It's official: Canon XL2 announced

Rob Lohman · July 14th, 2004, 06:54 AM

And I doubt they will, and who cares. It's about the end product
anyway. And since it is plain DV bitrate is easy to answer: 25 mbps.

And MHz of what? Who cares if the colorspace is RGB or YUV?
It is stored as YUV in DV anyway.

So Laurence, what do these answers tell you and in what way
do they relate to the camera? I'm really interested in your MHz
question....

I'm keeping my eye on this thread. To judge final quality.

Greg Matty · July 14th, 2004, 07:50 AM

<<<-- Originally posted by Rob Lohman :
I hope this explained things a bit better. -->>>

Rob,

That was a great explanation. Thanks for taking the time.

My DVX-100 only has the letterboxed 16:9 where the camera merely adds black bars on top and bottom of the screen. The DVX-100a has the digital squeeze and I always thought is somehow shot a wider angle image, compressed it horizontally and then recorded it to tape. Then the NLE would re-sample the image to the correct number of horizontal pixels and you are good to go.

I did not know the horizontally squeezed image was actually stretched vertically. Thanks for clearing that up.

Greg

Rob Lohman · July 14th, 2004, 08:10 AM

The easy way to see if you have true 16:9 is if your FOV (Field Of
View) widens when you switch from 4:3 to 16:9. So you should
actually SEE a WIDER picture then before.

If it uses a fake 16:9 stretch you will either just see black bars
being added or everyone will suddenly look taller.

There is only one catch. If you have a true 16:9 camera (or anamorphic
attachment) but with a normal viewfinder your people will look
taller AS WELL, but the FOV should ALSO get wider!

Kevin Lepp · July 14th, 2004, 08:21 AM

Im just wondering why Canon decided not to use the whole area of the 1/3 ccd that they created. Since the XL2 (in the viewfinder) shows you the 16:9 with black bars on top and bottom, and the 4:3 covering the whole screen, why didnt they just use the whole 1/3 ccd to capture the 4:3 image? (Because it wouldnt change the shape in the viewfinder- it would just change the DoF of the 4:3 making it more of a wide angle, but that would be welcome) That would obviously give the 4:3 image a huge jump in resolution.
They went through all the trouble to put more pixels in the ccd, but now are only using this advantage in the 16:9 aspect. Obviously the 4:3 is elarged compared to the XL1s, but none the less, its still not using its full capacity technically (that is if they put the rest of the non-firing pixels into use).

I just wonder why they chose not to use that extra space.
Anyone have any theories or knowledge of why they didnt?

And I really hope that they didnt choose to not use that space because it "technically" wouldnt make the XL2 a true 16:9 camera.... that would be very pitiful to me- because obviously it wouldnt change the area on the ccd where the 16:9 aspect recorded from, thus not changing any part of how it now actually is. It would be the same, but with a better 4:3 version.

Tommy Haupfear · July 14th, 2004, 08:57 AM

Kevin, as far as I understand it you don't need much over 340k pixels to get an excellent 4:3 720x480 picture. Just look at the VX2100 or other 1/3" 3CCD cams.

I think the extra pixels are there only for 16:9.

Michael Bott · July 14th, 2004, 10:44 AM

Kevin, my understanding is that in order to make the XL2 backwards compatible with the existing XL lenses and attachements, they were constrained to use a 1/3 chip. Now, Canon knew they would also HAVE to include native 16:9 in order to make the new camera a worthwhile upgrade in the customers mind. Obviously 16:9 was always going to take up the full width of the chip so to offer 4:3 on the same chip without altering the plane of the lens or the chip and incurring all the attendant focussing nightmares, this would HAVE to be acheived by chopping down the sides of the image.

Have I said that right??

Kevin Lepp · July 14th, 2004, 12:02 PM

Well, if the XL1s uses a 1/3 ccd then there should be no compatability problems if they wanted to use the whole chip. Though if its true that the extra pixels dont make a difference on the recording, then that may be why they chose to do it the way they did. But if not, then I guess the reason they did it was because they wanted a true 16:9.

I dont know.

What does HD record on? It doesnt use DV tapes does it?

thanks,

Michael Bott · July 14th, 2004, 12:20 PM

I'm trying to get my head round this myself.

