A (stupid) PAL XL2 question

Barry Green · July 23rd, 2004, 07:06 PM

Quote:

<<<-- Originally posted by Bill Anderson : But if it were as simple as that then every camera chip need be no more than the pal or NTSC frame size. -->>>

It is as simple as that. For optimal video quality, that is what you want. You want the biggest pixels you can get, as long as there are enough of them to cover the frame. That's why in broadcast cameras you'll always see CCD's with just enough pixels to cover the frame.

Barry Green · July 23rd, 2004, 07:09 PM

Quote:

<<<-- Originally posted by Russell Newquist :
However, the real point is that having more pixels on your CCDs than on your final image (as the XL2 does) is far better than having less pixels on your CCDs than on your final image (as the XL1 and XL1s do), so there should definitely be a noticeable picture quality increase when upgrading from the older cameras to the newer one. -->>>

True to a point. The # of pixels reflects the density of the pixels on the chip, and to some degree that reflects the resolution the camera is capable of. But there are other factors that are affected by the pixel count as well, namely latitude and low light performance. Tiny pixels have lousy light-gathering capability and can't hold as much of a charge so you get worse low-light performance. Resolution is not the sole determining factor for whether a video image looks good.

The optimal solution is to have just enough pixels to cover your frame size, and to have as large a chip as possible.

Bill Anderson · July 23rd, 2004, 07:32 PM

However Barry it would not be a benifit for Canon to reduce the pixel count. Would it? And according to your statement about having enough pixels to just cover the chip might read to a straight logic layman type like myself as meaning all chips have just enough pixels to cover them, otherwise they would simply not be covered (with those particular size pixels) Now it seems that there is an optimum size of pixel. What is its size? I don't understand how it could be just enough to cover the chip because you could do that with 4 or 6, but this would not be optimum. I'm going to assume that Canon optimised the pixel count for that size of chip otherwise I'll be wearing a straight-jacket before the XL2 hits the streets. Thanks for the info, it did help.

Barry Green · July 24th, 2004, 12:23 AM

I meant "enough pixels to cover the format's frame". i.e., for DV in an NTSC camera, you want at least 720 x 480 pixels. For a high-def 1080 camera, you'd want to have at least 1920 x 1080 pixels. Any more than that would mean that the pixels were smaller than necessary, any fewer than that would mean limited resolution.

Canon put in enough pixels to cover the DV frame at a 1:1 ratio, considering that the CCD pixels are square and the DV frame pixels are non-square. However, had they made the chip be actually 16:9-shaped, those pixels could have been much larger, which would mean better light-gathering capability and less issues with latitude and noise. Because they only use a smaller 16:9-shaped chunk off a 4:3 CCD, they had to cram a lot more smaller pixels onto the chip in order to have enough to cover the DV frame with enough pixels to provide adequate resolution.

Rob Lohman · August 2nd, 2004, 10:19 AM

Just to indicate it again, shooting in true/native 16:9 on DV will
increase your VERTICAL resolution, NOT horizontal. Yes, horizontal
will have a higher sampling rate and this wider field of view, but
as indicated only 720 pixels of those are stored.

So beside the wider FOV and higher sampling the real benefit is
in the vertical resolution.

Joe Carney · August 2nd, 2004, 12:29 PM

>>The question of why we're not all using PAL now is simple (as Chris implied) - namely that we're dealing with the matter of huge scale. North America is a big, big market<<

Colin, heres some trivia for you...

When I was stationed in West Germany in the late 70s, I brought my B&W NTSC TV with me and a step down voltage converter (from 220 to 110) but still at 50hz. I was able to pick up the local German TV station (video only) just fine. Great picture and everything.
Armed Forces Radio and Television Stations managed to broadcast 50hz with NTSC style interleaved audio so we had sound.

So... in order to save money...do the latest TVs have the built in ability to show either PAL or NTSC? And is the NTSC market simply not being told about it? hmmmmmm That would go with the North America gets dumbed down products theory, hehehehe

August 2nd, 2004, 10:19 AM	#35
Rob Lohman RED Code Chef Join Date: Oct 2001 Location: Holland Posts: 12,514	Just to indicate it again, shooting in true/native 16:9 on DV will increase your VERTICAL resolution, NOT horizontal. Yes, horizontal will have a higher sampling rate and this wider field of view, but as indicated only 720 pixels of those are stored. So beside the wider FOV and higher sampling the real benefit is in the vertical resolution. __________________ Rob Lohman, visuar@iname.com DV Info Wrangler & RED Code Chef Join the DV Challenge \| Lady X Search DVinfo.net for quick answers \| Buy from the best: DVinfo.net sponsors

July 23rd, 2004, 07:32 PM	#33
Bill Anderson Regular Crew Join Date: Jul 2004 Location: Ottawa. Canada Posts: 163	However Barry it would not be a benifit for Canon to reduce the pixel count. Would it? And according to your statement about having enough pixels to just cover the chip might read to a straight logic layman type like myself as meaning all chips have just enough pixels to cover them, otherwise they would simply not be covered (with those particular size pixels) Now it seems that there is an optimum size of pixel. What is its size? I don't understand how it could be just enough to cover the chip because you could do that with 4 or 6, but this would not be optimum. I'm going to assume that Canon optimised the pixel count for that size of chip otherwise I'll be wearing a straight-jacket before the XL2 hits the streets. Thanks for the info, it did help.

July 24th, 2004, 12:23 AM	#34
Barry Green Barry Wan Kenobi Join Date: Jul 2003 Location: North Carolina Posts: 3,863	I meant "enough pixels to cover the format's frame". i.e., for DV in an NTSC camera, you want at least 720 x 480 pixels. For a high-def 1080 camera, you'd want to have at least 1920 x 1080 pixels. Any more than that would mean that the pixels were smaller than necessary, any fewer than that would mean limited resolution. Canon put in enough pixels to cover the DV frame at a 1:1 ratio, considering that the CCD pixels are square and the DV frame pixels are non-square. However, had they made the chip be actually 16:9-shaped, those pixels could have been much larger, which would mean better light-gathering capability and less issues with latitude and noise. Because they only use a smaller 16:9-shaped chunk off a 4:3 CCD, they had to cram a lot more smaller pixels onto the chip in order to have enough to cover the DV frame with enough pixels to provide adequate resolution.

August 2nd, 2004, 12:29 PM	#36
Joe Carney Inner Circle Join Date: Jul 2002 Location: Centreville Va Posts: 1,828	>>The question of why we're not all using PAL now is simple (as Chris implied) - namely that we're dealing with the matter of huge scale. North America is a big, big market<< Colin, heres some trivia for you... When I was stationed in West Germany in the late 70s, I brought my B&W NTSC TV with me and a step down voltage converter (from 220 to 110) but still at 50hz. I was able to pick up the local German TV station (video only) just fine. Great picture and everything. Armed Forces Radio and Television Stations managed to broadcast 50hz with NTSC style interleaved audio so we had sound. So... in order to save money...do the latest TVs have the built in ability to show either PAL or NTSC? And is the NTSC market simply not being told about it? hmmmmmm That would go with the North America gets dumbed down products theory, hehehehe