and discovered that the GL2 has a 6 blade iris. Nice. Also, the GL2 electronically squeezes the 16:9 image while recording, and stetches it when the footage is played back. Nice. The claimed minimum illumination is 6 lux---again, nice!

Another thing that's interesting is the time lapse function : .5, 1, 1.5 and 2 seconds; 1/2, 1, 5 and 10 minutes.

<<<-- Originally posted by Frank Granovski : and discovered that [...] the GL2 electronically squeezes the 16:9 image while recording, and stetches it when the footage is played back.

I don't think I understand the significance of this bit. I thought that when recording in 16:9 mode you were simply using fewer horizontal lines. I'm not sure what stretching it when played back would mean, either. What happens when you play back in another deck, or capture into your NLE?

If you found some good tech information on the GL2, could you share the source here? I've been trying to get hold of all the good in depth info I can find, lately.

Thanks.

lyd

Frank

I believe the 16:9 mode on the gl2 is essentially the same as the xl1s, and not materially different from the "16:9 guides on 4:3" mode. Essentially the image is just cropped from the normal frame, and then blown up...so resolution suffers somewhat...It is possible, that the gl2, with it's higher "head" resolution, actually benefits somewhat from this scheme, as it is filling the compressed DV frame with more info...but I'm not sure that this is actually the case....I don't have a 16:9 monitor so I can't really test it out...I'll try importing both modes into FCP to see if there is a difference.

Regarding the minimum lux issue...these specs are the same as the gl1 100lux recommended....6 minimum...but what's interesting is that canon has finally posted an online brochure for the gl2, and it has a different number.

http://canondv.com/downloads/brochures.html

It states that with the lower shutter speeds. the gl2's minimum is .37 lux...at 1/8 sec. Now this speed is unusable for my purposes, but 1/30 sec is very usable....(even 1/15 for relatively static shots). This means the camera is certainly capable of 1-2 lux. I think that's pretty decent for DV.

Barry

I realize that the 6 lux is subjective and that a lot more light is actually required, but it gives a good indication of it's lower light ability.

No, the GL2 doesn't crop the 16:9 footage. That was my point. Some cams "crop" and some cams "squeeze." I think capturing squeezed (then stretched) is the better of the two.

Frank, I still don't understand the mechanism you are describing.

"capturing squeezed (then stretched)" sounds a little bit like my understanding of what the optical anamorphic adapters do; optically "squeeze" a 16:9 image onto a 4:3 CCD, to be "stretched" later in software. I think this works pretty well with an oversampling CCD, but with most prosumer cameras you loose a little horizontal (as opposed to vertical) detail in the tradeoff, because you have to interpolate when you stretch it out.

In any case. if you are not optically "squeezeing" the image *before* it hits the 4:3 CCD, I don't understand in what way (or for what purpose) you would squeeze and stretch it in the digital domain.

Help?

lyd

Frank

Regarding the 16:9 issue, and I think this has been covered before in these forums, so I hope I'm not mistating....the gl2 is no different than the xl1s or gl1 in this respect...The "optical" image projected onto the chip is the same as 4:3, and it isn't squeezed, but cropped..ie losing about 1/4 of the vertical lines, and then stretched vertically (digitally) to fill the dv frame. Later, when viewing on a 16;9 monitor, or using your NLE, the footage is stretched again this time horizontally, to create the 16:9 image.

Canon has confused the issue by stating that the image is squeezed first...it isn't...that couldn't happen without an optical adapter or a wider chip...one look at the image when changing modes on the camera shows that the horizontal field doesn't change one pixel...but rather the vertical field is stretched digitally.

Again, the gl2 might actually benefit a little from this strategy, where as the previous cameras didn't (their "head" resolution is slightly lower than the DV format itself)....but I'm not sure this is the case...it really depends on how the image is actually being processed by the camera. I did find this quote from chris ... "canon has always recommended shooting in 4:3 and converting to 16:9 in post".

Cheers all. Have a great weekend.

Barry


Here's one thread that shows considerable inconclusiveness regarding the 16:9 issue..

http://www.dvinfo.net/conf/showthread.php?s=&threadid=1787&highlight=169

The pointer to the other thread helped me a lot... I was stuck thinking in terms of viewing letterboxed on a 4:3 display. I realized after reading through those other posts that you are going to scale and interpolate at least *something* to get true anamorphic 16:9 unless you have a 16:9 CCD.

Wierd, wacky stuff...

lyd

My footage does not unstretch unless it's played back on a deck that supports wide screen or if the tv does. But straight out of the GL2 you say. Is there a playback setting that I'm missing?

<<<-- Originally posted by Frank Granovski : and discovered that the GL2 has a 6 blade iris. Nice. Also, the GL2 electronically squeezes the 16:9 image while recording, and stetches it when the footage is played back. Nice. The claimed minimum illumination is 6 lux---again, nice!

Another thing that's interesting is the time lapse function : .5, 1, 1.5 and 2 seconds; 1/2, 1, 5 and 10 minutes. -->>>

I don't know, it's just what I read on one of Canon's tech bulletins. It's when you output, it gets stretched, not played back in the cam. Many cams just crop, that's why I thought I'd mention this---for those that didn't know this (such as myself).