first hv20 review ( in japanese)

[Kurth Bousman]

http://www.watch.impress.co.jp/av/do...7/zooma294.htm

[Ross Hunter]

Kurth,

Thanks for posting the link. I used Google to translate the page and it is an "interesting" read. Some of the translation is awkward at best. I especially liked this line as translated by Google: "Last year for the consumer CANON “HV10” which carries out entry with HDV shook the heart of many consumers in the picture quality."

Ross Hunter
Orange, VA

[Holly Rognan]

That was one of the funniest articles I have read in a while. I couldn't find anything in the article that explicitly said if it was really improved in low light. Can anyone shed some light on this area? (I dont really like puns, but this was begging to be said.)

[Chris Hurd]

Don't laugh too hard at the translation -- imagine how they feel when reading some English web pages translated to Japanese. It's probably just as amusing (and somewhat confounding) for them as it is for us.

The statement "thorough(ly) improves the weak point of HV10" is self-explanatory, in my opinion.

[Wes Vasher]

Right, it's the translation engine that is so funny. Here's my favorite line...

“the monster”!? ~

[Holly Rognan]

Originally Posted by Chris Hurd

The statement "thorough(ly) improves the weak point of HV10" is self-explanatory, in my opinion.

I read that, but is the performance improved or just the "lux rating". I have seen many manufactures claim a better lux rating, but performance in the real world is another thing. I would be happy to get a 3 Lux (at a 1/30 shutter) and I just hope that it delivers.

Also, does this have a 1/48th shutter and a 1/24th shutter speed?

And wouldn't a 1/30th shutter provide true 30p (because the sensor is progressive)?

[Paulo Teixeira]

Originally Posted by Wes Vasher

Right, it's the translation engine that is so funny. Here's my favorite line...

“the monster”!? ~

http://www.engadget.com/2007/02/07/c...d-the-monster/

[Paulo Teixeira]

Originally Posted by Holly Rognan

Also, does this have a 1/48th shutter and a 1/24th shutter speed?

And wouldn't a 1/30th shutter provide true 30p (because the sensor is progressive)?

The answer for your first question can be found here.
http://dvinfo.net/conf/showthread.php?t=85905
The second question is no, although I wish it had a 30p mode.

[Lloyd Coleman]

Holly's question about 30p got me wondering. What does happen when the shutter speed drops below the frame rate. At 1/30 is the shutter open while two fields are recorded? How about at 1/6, how does a shutter speed that is 5 times as long as the frame rate work?

[Holly Rognan]

Originally Posted by Paulo Teixeira

The second question is no, although I wish it had a 30p mode.

I know it doesn't have 30p, but wouldn't a 30th shutter provide 30p? It is a progressive sensor, so wouldn't it work, or would it drop fields?

[Pieter Jongerius]

Holly, Lloyd, maybe it's really the same question... what is the relation between shutter speed and fields-or-frames/sec. Of course, the rate at which the CMOS is read and reset can be seen as beeing separate from rate at which the shutter opens and closes.

But I'm guessing this: shutter times can not be larger than sample time (ie seconds per frame or field). Otherwise, you would have some freaky effects of interference, with some frames brighter than others, because of the shutter beeing off-phase totally.

But then again.. it's a guess :)

----------
Edit: oh sorry that's silly of me, that's against specs:
http://www.dvinfo.net/conf/showthread.php?t=85905

Lloyd, the shutter must really stay open then... and produce motion blurs that are clipped by the frame cycle. I'm at least glad they kept the fractions clean.
To Holly: the frame rate that is recorded is determined by the read/clear cycle of the CMOS sensor, not the shutter.

[Michael Wisniewski]

Originally Posted by Lloyd Coleman

What happens when the shutter speed drops below the frame rate.

Each frame is exposed several times during the scan. That's why you start to get ghosting effects when you go below 1/30 or 1/15, the chip is getting multiple exposures while it's scanning for data.

[Holly Rognan]

Originally Posted by Pieter Jongerius

To Holly: the frame rate that is recorded is determined by the read/clear cycle of the CMOS sensor, not the shutter.

And the frame rate is also dependent on the codec, among other things, but that isn't my point. I understand you can shoot in 24p at a 120th shutter, or 48th shutter you still have 24p, albeit with differences in motion rendering. I guess you can do a psuedo 30p with a 30th shutter, I was just wondering if there was a way to trick the cam into getting a progressive signal if the shutter is below 1/60.

One thing I still can't figure out is if the CMOS sensor looses a stop of light and gains 6db noise during progressive recordings like the CCD cousins, because if so the lux rating that is being touted will just be a gimmick in marketing and only truley useful in 60i, although a 1/24 shutter for 24p would make up some of this disrepancy.

[Thomas Smet]

I'm not sure about a shutter of 30. Most cameras that use interlaced chips do usually loose a field of data but the same isn't always true for a progressive chip. For example the JVC HD100 cameras can use a shutter of 24 or 30 and they still sample the full 1280x720 pixels. So what you end up with is just a progressive frame but with a lot of motion blur that no longer looks natural. Since the HV20 does use a progressive chip it makes sense that a lower shutter speed might still use the full pixel raster. With a 1/30 shutter however you will always have that extra motion blur look which can look a little funky to some people. I guess you would have to try it and see if you are ok with the motion blur.