You've seen commercials where the bourbon oozes out of the bottle
and splashes so sensuously on the rocks?

High speed HD can also be used for shooting scientific things like bullets
hitting melons and water balloons dropping on people's heads. Cool stuff. There is something to me
that is intriguing about slow motion and time lapse video. The camera
can be so revealing in what it can see compared to the live human eye.

Siiiigh.

That's just not true.

http://en.wikipedia.org/wiki/Panasonic_AG-HVX200

I'm not saying that you should trust everything on the net, but according to wikipedia the resolution of the image sensors is 960x540. That said, I'm sure the camera is capable of caoturing some fantastic images, just not in 1920x1080 resolution.

the new sony 2/3" xdcam as well as the new Pansonic HPX will be 1920x1080 front to back, meaning that the image sensors are putting out 1920x1080 and processing that through to a format that records 1920x1080.

The HVX -- any DVCPro100 format -- records 1440x1080 with a PAR of 1.2 to fill a 1920x1080 frame. I understood the same to be true of XDCAM, which I though used an MPEG2 profile of 1440x1080 ... but I'd be happy to read a reference that refuted this as concerns XDCAM. It is irrefutable with DVCPro100, which has no ability to record 1920x1080 as a native resolution.

Note that this is not determined by the imaging chips, which may be capable of resolutions of 1920x1080 (or more likely actually quite a bit less) -- the recorded resolution is what I'm talking about here.

Which was rather my point ...

GB

The CCDs are 960x540.

But CCDs are analog, monochrome devices, not digital. A "pixel" on the CCD does not correspond to a "pixel" in the digital image. That has to do with internal processing of the light. What matters is the final image, and the HVX's 1080 resolution is on par with, or better than, any of the 1/3" cameras.

It should be noted that ReelStreem gets 2K uncompressed off the HVX's chips using their Hydra modification to the HVX, so the HVX, in fact, in recording in 1080 doesn't use the full resolution potential of the chips.

That said, with a half-pixel offset of the red and blue sensors (which is a long-standing industry practice not limited to the HVX200 or Panasonic by any means), you get an "effective" resolution of 1920x1080 from the sensors, anyway.

As someone else pointed out, if CCD pixel count were everything, then a simple 1-chip HV20 should give you a better picture than the HVX200, right? Hardly.

The HPX500, by the way, doesn't have a 1920x1020 chipset, either. Neither does the 2000. The 3000 does.

The wikipedia article states that "The green CCD in the array is physically shifted 1/2 pixel biaxially to achieve up to 50% higher horizontal and vertical resolution. This would make the theoretical maximum resolution of the image, 1440x810 pixels, even though each CCD has only 960x540 photosites.", and that is backed up by Panasonics own paper on the subject - see http://toshpit.blogs.com/the_toshpit...o_for_you.html . (And scroll down to the second table where Panasonic also quote max theoretical resolution - they themselves quote 1440x810 max.)

Note that Panasonic do emphasise that "Certainly this is a theoretical “best case scenario”.....", and for a decent mtf % it's more likely to be around the 1280x720 mark. But as you say, what matters is the final image, and that is in reality around what most of the other 1/3" cameras produce. The point Panasonic make about mtf over the resolved range, versus absolute max resolution is also very valid.

The articles also make clear that the internal processing of the camera is done at 1920x1080 (to make best use of the pixel shift) for very good reasons. That does not mean the chips are at all capable of that resolution, even in the limit. But then neither is any 1/3" camera, and few bigger chip models, let alone codecs capable of recording such a signal.

That's why I put the quotation marks around "effective."

But again, a pixel on the CCD does NOT equal a pixel in the image. They are pretty much unrelated. Like I said, they're getting 2K uncompressed (at 4:4:4) off those chips in the Hydra.

Not quite: in terms of measured TVL resolution the Canon XL-H1 trumped it by a slight margin:

http://www.adamwilt.com/HD/4cams-part2.html

But that's not particularly significant, the real question should be what are you trying to do rather than who records what. A recorded resolution of 1920x1080 pixels means nothing if it's coming from a cell phone with a pinhole plastic lens, so resolution isn't the only thing to consider if you're comparing "digital cinema" cameras.

I agree with the first sentence, and in the HVX (like many cameras) there are three things to consider - CCD pixels, "image pixels" (used for internal processing), and recorded pixels.

The HVX processes at 1920x1080 internally before downsampling for recording. Getting directly at the output of that processor will definately give you a 1920x1080 raster to record - but it can never give you any detail beyond about 1440x810, as Panasonics tech paper makes clear: "Spatial Offset technology will improve the resolution by a factor of 1.5 provided the gain in resolution is not offset by the quality of the lens."

The 1440x810 refers to the pixel count, the "resolution," of the CCD, not the processed image. As you agree, they're not the same thing.

In any case, you might want to check with ReelStreem, because they're pulling 2K off the CCDs -- that's BEFORE processing. If what you're saying were the case in the sense you're saying it, that would be impossible.

And it's worth noting that in real-world tests, the HVX200 yields measurable resolution about what you would expect from a good 960x540 sensor without any fancy processing. Same concept probably applies to the Sony V1U too, but I haven't seen definitive tests of that yet. Bottom line is that the native pixel count does matter, because it's difficult to create detail which isn't available at the sensor level.

I thought this was a discussion of cameras capable of 1920x1080 res, no more no less. All I was trying to get at is that, technically speaking, the HVX doesn't fit into this category. That isn't to say that the HVX is a bad camera, I have no idea to be honest. But from what I have read and the test charts that I have seen, the HVX does not fit into this category.

It's a good point that the JVC can beat an HVX in chart tests even if it can't in tech specs.

What's truly fascinating, is with solid state recording, full raster 1/2" or 2/3" chips and the new 50mbps xdcam and 100mbps AVCintra codecs, the under $10,000 camera market will all be 1920x1080.

Greg, I don't know exactly what the intent of this thread is -- but I too thought it was looking for those that did 'true' 1920x1080 ... and my point was that would exclude all DVCPro100 devices and I think all XDCAM HD devices ...

But if the question is just 'what devices offer 1920x1080 output' ... the answer is everything that claims to be 'high definition', regardless of what it actually can deliver, does deliver, or dreams of delivering.

GB

Geoff,

I'm definitely not trying to argue with you here. I agree, most of the camera's listed in this discussion don't offer true 1920x1080 res from pick-up to output. True 1920x1080 was once reserved for studio use and we are now begining to see full raster hd codecs. I could be wrong here, but since the second episode of Star Wars was shot with a F900 using HDcam, then it was captured at 1440x1080, and not 1920x1080. It's really exciting to see where technology is going. The number of digital cine cams out there now is very exciting.

if you want a list of HD cameras, check this one out:
http://www.hdcompare.com/Cameras.htm

there is a link to a specs comparison chart on the webpage