DV Info Net

DV Info Net (https://www.dvinfo.net/forum/)
-   Canon GL Series DV Camcorders (https://www.dvinfo.net/forum/canon-gl-series-dv-camcorders/)
-   -   GL2 / XM2 Frame mode (https://www.dvinfo.net/forum/canon-gl-series-dv-camcorders/3415-gl2-xm2-frame-mode.html)

Don Palomaki February 24th, 2007 07:24 AM

All NTSC video is interlaced, comprised two fields (the odd or the even scan lines) spaced 1/60th apart in time to form the full frame. With movie mode the fields are captured individually and their content is spaced in time by 1/60th. You can see this in a ful frame capture with a moving object. The edges of the moving object will have a comb-like effect resulting from the objects change in position between the fields.

In frame mode the full frame (both fields) are captured at the same instant in time (not as separate fields spaced by 1/60th as in movie mode), thus their content is spaced 1/30th apart in time. On NTSC playback the individual fields of the frame stream are still spaced 1/60th in time, but their content was capture at the same instant. This results in differnt image artifacts

With frame mode motion in the image is not as smooth (gives more of a film like or strobe effect), and the capture method results in a slight loss of vertical resolution (which it turns out just might tend to mask some of the vertical motion artifacts).

Jason Sanders April 15th, 2007 12:41 PM

GL2 and deinterlacing
Im trying to get better results when filming green screen. Would a deinterlacer program like Revision Fieldkits be better then the gl2's frame mode?


John Hudson April 15th, 2007 01:23 PM

I'd say it would not bring you better results due to losing half the resolution. I'd think your green screen needs those pixels

Tatsuya Graham April 16th, 2007 03:11 AM

doesn't matter
I recently did some green screen work for a school project, we used a smaller camera, panasonic gs-180, and it worked fine without being deinterlaced, your main concern should be proper lighting, That plays a HUGE role in greenscreening.

Here is what we did with the greenscreen:


Yev Belman April 16th, 2007 05:39 PM

Jason, what issues did you run into with shooting interlaced for green screen? Have you ever tried Ultra 2 for keying?
From trying fieldskit myself, i think it is pretty tricky. have you gone thru this tutorial ?: http://www.creativecow.net/articles/...iew/index.html

Did you go from 60i to 60p ? I never got it to work, i only used it on 29.97 compositions to convert it to 30p (and then twixtor for retiming).

What software do you use to key?

Janssen Herr April 26th, 2007 03:46 AM

Frame mode works for me
I used both the frame mode and normal mode on my XM2 to test which method is better. I de-interlaced the normal mode using MB and the G-Natress plugs. After many tests, looking at scopes and trying many types of keyers I found the best results came from the frame mode. The 'progressive' output was perfect to snap a key from. Almost one click.

The work I do with green screens is primarily for the small screen, so the "loss of resolution" issue with using it is never a concern.

I have some footage of a worst case scenario test online here:


its ugly camera work, but it helps me see what happens in the worst case.

Ryan Thomas May 3rd, 2007 06:55 AM

Frame mode mixed with 30p...
Is it possible to mix footage of frame mode from GL2 with footage shot in 30p by XL1? I guess the real question is, is frame mode considered 30p? Thanks.

Benjamin Hill May 3rd, 2007 09:14 AM

Ryan, as I recall the XL1 doesn't have 30P but actually the same Frame mode as the GL2. Even if you did shoot true 30P with an XL2 or a DVX100, you could mix the two. They will cut together fine on a 29.97 timeline.

Frame mode isn't technically the same thing as 30P because it is not progressively scanned, but rather 60i video that is deinterlaced in-camera.

Ryan Thomas May 3rd, 2007 07:18 PM

Thanks for the info. That is what I was looking for.

Norman Woo May 10th, 2007 02:23 PM

Yes you can mix and match from the 2 cameras.

I do this quite often. One thing you may notice is that the FRAME mode of the GL2 appears somewhat sharper than that of the XL1.

Dale Guthormsen May 11th, 2007 10:11 PM


I mix the frame gl2 and progressive xl2 all the time. On my big screen plasma you would not notice any difference than a bit of resolution difference.

Stephen B. Malkinson August 1st, 2007 11:06 AM

Is there a downside to using FRAME mode on GL2
Believing that there is no such thing as a free lunch, is there any downside to using FRAME mode? If FRAME mode is as good as the manual and other posts suggest, why would I not use it as my default mode. I use Adobe Premier Elements for post production. Target for video is both web, portable device (iPOD) and DVD.

Ian Holb August 1st, 2007 12:06 PM

Frame mode is fine if all you want to do is save a step in post (ie. de-interlacing). It does a good job interpolating the missing field. I'd say subjectively a good de-interlacer reduces resolution by about 33%, Frame Mode reduces it by about 25%.

Cons of Frame Mode:

1. Lose 60 samples per second of motion (good for slow-motion)
2. Some people don't like the look of 30-frame motion.
3. Conforming to 24P is very difficult.


1. Quasi film-like motion.
2. No need to de-interlace in post, thus saving a step.
3. Probably easier on the DV codec, thus less mosquito noise.

It depends on your application, I guess. If you shoot sports, you might want to turn Frame Mode off. If you shoot narrative, then turning it on might give it less of a reality look (unless you want a daytime soap feel).

Merlin Vandenbossche August 1st, 2007 01:30 PM

I've always wondered if "frame mode" doesn't affect a showing on a TV-set (DVD-video)? Since the image is made up 'progressive'? Will that give problems with interlacing artifacts or stroboscopic motion? Since most of my projects so far were payed I never really cared to try or take the risk. I had picked up somewhere that the GL-2 did not have a true 'progressive mode', but that frame mode was somewhere in between progressive and interlaced. I heard it had something to do with a specific usage of the green pixels CCD.

Don Palomaki August 1st, 2007 05:48 PM

Frame mode uses a vertical pixel shift technique to form a full frame (2 fields) that is captured from a single exposure. Thus it has somewhat reduced vertical resolution. It is recorded to tape (and played back from tape) as an interlaced video signal, but because the two fields were captured in the same exposure, the typical motion interlace artifacts between fields in a frame will be absent. This can give the motion a bit less smooth look (but not quite strobed to most eyes, unless you use a fast shutter), and that also accounts for much of the the film-like look.

Some folks like it, some do not. Give it a try (not on a paying job) and see if you like it.

All times are GMT -6. The time now is 02:53 AM.

DV Info Net -- Real Names, Real People, Real Info!
1998-2020 The Digital Video Information Network