Originally posted by mrmentera is that the K-1 is on pair with Canon 5D MkIII, Canon 70-80D and Panasonic m4/3 when it comes to the eastethics of videofiles?
I would expect that at HD resolutions, the K-1 would be the equal of the 5D3, when the Canon is using it's normal h264 encoding.
Once the Canon is switched to 'All I-Frame' with it's higher datarate, the Canon would retain more of the image detail in the files, but,...
I haven't had chance to pull apart K-1 files, but I would expect Pentax to have pulled it's usual act of recording data outside the IRE limits for video signals, which means the K-1 will retain it's full dynamic range in video, where most DSLR's 'clip' the signals down to IRE levels, throwing away two or more stops of dynamic range in the process.
The 5D2 and 5D3 for example, are only 12 stops dynamic range in Stills (yeah, a full stop behind the K-01, K-30 etc, and even more behind the K-1), but with the extra detail lost in the video signal clipping, they're only 9 stops range in video.
(The comparison tool on DXOMark's website shows the K-1 is far superior to the 5D3, for every area of Still performance except buffer size)
This is one of the main reasons why Pentax's video performance is so disappointing in other areas - They keep the dynamic range for highlights and shadows, and then throw it away with the over-compression in the h.264.
The KP's reported higher Datarate, jumping from 19mbits to 24mbits, should help retain more of that information. It's still nowhere near high enough - for HD you really want 45 to 50mbits, and a minimum of 100mbits for 4K, preferably 200mbits.
For reference - 200mbits is around 25megabytes per second, well under what any decent UHS1 spec SD-XC card can write at.
The cards recommended for HD CineDNG Raw in the BMD Pocket Cine Camera, write at 95MB/s, nearly 4 times as fast as what is required for 200mbits.
Aesthetics,... comparing the K-1 against a GH4 or the new GH5 for the 'look' of the footage is a subjective matter, and not a quality matter.
FF video is somewhere between s35 and 70mm in 'look', whereas the GH4, and the BMD Pocket and Cine cameras, are closer to 16mm film in their look.
APS-c is pretty much the same look at s35, with very similar crop factor.
Originally posted by mrmentera But doesnt FF video compared to aps video gives the same big advantage when it comes to noise and iso as for stills? I mean, the a7s is used by pros in many situations. And the bigger sensor must give SOMETHING to the eastethic also, or???!
I've used the A7s for video, it's a shallower DoF look, and once in sLog and using an external recorder, the footage is very good.
Noise levels in the image, it's not the size of the sensor, but the size of the individual photosites on the sensor that affect it.
The A7s has massive photosites compared to every other FF camera - it's only 12megapixel, where other FF's are 24, 36, and 50.
Not only does the size of those photosites determine the light sensitivity, but when used in decent lighting, it leads to an improved Signal to Noise performance.
So the A7s will have a cleaner image, where other FF cameras have the same noise levels as APS-c cameras.
Iff I was spending my own money, and buying a new DSLR or Mirrorless camera for what I do, the A7s is top of my list, followed by the GH4. The reason for this is, that I can take advantage of 4K recorded externally to increase the quality of the HD delivery.
If I was going to stay with 'only' HD, those two choices would be joined by K-1 or KP, if I could record externally to ProRes or DNxHD files. That is currently not possible with the K-1, and unconfirmed with the KP.
Originally posted by mrmentera The c100 has a fan that cools the sensor f.ex. Is it a problem with K-1, that it looses colours when filming longer periods?
I wouldn't worry about heat when filming with Any Pentax,... I know what it took to make my K-01 overheat, and the K-1 is much newer tech, I'd expect it's resistance to heat to be far superior.
Camera's like the C100, BMD's Cine Cam and Ursa range, all have fans and large heatsinks because they use a Peltier Cooler to chill the sensor for lower noise in the video signals.
Those cameras also do not apply any noise correction to the recorded files, as it would be expected for it to be done in post-production.