Thread: The next generation features of cameras

Reply to Thread
Page 2 of 4 FirstFirst 1234 LastLast
Results 11 to 20 of 32
  1. #11  
    Senior Member Christoffer Glans's Avatar
    Join Date
    Jun 2007
    Location
    Stockholm, Sweden
    Posts
    4,221
    Quote Originally Posted by Steve Sherrick View Post
    I think on the surface, this seems like something RED would be primed to do since their cameras are used a lot on visual effects films and they have a lot of experience now when it comes to pushing boundaries and developing a following. I simply just don't know the economics of their company to know where they currently stand in terms of investment into R&D. Can't imagine that Jarred will stand still. I would expect they are probably already looking into these things to see how they can take advantage of current tech and what is needed to develop things that aren't quite there yet.
    I can say that since Arri released their Mini LF, almost no one in Sweden rents a Red anymore. Sweden's industry has been pretty focused on Arri for the most part but with the Mini LF, it's a death blow to anything Red related. If this is true for more places around the globe then Red has to do something else than just Ks, DRs and so on. They need to focus on areas where Arri is lacking and will be lacking going forward. Red is often much further ahead than others, while Arri is behind but more reliable. The choice has to be one of those and if you are focusing on reliability you are competing directly with Arri on their turf, while competing in innovation you're on Red turf. If Red innovates with features like the ones mentioned, they would be the only ones having them for cinema cameras and will own the part of the global industry that needs such features. If anything, it would be a wise business move in order to compete against the Mini LF.
    "Using any digital cinema camera today is like sending your 35mm rolls to a standard lab. -Using a Red is like owning a dark room."
    Red Weapon 6K #00600
    Reply With Quote  
     

  2. #12  
    Senior Member
    Join Date
    Aug 2008
    Posts
    4,108
    As size, thermal and power envelopes drop with each new generation of chips and other electronic components it becomes easier to fit more powerful sophisticated tech into a given physical design package. Cost, vs development time, vs benefit trade offs for any particular set of design goals for a particular camera is an ever more complex engineering challenge.
    Jarred has mentioned that as they turn on software features in Komodo, power consumption, currently somewhere in the 27 W range, continues to rise. What is the ceiling for that? 30W? 40W? How will that impact battery life? Thermal loads? Image noise? Long term reliability?
    On the other hand there is much to be said for simplicity. I find shooting a camera whose outstanding feature is a total absence of features beyond the most basic necessities quite liberating.
    Reply With Quote  
     

  3. #13  
    I am for the most part an owner operator and in some of the projects I film I am a one man band. Here is my two cents.
    Don't need the built in camera stabilization, I use a gimbal with good to great results.
    Don't need the autofocus but it would be nice to have.
    3D tracking metadata is something many of us have been asking for, I would like to see it being implemented.
    Wireless video is the one feature I would want the most in this list, I think it's a must have for any serious production.
    Couldn't care less about wireless audio.
    What I really really want to see in the future is a full RGB sensor with no debayering of any kind and with global shutter.
    Just my humble opinion.
    Reply With Quote  
     

  4. #14  
    Senior Member Blair S. Paulsen's Avatar
    Join Date
    Dec 2006
    Location
    San Diego, CA
    Posts
    5,199
    Perhaps the most elegant solution would be a better selection of modules, targeted to particular use cases. Keeping the camera brain small is desirable so it can be rigged to suit myriad situations.

    One of those module offerings could support one man band/docco style shooting situations implementing several of the capabilities noted in the video. A different one could be more DXL-ish with features aimed at fully crewed shoots. I can also see use cases for a Ranger style rig - imagine a DSMC3 brain docking into a Ranger body. RED could still market a "known" configuration to suit the rental house market, with the advantage of potentially being able to swap out brains down the road as the tech evolves.

    IMHO, the modular approach of DSMC2 is a great concept that has not been well exploited to date. OMODs got some things right, others not so much. As noted by others, there's a lot steps on the development road. I know Mark has worked very hard to push forward through all the politics, technical issues, etc - but even with his knowledge and connections it still takes a lot of lead time. If DSMC3 retains the modular concept, I hope that RED is able to engage with 3rd party module makers far enough advance to support development of modules that offer desired capabilities.

