Getting maximum video quality from XL2

[Nadia Huggins]

Hi all

just curious as to what settings could I use on the camera to get the best quality out of the XL2..

I messed around with the colour, brightness and sharpness and i'm still not getting the richness.

 

am I doing something wrong?

[Steve Madsen]

Hi Nadia,

 

Canon's prosumer cameras tend to be fairly flat looking out of the box. Assuming your light is doing its job, try dropping the master pedestal and boosting the colour.

[Nadia Huggins]

Hi Nadia,

 

Canon's prosumer cameras tend to be fairly flat looking out of the box. Assuming your light is doing its job, try dropping the master pedestal and boosting the colour.

 

 

Thanks alot

i'll give it a shot tomorrow

:)

[Josh Bass]

A couple things:

 

 

When you make changes to color, pedestal, etc., you have to do it within a custom preset, and then make sure that preset is turned on to see the changes.

 

Secondly, if you're used to the DVX series from Panasonic, you'll find that the XL2 doesn't let you push the look as far as you can push the DVX in terms of saturation and contrast. The XL2 seems to be trying to keep your image legal and nice looking, so if you want to push your looks really far, you'll have to do in post.

 

That being said, create a custom preset with the following parameters changed:

 

 

 

color matrix - cine

gamma - cine

Knee - low

Black - press

 

This should give you a noticably snappier and cinematic look.

 

From there, if you increaste the color saturation, you'll get a more colorful saturated look. You can also manipulate the R,G, and B gain (increase them) in unison to get even more color saturation.

 

You can manipulate the white balance, to some degree, by messing around with the R, G, and B gain separately.

 

As you decrease master ped and setup, the image will get contrastier as the shadow portions of your image get darker.

[Alessandro Machi]

A couple things:

When you make changes to color, pedestal, etc., you have to do it within a custom preset, and then make sure that preset is turned on to see the changes.

 

Secondly, if you're used to the DVX series from Panasonic, you'll find that the XL2 doesn't let you push the look as far as you can push the DVX in terms of saturation and contrast. The XL2 seems to be trying to keep your image legal and nice looking, so if you want to push your looks really far, you'll have to do in post.

 

That being said, create a custom preset with the following parameters changed:

color matrix - cine

gamma - cine

Knee - low

Black - press

 

This should give you a noticably snappier and cinematic look.

 

From there, if you increase the color saturation, you'll get a more colorful saturated look. You can also manipulate the R,G, and B gain (increase them) in unison to get even more color saturation.

 

You can manipulate the white balance, to some degree, by messing around with the R, G, and B gain separately.

 

As you decrease master ped and setup, the image will get contrastier as the shadow portions of your image get darker.

 

When shooting low budget I'm not a fan of trying to make the on location image have a certain look. I'm more in favor of trying to keep as much of the image intact as possible, and then learn to do the video adjusting during video edit. In my tape editing room, I have huge, and I mean a huge ability to make changes to an image. I'm assuming that if it can be done in the tape world, you should be able to do the same level of adjusting in the computer world, if not more. If that concept is incorrect, please correct me.

 

The danger in doing the above adjustments will be evident when you shoot in high contrast situations.

[Josh Bass]

Yes, true, but I've also heard people who advocate getting as close to your final look in camera as possible, as opposed to shooting a clean, neutral image and manipulating it later.

 

At any rate, she was asking why her images lacked punch, so I was trying to give her some clues as to how to make it snappier using in-camera methods.

[Josh Bass]

Are we not able to edit posts anymore?

 

 

I remembered--

 

another reason I don't necessarily like to shoot using your method is that if you set the XL2 for maximum dynamic range by putting the black setting on "stretch", you get really weird results if portions of your image are severely underexposed (I'm talking around 5 stops under here). For instance, on a sunny day, if you shoot something with both shady areas and areas getting blasted by direct sunlight, and then stop down/use NDs so that the sunlit areas are correctly exposed, the shady areas in your shot, which should be nearly invisible, are instead perfectly visible and sharply defined, albeit very dark, and are almost colorless, everything being reduced to a desaturated blue/grey. It's really odd. I don't like the way it looks at all.

