Mixing HD & SD footage with FX1 and Canon GL2
 

Hi,
I just got my FX1 today and I have a wedding to film this weekend. So, I'm kind of learning as much as I can right now both with the camera and in these forums.
I hope someone doesn't mind answering a few questions.
I plan to shoot the wedding with the FX1 as my front (main) camera and film in HDV. I am going to use a Canon GL2 as a second camera.
When downconverting to SD for editing, it looks as though the video from the FX1 will have black bars at the top and bottom - something I don't mind, but my customer may have questions about.

1) Will it be odd to mix footage from the FX1 that has been downconverted with standard DV footage from the Canon GL2? Is there a way to import the footage (anamorphic, or whatever) so that there are no bars? Or, what is the correct setting for importing? I am using Final Cut Pro HD 4.5? Again, the bars don't bother me and if they have to be there, I can tell the customer that in the future I'll deliver it to them in HD.

2) When I do edit for HD, will it look odd when switching between the FX1 and GL2? I don't mind the GL2 having bars on the side.
If there is a better way to do all this, such as just filming in SD on the FX1, then plese let me know that because it may be best at this point.
Thanks and sorry the post is so wordy,
- John
Oh, and I am excited. I am filming stuff like crazy. Ahhh...HD!

John: I can tell you from personal experience that mixing HDV footage from a Sony FX1 and DV footage from a Canon GL2 is somewhat problematic. There are several ways to approach this, but my recommendation would be to run the FX1 in HDV mode, the GL2 in standard SD mode, and then trim the edges off the FX1 footage for SD editing and delivery. This way you at least get some HD footage you can use later for demo purposes, but your delivery to your customer will be SD only for now because otherwise the switch between cameras will be too noticeable.

Hi Kevin,
Thanks for your response. By trimming the edges, do you mean importing the footage as HDV, editing it with the GL2 footage, and then exporting in SD, which would cut off the left and right edges (pillars, I guess)? Is there enough information in the HD footage to expand it a little to cover the top and bottom of the screen so that the footage would be full-screen after exporting to SD?
Sorry for all the questions, but I'm just a little worried about getting this right so that I can give the customer what they are expecting, which at this point is SD. I haven't mentioned HD to them yet, and they really only have an NTSC tv anyway. They may not understand the bars, if they are there. :)
Thanks,
- John

And one more thing...
What about framing shots when you shoot HDV that will be downconverted to SD? I have turned on the center crosshair on the FX1, and I guess I'll try to use the rule of thirds as best I can apply it around the crosshair. The widescreen is throwing me off because I know that some of the edges will be lost when downconverting after editing, if I'm correct in my thinking.
- John

John: yes, the best thing I've found is to edit the HDV at full resolution with the left and right edges cropped such that you end up with a 4x3 frame, which has more than enough information to mix with standard SD footage. Not sure the best way to do that in Final Cut Pro, but that's the basic idea. As far as editing for HD is concerned, don't plan on doing that with any GL2 footage involved. After you've mixed FX

John: yes, the best thing I've found is to edit the HDV at full resolution with the left and right edges cropped such that you end up with a 4x3 frame, which has more than enough information to mix with standard SD footage. Not sure the best way to do that in Final Cut Pro, but that's the basic idea. As far as editing for HD is concerned, don't plan on doing that with any GL2 footage involved. After you've mixed FX1 and GL2 footage, you'll understand just how dramatic a difference shooting in HDV makes. At some point you'll probably want another HDV camera...

Your best bet will be to shoot in HDV and "down convert" to SD when logging. Reason: If you edit in HDV you will need to set sequence preset to HDV in FCP. (I am not sure you can do this with FCP 4.5. I use version 5.) Anyway HDV footage has more than 4 times the resolution of SD footage. So when you put GL-2 footage on a HDV timeline you will fine the SD footage only occupy a small portion of the screen and that's not gonna work at all (unless you treat it as a special effect.)

So downconvert to sd when you log and it will work with gl-2. and next you may need to deal with the noticable differences of footage between 2 cameras, which is not small (color balance, saturation, sharpness...). Your best bet will be using GL-2 only in well-lit, close up shots. GL-2 in close-up is at its best and you will have a relative easier time mixing them. I know, because I own both of them.

Good luck.

Kevin and Max, thank you for the help.