    FWIW, I also see the potential for a smart phone based approach. Very interested in seeing how the upcoming Komodos integrate with phones, could be a preview of coming attractions...

    Cheers - #19
    Reply With Quote  
     

  5. #15  
    What I cant understand is. Why on earth did the steadyXP build just a little plastic shitty GoPro battery powered unit?

    To me if they instead of the plastic thing would have made a Vlock 2 Vlock milled aluminium cake to sit in between camera and battery worked reliably and looked somewhat professional they could have sold those for + 2000USD a pop to steady cam ops etc.

    And yes they try to lock in their features into their own app instead of pushing out raw data or something that could be applied in any post app.


    And yes its a shame that red has such poor gyros in the DSMC2, or if its just the readout that is poor. I spent quite a bit of time and money investigating the RED SDK to extract gyro data, its there but its possibly the worst gyro data I ever seen. Its as bad and lagging as the leveling tool on the touch screen which I find very odd, every smartphone made has a really good gyro inside, one that can be calibrated and used as a quite accurate leveling tool with very little or close to no lag.

    I also looked into manufacturing a gyro module like the steady XP stuff described above, even got permission from Autodesk to use their universal FBX format for ti. But as camera developments goes so quick these days I simply did not pull the trigger as its not really complex stuff and as soon as one camera manufacturer does this, the rest will need to follow and in such scenario being a third party provider of such system would not fly well.

    This is how far we got with the DSMC2 gyro. Dont buy it as there is no real good use for it as the data coming out of the camera is simply very shitty. And the mac version does not kick on later iterations of OSX than I think, high sierra last Open GL OSX. on PC it should fly properly.

    http://www.syndicate.se/axis/

    But the program will work with komodo files and with a bit of luck Komodo has a better gyro than DSMC1 and 2.
    Björn Benckert
    Creative Lead & Founder Syndicate Entertainment AB
    +46855524900 www.syndicate.se/axis
    VFX / Flame / Motion capture / Monstro
    Reply With Quote  
     

  6. #16  
    Senior Member Tom Gleeson's Avatar
    Join Date
    Nov 2007
    Location
    Sydney
    Posts
    1,347
    Quote Originally Posted by Christoffer Glans View Post
    The next company that actually puts things like these into their next lineup of high-end cinema cameras will win first prize as the top brand for cinema cameras. And I'll bet that the next Emmy for technical achievement would be up for grabs for any company that actually puts this into their cinema camera.
    Christopher,

    I believe manufacturers have already started doing it. Check out the Sony FX9. It has an effective, configurable and practical auto focus system. Not perfect AF but I wish I had the same on my Monstro. The Sony also records gyro information into its video files that can be read by its own Catalyst Software to allow the stabilisation of footage. The Canon C500mk2 goes one step better and stabilises images actively in the body. (3D tracking in camera would be a game changer but accessing that IP may not be straightforward.) The Sony FX9 will also natively transmit video to a phone App that allows others to see the camera image. It's not high resolution and it has a small delay but the camera will even upload proxies to a server around the world as you continue to shoot. Sony has chosen to pioneer these "2020" features in the "medium level" FX9 but if Sony integrates this tech into their high end Venice its a game changer. I seriously hope Red is not ignoring the FX9.

    At the ACS we did an in depth look at the FX9 and you can find it here on https://vimeo.com/381979485 or https://www.youtube.com/watch?v=XkoG...r&index=2&t=0s. Unfortunately the Software stabilisation was released a few days after we published so we did not test it.
    Tom Gleeson
    Reply With Quote  
     

  7. #17  
    Quote Originally Posted by Tom Gleeson View Post
    Christopher,

    I believe manufacturers have already started doing it. Check out the Sony FX9. It has an effective, configurable and practical auto focus system. Not perfect AF but I wish I had the same on my Monstro. The Sony also records gyro information into its video files that can be read by its own Catalyst Software to allow the stabilisation of footage. The Canon C500mk2 goes one step better and stabilises images actively in the body. (3D tracking in camera would be a game changer but accessing that IP may not be straightforward.) The Sony FX9 will also natively transmit video to a phone App that allows others to see the camera image. It's not high resolution and it has a small delay but the camera will even upload proxies to a server around the world as you continue to shoot. Sony has chosen to pioneer these "2020" features in the "medium level" FX9 but if Sony integrates this tech into their high end Venice its a game changer. I seriously hope Red is not ignoring the FX9.

