IBC 2011: Testing the SONY VG-20 – recorded via hdmi out
SONY VG-20 – Interview with Kanta Yamamoto
Testing the SONY VG-20 – recorded via hdmi out
As you know by now the Sony VG-20 seems to be the only one of Sony’s newly announced large sensor cameras that has a clean hdmi output (no overlays or picture in picture).
Theoretically the advantage of a clean hdmi output is that we can record the feed directly to a harddisk recorder like the Atomos Ninja thus avoiding the bad internal AVCHD compression and giving us better quality and color correction possibilities in post.
At the IBC 2011 exhibition we got the chance to test the hdmi output of the VG-20 and you can see the results in the video above. On the left hand side are very low compression jpeg screenshots of the video to give you a better idea how the original footage really looks like as it is compressed to 8mbit and more on YouTube.
This is not a scientific test, just a quick grab of some footage to get a first idea of what we could expect from this camera. And to be honest, our shitty recordings are enough to make up my mind: This camera fails for me!
Unfortunately what we got out of the hdmi port of the VG-20 was only marginally better looking than the VG-20 internal AVCHD recording. If you can see what I see you might agree that the hdmi signal looks like it has undergone very much processing. There wasn’t much light at the Sony booth so the cameras are shooting very low light, but to me it looks like there is a lot of compression noise, even on the Atomos footage. Maybe the image is compressed before it gets sent out to hdmi. What do you guys think? Tell me in the comments.
I have included a still of the HX9v+Atomos grab and the noise looiks totally different there, more like real sensor noise (small dots) as opposed to the large blocky noise I can see in the vg20+Atomos footage here.
Maybe I’m doing the camera some injustice as the lighting conditions were really bad. Unfortunately the camera was chained to the table.
Despite the 7D being out of focus at one point, overall it looks less sharp than the VG-20. However the sharpness in the VG-20 footage looks a lot like it has gone through a heavy sharpening filter.
And our friends at eosHD were right: There’s also a lot of aliasing and some moiré in the footage, much more than there should be, to an extend that I’d suggest not to buy this camera, any Canon DSLR will do as good and the HX9v will do better (albeit not being a true video alternative).
How can a video camera have aliasing like this? Scary.
If you still want to buy this camera here’s a link to B&H: