Personal View site logo
Make sure to join PV on Telegram or Facebook! Perfect to keep up with community on your smartphone.
Nikon D5200 topic
  • 651 Replies sorted by
  • erased and inserted in previous post

  • @apefos Great song! Love it big time!

  • @Craftwerk As I said in some old posts in other topics, I just would like make friends here, (but I need the truth!)

  • @apefos "(but I need the truth!)" Aren't we all? ;-)

  • The horizontal pattern noise is there, you can see it in the blacks here @ 100 ISO if you push the exposure enough. It doesn't look as bad as Canon vertical noise so far. Looking at te histogram the noise is deep in blacks and goes away by slightly lifting them.

    Banding/posterization if it's there would be due to the 8bit codec, recording 10bit prores externally could remedy any banding. Even the 1DC has banding. The horizontal pattern noise is a sensor issue.

    Lets try to stay focussed on the facts yeah?

    Check the manual pp 86-88 picture profiles

    @apefos It was actually Samuel who put me onto the D5200 when he mentioned it in a post last week, it flew under my radar for a month! I did some pixel peeping and was blown away. I started my film career with a D90 so it's like returning to an old friend.

    My camera is in a warehouse 30min away but I'm "waiting for payment to process". grrr.

    I've put together a 2 camera rig for the D5200 and the MKIII. The D800 comparison test is interesting but it's only a true high ISO test if you're shooting everything @ infinity. I'll be testing the cameras with aperture set to match DOF and ISO set accordingly. I wanna know how it does shooting narrative not landscapes.

  • @Mimirsan

    Why you so grumpy?

    Because sometimes it seems like no one else bothers to track these rumors down:

  • Let me explain something to you all:

    When a camera is rated at 900 usd in USA market it is inexpensive for American people because the salary or freelance job is enough to buy it.

    Salary or freelance job here is something half value comparing to usa payments.

    Also, here in Brasil, 900 usd turn into 1800 usd due to import customs tax, so government get 900 usd.

    Also brazillian money is 2 times low value than usd dollar. So this camera here is rated at 3600 in our money, sometimes we can find a deal for 3000.

    So buying a D5200 here is the same when American people buy a 5dMarkIII

    So we need to be sure about what we are getting from a camera before buy, there is no room for experiment.

  • @Craftwerk This is why I will NEVER upload any tests I do. If I do the tests for my own interest and find something worth mentioning I will report it as a written post based on my own judgement. If someone's interested I assume they will follow it up for themselves with their own specific tests.

    Particularly for a cheap camera like the D5200 if someone is serious they can borrow test/buy the fucking camera themselves and make up their own minds instead of bleating on the forums...

  • The horizontal pattern noise is definitely there but if you look at these images it's there @ 100 ISO + 2 stops but not there at 2500 ISO + 3 stops which supports my theory that either some switchable camera function is causing the pattern noise or particular ISOs are susceptible. These are jpegs from the web so I need to shoot some video and RAW stills myself before I can draw any conclusions.

    100ISO +3stops.jpg
    1671 x 1001 - 421K
    2500ISO +3stops.jpg
    1603 x 1076 - 513K
  • @GH2UW It is easy to see when a test is true, I know when I can trust a test. I got all my gear seeing web tests and no sorriness so far. Do not be afraid showing tests. Easy way is upload footage straight from camera so nobody can tell you did any mistake. But some edited footage can be trusted also, just watch Samuel Flaat profiles edited tests and you will see there is no way to distrust it.

    Maybe sometimes it can be some codec issues, light situation issues, and so on... but for me, it is no problem because I know how to deal with the limitations of the tests.

    I used a strong talking just to provoke, because I found to lift the shadows that way from a total chushed to black into a green foliage was impossible. You need to be an experimented person to find if a test is reliable or not.

    Sometimes someone have a good willing to do a test, but not enough knowledge, so some mistakes can happen. I am not judging people intention, just evaluating the results.

  • @GH2UW I agree with you, but being enthusiastic about this new toy I like to share my findings so that others may benefit from it.

  • When I used a strong talking I was doing some king of a "Devil's Advocate" behavior to provoke all people into the problem. It was an intentional "fuel to the fire" action.

    I hope you all do not dislike me due to that.

  • @apefos I suggest to look at one of the low light street scenes I have uploaded and play with those in your editing software. They are quite dark so you can do a good test to see how much detail can be pulled out of those dark areas. Just do some tests with gamma or brightness.

  • @Craftwerk I already have downloaded all your new unedited footage and none of them shows any problem for me. I do not see any banding problem.

    So due to this I was thinking the previous banding problem you found before these new unedited footage was some kind of mistake you did.

    All new footage from you looks great in my full hd monitor and denoise great.

    Edit post: when I say new footage I mean the unedited footage you uploaded.

  • @apefos These banding artifacts only becomes visible when you start adding brightness or change gamma values. It is also discussed here that the standard picture style seems best to avoid it. BTW, I have uploaded the "questionable" video footage with the grass on the left side of the road. I say, download that one too and see for your self the details that come alive when adding a bit of brightness!

  • @Craftwerk did you do a new upload or a replace in that video? Please send me the link, I will download and take a look.

  • I think we have a terminology issue here. Banding aka posterization as opposed to fixed pattern noise aka digital rain. I can't really judge banding from vimeo and youtube and an external recorder should deal with any banding issues.

  • @apefos

    BTW, here is a link to a Dutch site where they talk about the noise and dynamic range of the D5200. Very interesting, they have a few samples to show how much can be pulled out of the high lights, amazing!

  • Thanks for posting this Hans. This gives a good idea of low light noise and handling...

  • @mojo43 Hi mojo, remember that you had a discussion on a forum about aliasing problems with the D5100? Someone told you there how you can use a AA Avisynth script to get rid of these artifacts, and I used it on my own footage and it really made my videos super smooth. The script also worked as some sort of denoiser.

  • Try doing this with a BMCC

  • @squig Amazing! No aliasing present! And look how natural that color rendition is. Glad I stayed with Nikon and upgraded from D5100.

  • @Craftwerk Yep, I remember. Good script!