I don't know for certain if it's the same transport mechanism across all those camcorders, but it wouldn't surprise me at all if it is.

Dear Chris,

Thank you for the update and all the work that you do in order to bring us the latest information.

Dan

Less We Not Get Confused 3 chips are generally better than 1!
 

One thing we don't mean to confuse the reader with is the thought that a one chip 2/3" CCD camera with an RGB filter will have more light gathering power than a 3 chip 2/3" CCD camera. I felt we need to clarify somehow that more chips are better generally better than single chip camera. This assumes the 14-bit DSP or whatever the processor component it's using, is approximately equivalent in both cameras. I was almost getting the impression that we were implying 1 chip is better than 3, that would not generally be true.

Well Douglas, just to stir things up a bit and to play the devil's advocate, I will go on record and say that I can demonstrate how a one-chip camcorder *can* outperform a three-chip camcorder. But it's got to be a newer one-chip vs. an older three-chip, because as I pointed out above, the DSP technology has a lot to do with it and the newer the processor, the better. Single-chip with RGB gets into the ballpark with image quality that isn't far from the three-chip. Newer, superior processing technology will then give the advantage to the single-chip camcorder.

Now I don't know of any single-chip 2/3rd-inch camcorders that I can get my hands on, but the example I used above should work very well: take an older 1/3rd-inch 1998 Canon GL1 and compare its performance head to head with either the 2003 Optura Xi or the 2005 Optura 600 or any Optura model between that spread.

The single-chip Opturas have RGB color filters, megapixel CCDs slightly larger than 1/3rd-inch, Digic DV processors, and either Optical I.S. or the new version of Electronic I.S. which rivals OIS quality. The 1998 GL1 was great for its time but really isn't superior to the newer technologies found in the post-2003 Optura line.

So yes, I'll stand by that claim and I urge you to check it out for yourself sometime. The results might surprise you. It's more a case of newer technology beating old than it is three chips vs. one anymore. I'm willing to bet that the demise of the three-chip camcorder is not too far off. We haven't seen the last three-chip camcorders yet but there are fewer of them now than there have been. At the same time image quality is better than it ever has been before.

Allow me to again point to digital still cameras: the best ones are all single-chip with RGB, not three-chip. Sooner or later, video camcorders will be the same way too.

This is what I've been waiting for to start shooting HDV big wall climbing videos - woo hoo!! I'm glad I didn't buy one of the Sony HDV cams - this looks like a high def version of my Optura 100MC's !!! ;^)

Doug, I'm a little confused about how Chris' comments on the tape transport have anything to do with the sensor?

But anyway, about the sensor. The HV10 will have a single CMOS sensor, not a CCD, so that changes the equation a bit. If everything else were to be equal (which in reality it never is), splitting the same amount of light amongst 3 sensors will reduce low light performance compared to an otherwise identical 1 chip camera. Three chip cameras generally do well because all else is NOT equal, such as absolute aperture, sensor photosite size and native sensitivity, etc, etc. So there probably isn't any real way we can know just how the low light performance of the HV10 with 1 CMOS will be vs its "big brother" 3-CCD cameras until we get our hands on them.

Another thing we'll have to wait and see about is how well the camera's footage can be matched to the images from other cameras. It would rock if little or no correction would need to be done to credibly B-roll this camera with an A1 or H1. Hoping! But with a CMOS sensor (color/gamma matching issues possible) and a much smaller lens (lowered image sharpness possible, despite the 1920x1080 sensor area), it won't surprise me too much if the camera doesn't cut easily with the "big brothers."

If low light performance turns out to be superb and the camera cuts well with an H1, that'll be a huge bonus. I think it is a bit beyond reasonable expectations for an inexpensive pocket camera, but we can hope. ;-) Be that as it may, this is going to be an amazing low-cost pocket camera that as a bonus acts as a deck for professional cameras...I'm thinking I'd better reserve mine now because these little hi tech wonders are going to SELL!

I agree with both Pete Bauer and Chris Hurd concerning the picture quality of this camcorder. Camcorder technology has come a long way especially considering the fact that the picture quality of the HC1/HC3 is nearly identical to the FX1/Z1u in good lighting. I expect the picture quality of the HV10 to definitely be close to the XH-G1/XH-A1/XL-H1 camcorders.

Call me completely insane but it would not surprise me one bit, if the picture quality of the HV10 ends up being a little bit better than the Z1u/FX1 during outside day shooting.

Actually the opposite is true. The prism in a 3CCD camera puts all the green light of the incoming light to the green CCD, all the red light to the red CCD and so on. So all the red, green and blue light will hit a CCD pixel.

On the other hand in a 1 CCD camera every pixel has a color filter in front of it which blocks about 66% of the incoming light. For example the red pixels will get only the red light, the green and blue will be absorbed in the filter.

So if you have a 1 CCD and a 3CCD camera and the 1CCD camera gets three times the amount of light the 1 CCD camera gets, they will have the same amount of light arriving to the CCD photosites.


This 1 CCD Canon CMOS camera is intresting for me because Canon manufactures the most light sensitive imaging chips for still cameras. And they are CMOSes. If they can get down that technology to this camera they may produce a 1 chip camera that is more sensitive than other 3 chip cameras with less light sensitive chips.

Thanks for the very nice clarification, Balazs. My (poorly stated) point was that this is a CMOS camera. The sensor block technology employed is going to be quite different from the guts of a 3-chip CCD block, so considering only the number of sensor chips in guessing the low light performance is not a useful exercise. Other aspects of design also matter tremendously. All we can do is wait for the camera and see how it performs.

1. if the HV10 is going to perform fairly close to its more expensive bros. in the XH series, why spend more $ when the lower price can achieve similar results?

2. hard to discern from the pictures, but does the HV10 accept filters/lenses? i'm starting to imagine people adding wider lens onto the HV10 =). ~$1,000 buys 1920x1080 AND very wide capture... this is pretty cool =).

I do not pretend to be an expert in this area, but both systems, 1 chip and 3 chip cameras have to have filters.

In my opinion, the 1 chip cameras have the filters in front of the sensor for each pixel, while the 3 chip cameras have a beam spliting prism and filters for each color.

One word: control. It all boils down to the basic differences between the consumer and professional realms (and it's the exact same situation that has always governed standard definition DV). There are no audio recording management capabilities on the HV10. There are no custom presets, no way to dial in a specific f/stop, no way to smoothly adjust zoom, focus or exposure during shooting. You're going to show up for a paying shoot with this little thing? It will have its limited and specialized applications for certain instances in professional shooting, such as sneaking shots, crash cam, helmet cam, etc. but that's only in situations where nothing else will do.

Yes it is a 37mm threaded filter diameter. Any 37mm attachment will go on there but it had better resolve for HD.

In 1 chip cameras the filters in front of the pixels simply absorb the filtered out colors. In 3 chip cameras the filters are much more sophisticated. Some of the surfaces of the prisms in the prism block are coated by a large number of layers with different optical properties. These surfaces reflect back part of the incoming light and let through the rest depending on the wavelength. So when the light comes into the prism it is going through a surface. But the blue portion of the light is reflected back in the direction of the blue CCD. Then the light goes through another surface where only the red portion is reflected back and is going to the red CCD. So here the advantage is that the filtered out light is not lost unlike in the case of the 1 chip camera filters.

Dear Balazs,

Thank you for the explanation.

I found a reference to how this works on wikipedia.org:

http://en.wikipedia.org/wiki/Dichroic_prism

Dear Dan,

Great link, thank you!