[neostyles]

So the more i shoot with the d600, the more im convinced that it's video doesnt make the same great low light that ive come to love for stillls.. So what is the reason for this? Is it the line skipping? This would explain why, if presumably its not using all of the sensor and not gathering the full amount of light that a ff sensor can. Thoughts?

[Steve.Ledger]

I'd say it has more to do with exposure of each frame to light per sec of the chosen frame rate (eg: 24fps = 5/2 of a second per frame (or 2.5 of a second)).

[Giklab]

I'd say it has more to do with exposure of each frame to light per sec of the chosen frame rate (eg: 24fps = 5/2 of a second per frame (or 2.5 of a second)).

I think that might be 1/25 of a second, at least...

[Fogel70]

So the more i shoot with the d600, the more im convinced that it's video doesnt make the same great low light that ive come to love for stillls.. So what is the reason for this? Is it the line skipping? This would explain why, if presumably its not using all of the sensor and not gathering the full amount of light that a ff sensor can. Thoughts?

I would say that line skipping is a big factor in low light video IQ.
But first of all video is usually recorded in 16:9 format which means that a large part of the sensor is cropped when recording video, then line skipping is cropping away even more of the sensor used, so in the end a FF DSLR might not use more sensor area for video than a P&S with much smaller sensor.

But some large sensor camera don't use line skipping for video which give them a huge advantage when it comes to low light video recording. FI Sony A7S, and I expect A7r II tyo be much improved low light video IQ over older A7r. Improved low light video will most likely be standard on future generation large sensors.

---

Last edited by Fogel70; 07-07-2015 at 12:11 AM.

[neostyles]

How much of the sensor would you estimate is beng lost?

[Fogel70]

I have no figures for this, but I would guess it's at least 75% lost (2+ stops)

[fuent104]

I think that might be 1/25 of a second, at least...

The standard shutter speed 1/48 second at 24fps.

[Cannikin]

Let's start out by simplifying things and assuming the D600 has exactly 24 megapixels and a resolution of exactly 6000x4000 (it actually has slightly more than that).

Case 1) If it were pulling exactly 1920x1080 pixels from that, it would result in a (1920*1080)/(6000*4000) = 8.64% area, a loss of about 3.5 stops. That's the worst case scenario.

Case 2) Assuming there is no line skipping, and the full 16:9 area is used, a straight crop from 3:2 to 16:9 gives a height that is (3*9)/(16*2) = 84.375% that of the original height (3375 pixels), and by extension area, assuming no width reduction. That's a 1/4 stop loss. That's the best case scenario.

Both case 1 and 2 are unlikely.

Case 3) Perhaps Nikon does something more along the lines of using the full width, but line skipping to 1080 vertically, in which case the effective pixel coverage would be 1080/4000 = 27% area, or a loss of a little less than 2 stops.

That's about as much as one can guess without knowing the exact video implementation Nikon uses.

[Giklab]

The standard shutter speed 1/48 second at 24fps.

That's true (1/50), but the camera can obviously go lower. Even lower than 1/25.

[neostyles]

So im looking at picking up a video camera from panasonic or sony. Do they get more area out the sensors? I Know some of them (the rx10ii and the recently released a7rii dont skip), so would this mean that we are getting full sensor area or atleast much more? So if im understanding this correctly the line skipping is more an issue than the video resolution?