I am importing now by downconverting in the camera and it seems as though I will have black bars at the top and bottom of the screen. Is that correct, or is there a setting I need to change? I don't mind losing footage on either edge of the screen, but I would like a 4:3 picture without bars at the top and bottom, if possible so that it doesn't look odd when switching to DV footage shot on the GL2.
Thanks again,
- John

In one of the menus in the downconvert neighborhood you can choose letterbox, squeeze or crop. Make a 4X3 target and shoot it while in 16X9/ HDV and use black tape (camera or paper tape is best) to physically crop the LCD. Then do a test and cut the two cameras together--- you will have to adjust a Picture Profile to match. That should do the trick. Test, Test and Test or everyone could be unhappy. Try a daylight scene and an interior scene to see if the PP you have customized tracks between the two.

I bet there is someone out there who has already matched the two cameras....they could save you the work of painting the Sony.

George

Hi,
Thanks. I found the setting your talking about that gives the 4:3 ratio markers on the viewer. I may mark the LCD's edges with tape.
I don't have access to a computer that's capable of editing HDV until after the wedding, so I am still wondering if someone can tell me if an HDV picture will fill up all sides of the video when editing for regular DV. What I mean is, it seems like the number of pixels in HDV would be so much greater, 1920x1080 in HDV versus 702X480 in DV, that the image would allow me to crop both the top and sides of the HDV video in order to export it as standard DV.
Am I correct on this?
Thanks,
- John

If your goal is to be able to zoom in on your HDV footage to select a portion of it, that is possible if you capture at HD resolution. But don't count on being able to zoom in too much: if you enlarge much more than 50% the image quality starts to deteriorate.

Is there really much advantage over shooting in HDV or DV when you are planning on downconverting to SD anyway? Since it's a two camera shoot with the other being SD you won't be holding onto the HDV for a HD render later, so that may not be an issue. Just curious because I was under the impression that there wasn't any significant quality difference between shooting HDV and downconverting to SD vs just shooting in SD.

I recently did a 3 camera shoot with the FX1 in DV mode, a VX2000 and a Panasonic AG-DVX100A. There was no black bar issue then, no cropping to do, but maybe the image quality was compromised. Unfortunately, I am still editing it so haven't been able to make an objective evaluation of the different videos.

There's quite a few threads about the "downconvert HDV to SD quality vs shoot as DV" issue here at DVinfo. Many who have gone the downconvert HDV route feel that it's superior. My tests have found that it is superior, but only if it's done "properly"... with a decent 'uncompressed' intermediate format - like Cineforms' CFHD avi codec, and a knowledge of the proper settings (in the NLE of choice) for the re-encode to limit re-encoding artifacts.

I don't personally have a VX2000 or DVX100A to compare DV clips shot on my FX-1 with, but I doubt that the quality would be worse - as long as the exposure variation and lens length differences between the different cams were accounted for. Correct DV aspect ratio would also need to be set on the FX-1 if shooting to DV, unless you're prepared to crop/pan HDV into a 4:3 project... which takes reasonably significant render time.

I guess it's fair to say - if you want superior quality DV from a HDV camcorder; shoot in HDV... but be prepared for longer editing/render times, and make sure you've got decent HDV capable software.

If you want quick results that don't require toying with software settings etc to extract the better downconvert result... shoot using DV mode.

I was thinking that I would be able to crop the sides, top and bottom to match it with the DV shot on the GL2, without zooming in on the HDV footage.
Maybe it would be better to shoot in DV mode, at this point. I haven't mentioned HD to the people I'm shooting for. I haven't played with regular DV on the FX1 yet, either.
But without zooming in, there is a big enough picture to crop into 4:3 with no bars on top and bottom...correct? Again, the problem I have is not being able to play with this in a timeline because my system isn't fast enough to import HDV. I will be able to experiment next week on an iMac G5, but this will be after the wedding, which is tomorrow. Doh!
In Final Cut Pro's preview windows, there are overlays for title-safe and what I guess would be the "action-area," or roughly the borders of a 4:3 TV screen. I can't verify at this point that HDV covers all of that (I think the pixels would be more than enough). I would assume that the 4:3 box would fit inside the HDV picture and leave no black bars and room to spare at the top and bottom, and plenty of room to pan across the picture if necessary. And, I can see that it would take a long time to render.
The problem with downconverting is that it gives the black bars and DV quality. Wouldn't importing full HDV let you crop the picture with no bars and mix it with SD footage, since the HDV footage would be exported as DV?
- John

I don't have FCP to know how it would deal with the crop of 16:9 HDV into 4:3 DV, but in Vegas on PC it's relatively painless and crops fully without 'bars'.

It's a shame you don't have the time to verify that this'll work in your case.