So - same lens and a 1/3 chip in the same place as before, right? OK, so if we use the whole of the chip for 4:3, how do we get 16:9 out of the same configuration without employing squeeze or crop?

You see what I mean?

Canon have gone the other way round - ASSUMED 16:9 and cropped to get 4:3. That way, we don't have to think hard about loss of horizontal resolution when choosing to shoot 16:9 over 4:3.

Thing is, because they've managed to fit more pixels in, even wasting the top and bottom of the chip we still get better resolution.

Have I got this right? Somebody clever help me here!

Kevin Lepp · July 14th, 2004, 12:30 PM

I hear what your saying, but either way you have to crop. So if you're going to have to crop, why not crop with the bigger 4:3? The only thing that happens if you crop from 4:3 to 16:9 is that you cant say its "true" 16:9. But if you do it the way they did it, you can say its "true" 16:9.... either way though, your 16:9 picture remains the same.

Milosz Krzyzaniak · July 14th, 2004, 12:47 PM

I agree, Kevin. The method that canon chose is the true 16:9, but we 4:3 players could argue that we don't have the true 4:3 mode right now....

Shame for the DOF that's going to be sighnificantly worse I suppose.

Laurence Maher · July 14th, 2004, 01:13 PM

Rob,

Good points, and I'll tell you what they tell me . . .

For all the hype, etc. . . . this camera SUCKS!

LOLOLOLOLOOLOLOOLOLOL

Michael Bott · July 14th, 2004, 01:48 PM

Yeah, I get that the chip has been *hobbled* in that way. But if I were Canon I'd have done the same ...

Imagine them coming to market with a fine 4:3 cam with great resolution but that when you wanted 16:9 you were into the losing 25% res game all over again. Exactly the same frustrations of trying to find an anamorphic adapter to extract the resolution that you just *know* is there. It's a human nature thing. And it's a marketing thing. Canon had to bring *native* 16:9 to this release for it to be a worthwhile upgrade and they had to stick with a 1/3 chip to remain backward compatible. I can see their dilemma.

Aaron Koolen · July 14th, 2004, 02:45 PM

I hope that marketting isn't the reason. I've seen so many companies do this sort of thing. Release sub-par products (And for me, <1/3" 4:3 is subpar) just to have a tick on a box.

Aaron

Michael Bott · July 14th, 2004, 02:55 PM

<And for me, <1/3" 4:3 is subpar>

Even when it's way better than what we've got now?

Aaron Koolen · July 14th, 2004, 04:52 PM

Michael, I'm not bagging it. I'm just not impressed and just have this "gut" feeling that Canon isn't a company that's wants to break new ground. I love innovation and, companies outdoing each other with all these great enhancements. It benefits us, the end user immensely. I just don't (so far) see that the XL2 has really jumped up and kicked me in the nads to say it's here and gonna challange the DVX etc.

I'm sure it will be way better that the xl1s, but as far as CCD size goes I was thinking about people who want to squeeze whatever shallow DoF they can out of it, and also usually the smaller pixels on the CCD's mean lower sensitivity right?. If these aren't issues then it will probably be fine for people. I guess there's this mindset that 1/3" is the standard for a "prosumer" camera and to go back on that sounds risky.

As I keep saying "The proof is in the pudding". I want to see footage.

