Jump to content

paulraymaekers

Members
  • Posts

    65
  • Joined

  • Last visited

Posts posted by paulraymaekers

  1. If you research back throughout this thread, you'll find the answer you're looking for.

    4:2:2 means higher bitrate too and all this means a obvious benefit.

    and can anyone confirm that the A7S outputs 1080P 50frames/sec over the HDMI?  (in crop mode for sharpness)

  2. LOL

    I'm sorry to disappoint you when I can't neglect the benefit of higher bitrate 4:2:2 even if 8-bit...

    so then the question comes...  how big would the benefit be to start recording externally on my A7S  (just for 1080P) ?   to go from 8 bit 420 XAVC to 8 bit 422 (proress probably)

  3. Oh my, I never saw that one. There is a little flaw in his test that probably explains the huge difference between the demonstrated quality of 8bit 4:2:0 vs 8bit 4:2:2: he's recording externally to the Shogun with the 8bit 4:2:2 and internally to the GH4 for the 4:2:0. I'm gonna bet a dollar that's why they look so different. Maybe it's a data rate thing, maybe its smoothing happening with the Shogun, Maybe it's the GH4's implementation of 8bit 4:2:2, I don't know. Below is just a radial saved as 8bit dpx, then two conversions to respective chroma subsampling schemes. One is yuv420p 8bit and the other is yuv422p 8bit,  Same very high h264 constant bitrate settings for both. Re-imported, gamma curve applied and saved out as a sRGB png.

    radial_420test.png

     

     

    please show the same test in colour? :-)

  4. His thread title/post....... He's worried about losing sharpness with these lenses on Sony A7s, my point being he already is losing sharpness as those lens are not sharp to begin with, especially shooting at T1.5 which those lenses should not even be rated as, as you really cannot use them until T4 and above, specially if he is using the 24mm lens which is their worst lens in the series.

    I don't think this is general for Samyang/rokinon: I have always had the impression that the lens outresolves the sensor on the A7S. I use an 85mm​ T1.5, and when shooting stills with it on A7S, they are sharp up to pixel level

    so for HD film it just works perfect?

  5.  

    have been using metabones nikon to E mount on A7S, both with and without aperture control
    both work good

    used it in FF with sigma 12-24,  nikon 24-85 ED, sigma 50 2.8 macro, samyang 80 1.4, nikon 80-400, nikon 200 f4 micro  all fine

    and in APS-C mode, with nikon 10-24 and nikon18-200  just works

  6. hi all, sorry if this was already asked, but I cannot find it in the archives.

    Full frame, at 50p, 1080, the A7S is less sharp when recording internally, compared to 25P

    now if I would record 50P 1080 externally, do I get better resolution? or does it not matter?

    paul

     

  7. try rolling past your 100% and black a couple of stops while recording, then pull those into your coloring suite of choice and see what the results are. my guess is %100 is clipping a bit earlier than you're actually losing all detail. from what i've seen you should have no problem getting to at least 12 stops in slog2

    ​so did the test again, both in PP0 (contrast -3) at 800 iso, and SLOG2 at 3200

    starting 2 stops over zebra clipping down to 2 stops after pitch black

    put it into resolve to see the waveform

    PP0 I got like 9 1/3 stop  from 1024 down to zero noise

    sLog2 got like max 11 stops from 1024 down to zero

    what I notice in PP0  I go from 1024 to 512 in 3 stops   so very small  'bandwide' in the highlights

     

    (not saying I'm not happy with my A7S, but till now I find only very small advantage between PP0 contrast -3  and Slog2: for everything where I don't need the absolute highest DR, I'm using PP0) 

    (if AS7 would have possibility of contrast -6 or lower in PP0, probably there would be no reason to use SLOG2)

     

  8. hello,

    can someone tell me what is wrong in the next test with my A7S

    I set zebra's to 100

    PP0   ISO 800

    I start shutter at 1/25 and aperture 2.8, I illuminate a white paper so I just get zebra's, this is my starting point

    by increasing shutter and closing aperture, I click down till the white paper is pitch black so I don't see anything anymore..

    from just zebra's to pitch black, I just get 8 1/3  stop  when contrast is -3    (contrast 0, I get 7 1/3 stop)

    same test for Slog2, gamut, but ISO 3200

    10 or 10 1/3 stops from zebra's to pitch black...

     

    where is the test wrong?

    paul

     

  9. Yes, I do, the moire was subtle btw and the picture looks incredibly sharp otherwise. 

     

     

    if it were only for sharpness, all we would need is the GH4 having an EXT TELE CONV of  x1.2  and internally recording 1080P

     

    this would use the native pixels of the 3840*2160, and downscale it to FHD, like it does now on the HDMI output..

     

    a 'reasonable' upgrade request...?

  10. Yes it's the Ninja that converts the 4K to HD. That's what you are seeing in the Dropbox video above. This is why I can't see any benefit to buying a Shogun if you're going to be downsampling in post anyway. As I understand it, the GH4 gives full sensor readout when using 4K whatever. If this really is the case it actually makes more sense to record to 1080 prores. In which case a standard Ninja is probably better - saving yourself money, time, hassle and space on your hotshoe  ;)

    ok, this even makes it nicer. I assume it's not the full sensor coming out of the HDMI, but the center 3840*2160 pixels, same as what is recorded when filming UHD. and the ninja just bins one pixel every 2 and one line every 2 lines  to convert it to HD...

     

    so the workflow become so easy: film in UHD (ok, with the 1.2 crop from the full sensor), record 1080p 10bit 422, and you have perfect footage to work on.... 

  11. Just to clarify what I said above, the footage was shot at 4K on the GH4 but recorded to the Ninja as 1080p. I have no idea if you will get superior results recording 10-bit/422 4K to a Shogun and then downsampling in post. I don't see why you would really.

     

    and this brings us to another question again :-)  (as we seem to ask good questions :-))

    the HDMI out 1080p, is it compressed in a different way from the full sensor to 1080, compared to the internally recorded 1080?

     

    I know it is 10 bit 4.2.2, but the downsampling in the camera from full sensor to 1080 is the same algorithm for internal recording and external to HDMI, so it would have the same sharpness, and not sharper as some might hope?

     

    OR

     

    you write: shot at 4K but recorded to Ninja as 1080p. Does the Ninja convert the 4k HDMI signal to 1080p?

×
×
  • Create New...