    At the ACS we did an in depth look at the FX9 and you can find it here on https://vimeo.com/381979485 or https://www.youtube.com/watch?v=XkoG...r&index=2&t=0s. Unfortunately the Software stabilisation was released a few days after we published so we did not test it.

    Camera gyro stabilization needs to know the lens, Red had this for a longer than all the others trough their early EF an nikon mounts. Aslo they been on the forefront on resolution. With a decent gyro and very little bit of coding red would have killed the competition for kinds of camera stability work. It hurst, it´s like watching someone jumping up way above the basket holding the ball firmly with both hands... and then for some odd reason not slam dunking it.
    Björn Benckert
    Creative Lead & Founder Syndicate Entertainment AB
    +46855524900 www.syndicate.se/axis
    VFX / Flame / Motion capture / Monstro
    Reply With Quote  
     

  8. #18  
    Senior Member Nick Morrison's Avatar
    Join Date
    Nov 2011
    Location
    Brooklyn
    Posts
    8,829
    Quote Originally Posted by Tom Gleeson View Post
    Christopher,

    I believe manufacturers have already started doing it. Check out the Sony FX9. It has an effective, configurable and practical auto focus system. Not perfect AF but I wish I had the same on my Monstro. The Sony also records gyro information into its video files that can be read by its own Catalyst Software to allow the stabilisation of footage. The Canon C500mk2 goes one step better and stabilises images actively in the body. (3D tracking in camera would be a game changer but accessing that IP may not be straightforward.) The Sony FX9 will also natively transmit video to a phone App that allows others to see the camera image. It's not high resolution and it has a small delay but the camera will even upload proxies to a server around the world as you continue to shoot. Sony has chosen to pioneer these "2020" features in the "medium level" FX9 but if Sony integrates this tech into their high end Venice its a game changer. I seriously hope Red is not ignoring the FX9.

    At the ACS we did an in depth look at the FX9 and you can find it here on https://vimeo.com/381979485 or https://www.youtube.com/watch?v=XkoG...r&index=2&t=0s. Unfortunately the Software stabilisation was released a few days after we published so we did not test it.
    I have a feeling RED understands where the industry is going.

    You can see them testing ideas in the Komodo.

    A lot of that will get folded into DSMC3.

    I have no fear.

    Also I'm looking forward to an 8K Gemini sensor, or something similar that's IMAX size.

    Full Frame is starting to feel so 2019.
    Nick Morrison
    Founder, Director & Lead Creative
    // SMALL GIANT //
    smallgiant.tv
    Reply With Quote  
     

  9. #19  
    Senior Member
    Join Date
    Dec 2010
    Location
    Toronto & Vancouver
    Posts
    3,906
    Quote Originally Posted by Nick Morrison View Post
    I have a feeling RED understands where the industry is going.

    You can see them testing ideas in the Komodo.

    A lot of that will get folded into DSMC3.

    ...

    Full Frame is starting to feel so 2019.
    Yeah no doubt, but Komodo isn't out yet (much less DSMC3), whereas most other manufacturers have a lot of these features implemented already (especially the FX9) and have had some for quite a while now (like AF). If RED understands where the industry is going, it'll be because everyone else is already there (except Arri, but they still have the best image with 10year old sensor tech, so they get a pass).