Aaron

July 14th, 2004, 06:54 AM	#106
Rob Lohman RED Code Chef Join Date: Oct 2001 Location: Holland Posts: 12,514	And I doubt they will, and who cares. It's about the end product anyway. And since it is plain DV bitrate is easy to answer: 25 mbps. And MHz of what? Who cares if the colorspace is RGB or YUV? It is stored as YUV in DV anyway. So Laurence, what do these answers tell you and in what way do they relate to the camera? I'm really interested in your MHz question.... I'm keeping my eye on this thread. To judge final quality. __________________ 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 14th, 2004, 08:10 AM	#108
Rob Lohman RED Code Chef Join Date: Oct 2001 Location: Holland Posts: 12,514	The easy way to see if you have true 16:9 is if your FOV (Field Of View) widens when you switch from 4:3 to 16:9. So you should actually SEE a WIDER picture then before. If it uses a fake 16:9 stretch you will either just see black bars being added or everyone will suddenly look taller. There is only one catch. If you have a true 16:9 camera (or anamorphic attachment) but with a normal viewfinder your people will look taller AS WELL, but the FOV should ALSO get wider! __________________ 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 14th, 2004, 08:21 AM	#109
Kevin Lepp Regular Crew Join Date: Jul 2004 Location: Charlotte, NC Posts: 38	Why not use the whole area? Im just wondering why Canon decided not to use the whole area of the 1/3 ccd that they created. Since the XL2 (in the viewfinder) shows you the 16:9 with black bars on top and bottom, and the 4:3 covering the whole screen, why didnt they just use the whole 1/3 ccd to capture the 4:3 image? (Because it wouldnt change the shape in the viewfinder- it would just change the DoF of the 4:3 making it more of a wide angle, but that would be welcome) That would obviously give the 4:3 image a huge jump in resolution. They went through all the trouble to put more pixels in the ccd, but now are only using this advantage in the 16:9 aspect. Obviously the 4:3 is elarged compared to the XL1s, but none the less, its still not using its full capacity technically (that is if they put the rest of the non-firing pixels into use). I just wonder why they chose not to use that extra space. Anyone have any theories or knowledge of why they didnt? And I really hope that they didnt choose to not use that space because it "technically" wouldnt make the XL2 a true 16:9 camera.... that would be very pitiful to me- because obviously it wouldnt change the area on the ccd where the 16:9 aspect recorded from, thus not changing any part of how it now actually is. It would be the same, but with a better 4:3 version. __________________ Kevin

July 14th, 2004, 12:02 PM	#112
Kevin Lepp Regular Crew Join Date: Jul 2004 Location: Charlotte, NC Posts: 38	Well, if the XL1s uses a 1/3 ccd then there should be no compatability problems if they wanted to use the whole chip. Though if its true that the extra pixels dont make a difference on the recording, then that may be why they chose to do it the way they did. But if not, then I guess the reason they did it was because they wanted a true 16:9. I dont know. What does HD record on? It doesnt use DV tapes does it? thanks, __________________ Kevin

July 14th, 2004, 12:30 PM	#114
Kevin Lepp Regular Crew Join Date: Jul 2004 Location: Charlotte, NC Posts: 38	I hear what your saying, but either way you have to crop. So if you're going to have to crop, why not crop with the bigger 4:3? The only thing that happens if you crop from 4:3 to 16:9 is that you cant say its "true" 16:9. But if you do it the way they did it, you can say its "true" 16:9.... either way though, your 16:9 picture remains the same. __________________ Kevin

July 14th, 2004, 07:50 AM	#107
Greg Matty Regular Crew Join Date: Mar 2002 Location: Kirkland, WA Posts: 177	<<<-- Originally posted by Rob Lohman : I hope this explained things a bit better. -->>> Rob, That was a great explanation. Thanks for taking the time. My DVX-100 only has the letterboxed 16:9 where the camera merely adds black bars on top and bottom of the screen. The DVX-100a has the digital squeeze and I always thought is somehow shot a wider angle image, compressed it horizontally and then recorded it to tape. Then the NLE would re-sample the image to the correct number of horizontal pixels and you are good to go. I did not know the horizontally squeezed image was actually stretched vertically. Thanks for clearing that up. Greg

July 14th, 2004, 08:57 AM	#110
Tommy Haupfear Wrangler Join Date: Apr 2003 Location: Greenville, SC Posts: 1,415	Kevin, as far as I understand it you don't need much over 340k pixels to get an excellent 4:3 720x480 picture. Just look at the VX2100 or other 1/3" 3CCD cams. I think the extra pixels are there only for 16:9.

