Which to buy for low light? Sony HDR-HC3/5/7 or Canon HV20

[2Bdecided]

I was waiting for an HD camcorder to come along with good low light performance at 50i (i.e. not 24p/25p because I hate the filmic look) but unfortunately my trusty Sony SD miniDV camcorder has died, and my kids birthdays are coming up, so I'm looking for an HD replacement very quickly.

Some reviews suggest the HC7 is poorer in low light than the HC3 due to higher pixel count in similar space - is this true? Where does the HC5 fit in? It appears to have a similar sensor to the HC3, but I can't find any definitive reviews.

Most HV20 reviews concentrate on the 24fps mode. I won't use this. So, honestly, how does the HV20 compare with the Sony models in low light? I film a lot in doors (especially in the winter) – daytime and evening. I don't expect brilliant results, but my existing miniDV camcorder has always been poor in less than bright sunlight, and horrible with subdued (normal!) lighting indoors at night.

The other issues are

* ease of use
* easy manual override for things that may go wrong (e.g. focus, exposure) (I believe the HC5 is worse than the HC3 because the manual button near the lens has been removed - is this right?)
* a decent image stabiliser
* playback of my old SD miniDV tapes recorded in SP and LP on a 9 year old Sony machine

I don't think the rolling shutter of the HV20 is going to be an issue.

Any advice?

Thanks in anticipation,
David.

[wonkey_monkey]

Most HV20 reviews concentrate on the 24fps mode. I won't use this.

I don't blame you, it looks terrible. It uses the full 25th of a second, resulting in unnaturally blurry movement (see the Channel 4 comedy Spaced or the first few episode of Torchwood, which seem to suffer from this problem).

I've had a quick go with one, and all I can tell you about it's low light performance was that it didn't seem much better than a Sony PD170 I've used (semipro camcorder, a few years old), but it was in no way an extensive test. Not great, but not bad.

One other thing is that the HV20 must use some weird shutter system, because there is some crazy warping of the picture when you move the camera, like someone's tugging at the corners. It's not very noticeable unless you step through the picture, but it's enough to annoy you if you're looking for it. But perhaps all HD camcorders are like this - I've only tried the HV20.

ETA: I really should read posts more carefully, since you already mentioned the shutter.

David

PS I remembered this morning that I still have the HV20. The only other camera I have for comparison is a Canon MV330i, so if it's any use to you, the low light performance is about the same.

[2Bdecided]

Thanks for the response David. I hate those first few episodes of Torchwood (visually).

Maybe I would be better with an SD camcorder like the DCR-HC96?

What I'm trying to figure out is, for a given sensor size, does increasing the pixel count increase the noise, or does it just make it more visible? e.g. if I took the output from a 3" HD sensor, and scaled the results to SD, would it look more or less noisy than the output directly from a 3" SD sensor?

Cheers,
David.

[wonkey_monkey]

Maybe I would be better with an SD camcorder like the DCR-HC96?

While the PD170 is highly rated as a near-broadcast quality camera, the scaled down HD pictures from the HV20 blow it completely away for sharpness (although I'm not entirely convinced that the PD170 isn't a bit out of whack).

In theory, scaling any picture down will reduce the perceived noise, since you're averaging neighbouring pixels - but comparing SD to HD is a little bit apples & oranges since the resolution is so different. I haven't tried comparing scaled-down HD with SD from the same camera, so I don't know what difference the DV codec makes.

David