    Quote Originally Posted by Björn Benckert View Post
    Camera gyro stabilization needs to know the lens, Red had this for a longer than all the others trough their early EF an nikon mounts. Aslo they been on the forefront on resolution. With a decent gyro and very little bit of coding red would have killed the competition for kinds of camera stability work. It hurst, it´s like watching someone jumping up way above the basket holding the ball firmly with both hands... and then for some odd reason not slam dunking it.
    On the topic of gyro data, what's to stop a phone from being used like a "SteadyXP unit"? That'd be pretty ideal for Komodo since it's looking like phones will be the primary display for a lot of people (so it could be built into the RED display app). Similarly, it wouldn't be surprising if Komodo's "global shutter" is actually in-camera/actively applied gyro stabilization (like the C500m2's) combined with a really-fast rolling shutter that mitigates 99.99% of skew.

    As other's have mentioned, I'm inclined to believe the reasons for these things not being implemented sooner/faster is likely IP, licensing, and legal reasons, more than anything else/engineering.
    Reply With Quote  
     

  10. #20  
    Quote Originally Posted by Mike P. View Post
    Yeah no doubt, but Komodo isn't out yet (much less DSMC3), whereas most other manufacturers have a lot of these features implemented already (especially the FX9) and have had some for quite a while now (like AF). If RED understands where the industry is going, it'll be because everyone else is already there (except Arri, but they still have the best image with 10year old sensor tech, so they get a pass).



    On the topic of gyro data, what's to stop a phone from being used like a "SteadyXP unit"? That'd be pretty ideal for Komodo since it's looking like phones will be the primary display for a lot of people (so it could be built into the RED display app). Similarly, it wouldn't be surprising if Komodo's "global shutter" is actually in-camera/actively applied gyro stabilization (like the C500m2's) combined with a really-fast rolling shutter that mitigates 99.99% of skew.

    As other's have mentioned, I'm inclined to believe the reasons for these things not being implemented sooner/faster is likely IP, licensing, and legal reasons, more than anything else/engineering.
    Yes I guess it would be quite easy to make an iPhone app making he phone do the same job as a steady XP. But there is a few buts... The phone would need to be mounted in a known way, like flat ontop of camera facing a certain direction. You would also need to feed synk from the phone, so you would need to wire the phone headphone jack / cable into the camera sound in and feed sound tc/sync signal. And Im not sure possibly the phone could run such app, recording the gyro data from the phone while also getting a feed from camera for picture etc. But to me thats a bit messy, not so much better than the steady XP, yes you dont have to deal with the GoPro battery, and a phone is possibly more reliable. But it would kind of, to me, be like replacing one bad solution with another and I think the steady XP is a really really good gyro, phones might not be having the same spec. I think they sample very precise rotations at 160Hz or such. Dont know the frequency of Iphones possibly its the same but could be less. The DSMC2 got something like 10Hz...

    Now I would be very surprised if komodo does not have an internal gyro. Very likely Komodo has the same kind of horizontal indicator etc as the DSMC2 has and if it does it also very likely write the gyro data into the r3d´s the same way as DSMC2 does. And at this date it would be strange if the gyro on Komodo, or / and the readout from it is not more accurate and higher frequency than the DSMC2. If so, if its actually fast and accurate then I will give my little Syndicate AXIS app just a little some more attaention and make it so it creates one fusion camera stab setup for each clip on a mag right there in the r3d folders as a one button click operation.

    We pretty much have the above done allready, but with the current quality of the DSMC2 gyro output, I just halted the AXIS project as the noise introduced from poor gyro data is bigger than the shakes you normally would want to remove. But we have built it on the last SDK so it will in its current state be able to detect and read Komodo files and if there is a gyro and the gyro data is good. Its basically a gun ready to fire.

    At around 55s into this rough demo you see DSMC2 data pulled using AXIS play inside Flame, if this was good gyro data coming in then: roll, pan, tilt would simply be locked / not moving. Now there is a heavy gitter and also a drift due to the inaccuracy of the DSMC2 gyro.

    Björn Benckert
    Creative Lead & Founder Syndicate Entertainment AB
    +46855524900 www.syndicate.se/axis
    VFX / Flame / Motion capture / Monstro
    Reply With Quote  
     

Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts