Initial thoughts from a HDR-HC1 Switcher

[Dave Lammey]

Originally Posted by Mark Fry

I'd second that, since it is a little smaller than the XH-A1. I'd add that, since it uses CMOS chips, it should give better lattitude (more details shown in darkest and lightest parts of the pictures) than the Canon's CCDs, which could be very useful in low-light situations. This is one area where I really want to see some side-by-side test results...

Ditto... though it doesn't look like anyone's buying the FX7, so we'll have to wait until the V1 comes out ... hopefully someone will buy that!

[Thomas Lundberg]

OK, I'm fully convinced the A1 performs far better than the HC1 in low light.

Alex, you report a patterning of out-of-focus objects in the HC1. I don't think this is due to a "sharper lens than sensor"-effect. In high-frequency parts (fine texture, like in a tweed jacket) of the image, "aliasing" will appear if the sensor resolution is lower than the optical resolution. However, this applies to in-focus objects and the aliasing will vanish as the object moves out of focus. Maybe you see an effect of the compression algorithm(?). In any case, your observation is very much appreciated!

As for the CMOS vs CCD comparison, which perhaps is sliding away a bit from the original topic, I am all ears/eyes for any improvements of the dynamic range. Moving from good-old film into video has been painful regarding this.

[Alex Leith]

I've attached an image. You'll need to open it to full size to see the noise.

If you look at the trees in the background you see a kind of patina over the out-of-focus parts of the image. It looks almost like it was shot through a bug screen (it wasn't - that's clear air between the camera and the background).

I remember reading somewhere what caused this - and I seem to recall the suggestion that the lens was too good for the camera (the pattern you are seeing is something to do with the sensor array)... but I may be recalling that wrong.

Anyway, however it's created, it's there.

[Thomas Lundberg]

Thanks, Alex, for that illustration. Very clear distortion of the background, indeed. It is very similar to image processing induced artefacts I've seen in highly compressed airborne surveillance images. I would place a little bet on the compression after all.

However, what's interesting is, as you say, it's there! So this is a typical HVR-A1/HC1 behaviour that is not seen with the Canon XH A1? Then Canon scores one more point with me...

[Tom Roper]

Originally Posted by Alex Leith

I've used the Sony HVR-A1 / HC1. I love the size, and the images are generally excellent - but there were two problems we noticed:

1. Low light performance is rubbish, and the image quality rapidly degrades and colour balance dramatically shifts when things get a bit dull.

2. Out of focus objects display a sort of moire patterning. Apparently this is something to do with the lens resolving more than sensor can cope with?

The screen door effect is the bayer filter. My Canon HV10 does this. My JVC GR-HD1 did this. It's what you avoid when you get 3CCD/3CMOS as opposed to the one chippers. The 3 chippers don't need this filter, the single chippers do to produce color.

You'll avoid this moire pattern with any 3 chip camcorder, Sony, Canon, Panasonic, JVC et al