Announcement

Collapse
No announcement yet.

4.0 Public Beta 2

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #76
    Any chance BM could add a way to calibrate in the future to completely get rid of the FPN?

    Comment


    • #77
      Originally posted by mrbrycel View Post
      Any chance BM could add a way to calibrate in the future to completely get rid of the FPN?
      I will never understand why BMD doesn't have blackshading considering the amount of noise these cameras get complained about.

      Comment


      • #78
        I'm having trouble importing Denver Riddle's Impulz LUTs onto the camera, is that a firmware issue or some source of built in security in the LUT itself? All the other LUTs i've tried import fine.

        Comment


        • #79
          Originally posted by townio View Post
          I will never understand why BMD doesn't have blackshading considering the amount of noise these cameras get complained about.
          How do you know they don't ?

          Do you see it on an Alexa ?

          It's there, it's just not done by the user.

          JB.

          Comment


          • #80
            I keep getting all my overlays disappearing and wont come back on until I factory reset.

            Comment


            • #81
              Originally posted by John Brawley View Post
              How do you know they don't ?

              Do you see it on an Alexa ?

              It's there, it's just not done by the user.

              JB.
              That's right the camera is supposed to auto black shade on its own, the difference is that you never see FPN on an Alexa when properly exposed.
              My point is that I wouldn't mind to wait a few minutes for some manual calibration if it meant getting rid of the noise like on the RED cameras.

              PS:Of course it's not the same price, but maybe something could be done in the software in that regards?

              Comment


              • #82
                The Red Epic is quite a powerful beast and it blows through about twice the amount of electricity and heat…

                OTOH, a perfect sample of the UM46 doesn't show any FPN if you don't lift the shadows beyond a point where the random noise in a Red Epic Dragon will make the image ugly too.

                IMHO, you either need to RMA your camera or you have unrealistic expectations.

                Comment


                • #83
                  Seems no one can agree what is acceptable FPN and what is not.
                  Any chance someone from BMD post some samples of FPN with camera settings on what is acceptable and what need to be RMA?

                  Comment


                  • #84
                    Originally posted by amsh89es335 View Post
                    I keep getting all my overlays disappearing and wont come back on until I factory reset.
                    Do you mean the HUD display which will return if you swipe up or down on the monitor?

                    Comment


                    • #85
                      I have owned Red cameras since 2008. I used my 2 Epic Dragons for TV programs, movies and TV spots and IMHO the noise is very comparable to the 2 4.6k's I received early this year. Noise depends on exposure on both systems. I rate my 4.6Ks at 800 and monitor output film mode connected to a monitor with a waveform. When I set the iris for white poont at 90 percent I am actually shooting at ISO 400 according to my light meter. I shot my Dragons at 250. If you shoot at 800 on the 4.6K you can stretch up the gamma a little but forget about gaining up unless you want noise. I don't feel 1600 is useable unless your doing day for night. But everybody has there own opinion. If you havent worked with underexposed Red and Arri footage I can see where you might expose at 800 and be dissappionted and feel the camera must have a problem. Both my cameras make spectacular images and I see no fixed pattern with proper exposure. I shoot only RAW 3:1.

                      Comment


                      • #86
                        There is no standard for what's acceptable and what's not noise or FPN.

                        In fact, does anyone know how they measure noise when doing tech checks on deliverables for broadcasters ?

                        There's no SMPTE standard for it, there's no machine that measures it.....they EYEBALL it.

                        It's somebody with experience sitting in a room subjectively saying "yeah" or "nah" to the noise. It's often done in context to the rest of the material. You establish a noisy look up front and if you're consistent in the application of it, chances are it will get through. You have a show that's super clean but one shot is a bit noisier than others (maybe it's a stunt or rig camera) then it will get bounced because it draws attention to itself.

                        It's ALL subjective.

                        I think 9 out of 10 FPN problems are users who haven't previously dealt with LOG footage before looking at LOG footage on a computer monitor pre grade...you have the double whammy of LOG and Computer monitoring of a video signal artificially revealing the blacks.