[Michael Nash]

When shooting low budget I'm not a fan of trying to make the on location image have a certain look. I'm more in favor of trying to keep as much of the image intact as possible, and then learn to do the video adjusting during video edit. In my tape editing room, I have huge, and I mean a huge ability to make changes to an image. I'm assuming that if it can be done in the tape world, you should be able to do the same level of adjusting in the computer world, if not more. If that concept is incorrect, please correct me.

 

As I've taken to saying lately, "it's usually best to do in-camera what you can do well in-camera." That means that the closer you can get the image to looking "right" in camera, the better it will look after post tweaking.

 

Whatever camera system you're using (digital or film), there are limits to how far you can push the image in the camera before artifacts become visible or objectionable. After that, the recording format limits what you can do. And with most video cameras, the camera generally offers better colorspace and bit depth than the recording format. So for a desired look, you usually get the best results by tweaking the camera's DSP (or filter pack and film stock, in the case of film), and then use post for a final "polish."

 

That doesn't mean that you should try to do final color-correction in the field. Instead, if at all possible you should decide on the desired look ahead of time and pre-program those looks into the camera, just as you might choose a certain film stock, exposure, and filter pack with film. Of course that's not always possible, and you may end up trying to stretch the image in post. In that case, just remember that the farther you push the image in post, the more you stress the limits of the recorded image.

 

So if in-camera tweaking gives you odd color artifacts when attemting too much black stretch or knee compression, then you know you've reached your limit of what you can do well in-camera. Then it's time to back off some and do the rest in post.

 

Re: the original post; flatter gamma curves like "cine-like or "film-like" also tend to reduce color saturation (saturation is tied to gamma). When using one of these gammas, it's important to also change the color saturation to compensate.

 

Also, make sure the monitor you're viewing the footage on is properly set up.

[Alessandro Machi]

I used the word "low budget" to make a distinction. Perhaps I should have used the phrase E.N.G. as well. If one has time, the budget, crew to properly light, and an accurate monitor, and the FINAL DECISION MAKER ON SET for APPROVAL, than I'm not against making the image look as close to what is wanted. Should a DP be concerned, or at least be made aware about someone who isn't on set but who actually has the final visual say? Might make an interesting topic.

[Michael Nash]

I used the word "low budget" to make a distinction. Perhaps I should have used the phrase E.N.G. as well. If one has time, the budget, crew to properly light, and an accurate monitor, and the FINAL DECISION MAKER ON SET for APPROVAL, than I'm not against making the image look as close to what is wanted. Should a DP be concerned, or at least be made aware about someone who isn't on set but who actually has the final visual say? Might make an interesting topic.

 

 

Of course it's a gray area; where the bulk of the "look" should be created, and by whom. I don't think there's a one-size-fits-all answer. Music videos and commercials have been dealing with this for years, where the DP may not be present during the color correction. I think for the good of any project there should be one person who guides the look of the image all the way through, and if that's not the DP then it should probably be the director. But we know it doesn't always work out this way, and sometimes DP's have to delegate and trust their colorists (as is usually the case with episodic television).

 

I was just pointing out that the closer you get the image to looking "right" at every step of production, the better it will hold up in its final form. But how far you want to carry that concept on your project is up to you -- we're always making judgements on how far we can or should push something during production, and still be able to massage the final image into line.

 

With film and high-end telecine/color correction, you've often got so much raw information to work with that you can afford to shoot "clean" and manipulate the image heavily without too many artifacts. But with more compressed formats like DV and HDV, you know that color and contrast start to fall apart pretty quickly in post, so you want to limit what you have to do there. Higher-end digital systems like HDCAM are somewhere in between, with the camera outperforming the tape format. It's all about guiding the image where you want it to go, keeping it within the "margins" of the colorspace you're working with at any given step.