Carl “We have an up-to-date, ready to go model, and first things first; we head out to inspect the image quality. Firstly trying out the various Gamma profiles. EVA has 5 different profiles that they call “Scene files” 1-5 and then of course V-Log.
After a quick check for rolling shutter – Barely noticeable, but there in small amounts at high speeds – we test out the Image Stabilisation. EVA has a 5.7K sensor which it crops into a 4K image, using the extra space to stabilise the shot. Whilst only making a slight difference during walking shots, the handheld stationary shots proved much more noticeable.
200FPS made for some impressively smooth slow-motion shots, if a little noisy. 100fps was a lot cleaner.”
“You can see for yourself how the picture compares, and – more than ever – we’d love to hear your reaction, if you already own one of these other cameras, how does EVA compare?”
“After … – we test out the Image Stabilisation. EVA has a 5.7K sensor which it crops into a 4K image, using the extra space to stabilise the shot.”
Errr, what!!? I sincerely hope it does not “crop into 4K mode”? The whole promise of the 5.7K sensor is that after debayering to 4K it will give full 4K luminance resolution – that’s the whole point of the extra photosites!
In general, all chips will have extra photosites around the edge to provide for various factors – including stabilisation. The 5.7k aspect is surely totally irrelevant here?
And is it just me, but on the “test” shown here, the rolling shutter didn’t seem very good? It really underlines how if you’re going to test a camera, it really needs to be done scientifically to draw any sensible conclusion – not just wave it past trees. (And rolling shutter is probably one of the easiest things to properly test quantitatively – have a vertical line, pan so it moves from one edge of frame to the other, and measure the angle off the vertical, and the number of frames it takes to travel from left to right. From that it’s a relatively simple calculation to get a figure in mS.)