just catching up with DSMC2 tech IMO and most still do not offer good codecs and 16-bit RAW at least not internally. Arri and RED are still the top dogs for RAW acquisition and just increasing resolution (albeit at 10-bit) by competitors doesn't threaten DSCM2 to my eyes. Things like Global Shutter, Autofocus Tracking, Lidar Focus, Internal ND's, Gyro Data, etc. are really the game changers here, not more frame rates or higher resolution formats. Full Frame sensors are also a thing of the future and I expect that to play a large role in DSMC3 (think FF global shutter sensor in Komodo sized or slightly larger body).

It will be interesting to see if RED sticks with more of an Arri position in the future with staying on track with current development and not feeling the pressure to drop a new model every 1-2 years like Sony and Canon have begun to do. I sure hope so.... as that is one of the biggest reasons of owning an Arri/RED.... Industry familiarity/reliability and long term value. Those in this business seriously don't want to have to sell their cameras every 1-2 years just to get a few upgrades that are basically firmware updates.

 

[Bérenger Brillante]

No

...Like, if there was one major disruptive firmware change that is much more interesting for this camera it would be a face/eye-tracking feature for its autofocus. That would be a lot more important than 8 more frames in full frame. You can still shoot 5K at 48fps, so I don't see the importance of it compared to something like a fully functional facetracking auto-focus.

This.

40fps is plenty enough slowmo for 6k. You can shoot 6k WS and crop on the side if you need more. the crop is minimal.
Eye tracking/Face tracking/ 3D tracking would be the thing.

I came across one of my post from something around 4 years ago saying the same thing describing a Red dslr.

About DSMC3, I think they said a while ago the annoucement would come before the end of 2021, but delivery is a whole other thing.
I'm eager to see what a new body could do to the line-up, but as most of Redusers, I'm already waiting for a Komodo, well 2 actually, and it has not earn any projects and money yet.

I don't think I will be able to buy a DSMC3 if it comes too soon and I'm pretty sure lots of us are in the same situation, and Red probably knows it.

Considering the semiconductor shortage is here to stay for another year if I believe the media forecast, it wouldn't be a smart move to announce something to deliver it the year after at best. Again.

 

[Christoffer Glans]

As much as I was originally expecting DSCM3 in 2021, I am starting to doubt that. Maybe announcement of DSCM3 could be a possibility in Fall/Winter of 2021, but I highly doubt DSCM3 cameras would actually be be shipping anytime soon.

I wrote in another thread about DSMC3 that I believe that it's reached a point of diminishing returns. When we start asking about what we really need and what we don't need. As I see it, Komodo is in a sense the DSMC3, because it has pushed the way we use cinema cameras to be smaller and more flexible. What else would a DSMC3 lineup bring to the table that hasn't already been done? The Ranger models already feature most things needed for larger sets and Komodo is the go-to indie camera.

A DSMC3 lineup might become just a cosmetic change that would require a huge re-invest that most won't find rational to do. So I proposed that a DSMC3 lineup needs to have a lot more that pushes the industry than just a new camera body or new sensor. There are still projects shot on Dragon that holds up perfectly to the 4k HDR standards of today, so I don't think people will find the urge to upgrade if there's no real heavy improvements to their workflow.

So I think a DSMC3 lineup needs the features that would revolutionize the link between shooting and post. Z-depth information and in-camera tracking data for 3D software and compositing including post-stabilizing (to be able to one-click extract a virtual camera that is synced to what was shot) are two of those types of improvements that is a major industry leap. People seem to think that this would only benefit big VFX driver projects, but even minor VFX and compositing edits are done today on small projects. It would also help with normal stuff like removing unwanted objects in frame, stabilize shots as a RAW slider function rather than time-consuming-trying to get tracking to work-and then everything just warps badly.

Unfortunately, I think that the people who are close to camera manufacturers like Red are often people who aren't concerned with changing the status quo, they don't care for things like the above mentioned so they don't lobby the manufacturers to design any such features. They want more stops, and more of the things we already have. But since the entire camera market starts to get over-saturated with cameras that are on the same level or so similar that people have a hard time deciding what to get, more of the same won't really cut it anymore. For a lot of projects, I don't see any reason why shooting on Monstro would benefit over shooting regular Dragon 6K. I mean, we can pixel peep and go into the fine details to find differences and reasons, but for the workflow and delivery of a regular project, the final result will still be more than enough for almost 99,99% of the viewers, which includes everyone watching on expensive 4K HDR TVs at high bitrate streaming. It becomes esoteric rather than rational.

As someone who does most things from shoot to delivery, the amount of time I put into tedious post fixes with tracking, stabilizing, object-removal, compositing and so on, if all that were close to instant, it actually opens up how you can plan and shoot something more efficiently and less costly while also giving you more power to make advanced stuff that was previously only available to high budget projects.

There's so much innovation that could be done on the camera side that I don't see any camera manufacturer really taking advantage of. Having more advanced measurements while shooting and more options available in post doesn't just help VFX heavy shoots but everyone who needs to fix something in post, which happens more often than people who just work on shoots seem to understand.

 

[Han Vogen]

This is a nice to have feature, but I’d prefer that they enable the gyroscopic sensor. Doing this would be a huge step up in terms of post stabilization effectiveness.