July 14th, 2004, 10:44 AM	#111
Michael Bott Regular Crew Join Date: Jun 2004 Location: Stratford-upon-Avon, Warwickshire, UK Posts: 47	Kevin, my understanding is that in order to make the XL2 backwards compatible with the existing XL lenses and attachements, they were constrained to use a 1/3 chip. Now, Canon knew they would also HAVE to include native 16:9 in order to make the new camera a worthwhile upgrade in the customers mind. Obviously 16:9 was always going to take up the full width of the chip so to offer 4:3 on the same chip without altering the plane of the lens or the chip and incurring all the attendant focussing nightmares, this would HAVE to be acheived by chopping down the sides of the image. Have I said that right??

July 14th, 2004, 12:20 PM	#113
Michael Bott Regular Crew Join Date: Jun 2004 Location: Stratford-upon-Avon, Warwickshire, UK Posts: 47	I'm trying to get my head round this myself. So - same lens and a 1/3 chip in the same place as before, right? OK, so if we use the whole of the chip for 4:3, how do we get 16:9 out of the same configuration without employing squeeze or crop? You see what I mean? Canon have gone the other way round - ASSUMED 16:9 and cropped to get 4:3. That way, we don't have to think hard about loss of horizontal resolution when choosing to shoot 16:9 over 4:3. Thing is, because they've managed to fit more pixels in, even wasting the top and bottom of the chip we still get better resolution. Have I got this right? Somebody clever help me here!

July 14th, 2004, 12:47 PM	#115
Milosz Krzyzaniak Regular Crew Join Date: Mar 2004 Location: Germany Posts: 64	About The Truth I agree, Kevin. The method that canon chose is the true 16:9, but we 4:3 players could argue that we don't have the true 4:3 mode right now.... Shame for the DOF that's going to be sighnificantly worse I suppose.

July 14th, 2004, 01:13 PM	#116
Laurence Maher Join Date: May 2004 Location: denton, texas, usa Posts: 416	Rob, Good points, and I'll tell you what they tell me . . . For all the hype, etc. . . . this camera SUCKS! LOLOLOLOLOOLOLOOLOLOL

July 14th, 2004, 01:48 PM	#117
Michael Bott Regular Crew Join Date: Jun 2004 Location: Stratford-upon-Avon, Warwickshire, UK Posts: 47	Yeah, I get that the chip has been hobbled in that way. But if I were Canon I'd have done the same ... Imagine them coming to market with a fine 4:3 cam with great resolution but that when you wanted 16:9 you were into the losing 25% res game all over again. Exactly the same frustrations of trying to find an anamorphic adapter to extract the resolution that you just know is there. It's a human nature thing. And it's a marketing thing. Canon had to bring native 16:9 to this release for it to be a worthwhile upgrade and they had to stick with a 1/3 chip to remain backward compatible. I can see their dilemma.

July 14th, 2004, 02:45 PM	#118
Aaron Koolen Trustee Join Date: Apr 2002 Location: Auckland, New Zealand Posts: 1,727	I hope that marketting isn't the reason. I've seen so many companies do this sort of thing. Release sub-par products (And for me, <1/3" 4:3 is subpar) just to have a tick on a box. Aaron __________________ My Website Meat Free Media

July 14th, 2004, 02:55 PM	#119
Michael Bott Regular Crew Join Date: Jun 2004 Location: Stratford-upon-Avon, Warwickshire, UK Posts: 47	<And for me, <1/3" 4:3 is subpar> Even when it's way better than what we've got now?

July 14th, 2004, 04:52 PM	#120
Aaron Koolen Trustee Join Date: Apr 2002 Location: Auckland, New Zealand Posts: 1,727	Michael, I'm not bagging it. I'm just not impressed and just have this "gut" feeling that Canon isn't a company that's wants to break new ground. I love innovation and, companies outdoing each other with all these great enhancements. It benefits us, the end user immensely. I just don't (so far) see that the XL2 has really jumped up and kicked me in the nads to say it's here and gonna challange the DVX etc. I'm sure it will be way better that the xl1s, but as far as CCD size goes I was thinking about people who want to squeeze whatever shallow DoF they can out of it, and also usually the smaller pixels on the CCD's mean lower sensitivity right?. If these aren't issues then it will probably be fine for people. I guess there's this mindset that 1/3" is the standard for a "prosumer" camera and to go back on that sounds risky. As I keep saying "The proof is in the pudding". I want to see footage. Aaron __________________ My Website Meat Free Media