                        Blackmagic's LOG is also especially flat and lifted compared to most others. In fact a big difference with the 4.6K is the LOG starting point. It's now much closer to a more normalised image.

                        So, take some users that aren't used to seeing LOG pictures, show them on a computer pre grade and all of a sudden you see a lot of faults that aren't really faults by the time you normalise the grade and look at them on a television.

                        So what I see 9 times out of 10 are people posting SCREEN SHOTS of the PREVIEW images. That's a big clue.

                        I think there are definitely cameras out there made by Blackmagic that have unacceptable FPN and they should be RMA'd. But I don't think it's nearly as bad as it "appears" to be because it's pretty easy to see FPN in near normal looking images when you're looking in BM's log profile.

                        Maybe what BM should have done is truncate the bottom end and hard clip it more (like other baked in cameras do) and no one would be the wiser. In attempting to give the user as MUCH DR as possible, they give you everything including the FPN. That way you choose where to put your hard clip for black.

                        So what do you want ? A camera that gives you less DR but no FPN, or more DR and the cost of sometimes having to vary the black point in case you get some FPN showing up ? It won't always happen cause it depends on what you're shooting visually and how your workflow is....

                        What do you prefer ? I think most would, like noise reduction, prefer to have more DR and then choose where they clip the blacks.

                        Unless you have an especially poor camera, you should be looking to make you blacks black and instead of making what is photographic black lifted so you can see everything, including the sensor faults.

                        We have the same discussions when it comes to measuring DR. There's no real agreed way to do it and the results ALWAYS vary because it requires a subjective judgement about the noise floor. What's photographic dark detail and what's noise...no one ever agrees on where that line is....

                        That's why most camera manufactures do a theoretical calculation based on an ideal scenario and why rarely are the theoretical numbers ever be photographically achievable. Arri are the only ones I know of that seem to have deliberately UNDER quoted their DR.

                        JB

                        Comment


                        • #87
                          Thank you, John!
                          That's very close to my observations. IMHO the best way to judge what you really have is comparing cameras directly under controlled conditions. I have carefully tested the UM46 next to a Red Epic Dragon and I can just say that the Red has more noise and less headroom too (not counting HDRx). The UM46 has some FPN if you pull the shadows too high, the Red has none (if calibrated properly by myself). BUT at the same point the Red already has more random noise. In the highlights the Red shows some pink clipping where the BM is still clean. How's that for a camera for a fraction of the price?
                          I didn't have an Alexa next to it for testing, but from my former tests between the Epic and Alexa (even using a Xyla 21 at the time) I'd say that the DR of the UM46 gets close to Alexa territory, but the Alexa is keeping better colors into the highlights and shadows.

                          So, my advice to newcomers who never operated a professional digital camera before, try to compare before you RMA yours. BTW, did you ever have a close look at analog 35mm film of high sensitivity pushed by a stop or even two?

                          I don't deny that BM cameras lack some proper QC, having been bitten myself, but please avoid unrealistic expectations. Would you rather have a crappy camera with perfect QC or a quite capable one with some foul apples in the basket at this price point?
                          Last edited by Nomad; 09-23-2016, 11:54 PM.

                          Comment


                          • #88
                            Originally posted by Nomad View Post
                            I have carefully tested the UM46 next to a Red Epic Dragon and I can just say that the Red has more noise and less headroom too.
                            Yep. I always found that Red seemed to weight their middle exposure very high. Once you middle the exposure it always feels like you run out of highlights very quickly, like maybe a couple of stops, but the shadows go on for a lot longer...

                            Originally posted by Nomad View Post
                            The UM46 has some FPN if you pull the shadows too high, the Red has none (if calibrated properly by myself). BUT at the same point the Red has more random noise already.
                            And again, I see people posting the UM4.6K is noisy when it really isn't compared to others. But it probably LOOKS noisy when you're graduating from a baked in NR camera...

                            JB

                            Comment


                            • #89
                              Right, even the" low-light wonder" Sony A7S has noise in RAW photographs…

                              Comment


                              • #90
                                Some LUTs load just fine, others seem to not load at all, is there something in a .cube file itself that restricts it from working?

                                Comment

                                Working...
                                X