Shooting 60P

[Michael LaVoie]

Was wondering if you're shooting 60P and you're previewing the image in real time on a monitor, how can you get an accurate reading from the monitor if there's loss occuring from the 60P that you're not seeing untill you playback? What is the loss approximately? If I was shooting at a 4.8 in 24P, what would I open up to in 60P? Or is it the same as in film?

[David Mullen ASC]

If it looks correctly exposed on the monitor, and the monitor is set-up properly, then it's exposed correctly. You'll find just as with a film camera, changing the speed from 24 fps to 60 fps loses about a stop and a half, just as changing the shutter speed from 1/50th to 1/100th would lose you a stop of exposure. But you'd see the chance in brightness in the viewfinder / monitor once you changed the frame rate.

[Michael LaVoie]

My problem isn't with the Varicam, I just thought it might be too early to ask this question in the HVX forum. I had a greenscreen shoot with the HVX-200 at 60P. The director wanted a very contrasty look on the subjects. I rated the camera at 500 I used a CRT monitor and my light meter and both readings suggested to me I was exposing properly and yet the folks who were keying the image later complained that their desktop vectorscopes said the footage was a little underexposed. Not to any unusable degree but it was lower than they would have liked. I'm wondering if this is caused by lighting the subjects less evenly for a desired look? Did that throw off the editors?

 

If not I'm wondering what may have caused this. The monitor was calibrated properly and my meter is not off. I was assuming light loss from the 60P that didn't show up on the preview. If that's not the case than perhaps I rated the camera too fast for greenscreen but then why did it look good on the monitor live? I had the shutter off on the camera. Any ideas?

[Michael Most]

My problem isn't with the Varicam, I just thought it might be too early to ask this question in the HVX forum. I had a greenscreen shoot with the HVX-200 at 60P. The director wanted a very contrasty look on the subjects. I rated the camera at 500 I used a CRT monitor and my light meter and both readings suggested to me I was exposing properly and yet the folks who were keying the image later complained that their desktop vectorscopes said the footage was a little underexposed. Not to any unusable degree but it was lower than they would have liked. I'm wondering if this is caused by lighting the subjects less evenly for a desired look? Did that throw off the editors?

 

I think you're taking the comments of non-camera trained post people far too seriously, and trying to relate it to what you saw and what you wanted. "Underexposed" compared to what? My guess is that what you wanted and what they expected are two different things, not an uncommon occurence in the translation of images from production to post, especially in video. As far as the greenscreen, if the green information was well separated from the foreground elements, was lit evenly, and is at a decent video level, there shouldn't be any problem, regardless of how the post people might want to complain about it.

 

My only question to you would be whether you had access to the surrounding footage that the composite was supposed to be cut with. This would determine an overall light level that they might be trying to reach in the composite, but might be a bit "hotter" than what you shot.

[David Cox]

Are you able to post an image of your shoot?

 

My guess is that the post guys were moaning about the green screen being too dark. If this is the case, the separation between the green screen and the foreground subjects is limited and its this that the software looks at to create the key. The more limited the separation, the worse the key is (or the longer / harder the post guys have to work to make something of it). The reason is that if you imagine a very dark green, you can imagine that in terms of RGB level it might have a green level of 10% or so. Your foreground people might have a green value of 3% to make up their skin tones, so you have 7% of separation. If your green screen was 100% green (RGB green level), you would have a 97% separation.

 

The age old problem is how to brightly light the green screen and leave the foreground people contrasty. Distance to the green screen and using a grade to restore contrast after deriving the matte are both common techniques.

 

But see if you can post an image of your shoot and I'll tell you if the post guys were just moaning cos their one-press green keyer wasn't up to a little challenge :-)

 

David Cox

www.baraka.co.uk

[Michael LaVoie]

I don't have an image to post yet. I rated the camera at 500 and lit the background with three kinos and got a F4 pretty evenly all around. I lit the subjects with tungsten and just barely got a 5.6 on my key and a 2.8 7 on my fill and a 2.87 on my backlights only they were coming through thinner diff so they looked hotter.

 

I went from a 4 to a 4.8(camera fstop) depending on the shot. I checked my meter and it's fine. Both monitors looked almost overexposed. Yet the vectorscope readings were low. Is it that I should have rated the camera at 320? Should the greenscreen be lit above key?

[Christopher Bell]

Did you set up the monitor with bars? An improperly set up monitor can lie to you. I like the Panasonic LCD monitors as they have a built in waveform. It's a great tool to make sure you are in the range.

 

I tend to shoot a lower contrast image for acquisition, then dial in the contrast later. That way the client can get the look they want.

 

Chris Bell, Seattle

[Micah Ellars]

If it looks correctly exposed on the monitor, and the monitor is set-up properly, then it's exposed correctly. You'll find just as with a film camera, changing the speed from 24 fps to 60 fps loses about a stop and a half, just as changing the shutter speed from 1/50th to 1/100th would lose you a stop of exposure. But you'd see the chance in brightness in the viewfinder / monitor once you changed the frame rate.

 

 

With the Varicam I thought it recorded 60p all the time and simply flagged the frames you need based on what speed you had set. If that is the case, then how does the varicam suffer additional light loss from overcranking? I may be way off here, but perhaps someone could explain how it works exactly. Thanks,

 

Micah

[David Mullen ASC]

With the Varicam I thought it recorded 60p all the time and simply flagged the frames you need based on what speed you had set. If that is the case, then how does the varicam suffer additional light loss from overcranking? I may be way off here, but perhaps someone could explain how it works exactly. Thanks,

 

Micah

 

It always RECORDS at 60P but it shoots (captures) at the frame rate you select -- the extra frames are added after capture to add up to 60 for recording. So capture rate does affect exposure time per frame.

[Micah Ellars]

Thanks David. I appreciate you clearing that up for me.

[Micah Ellars]

It always RECORDS at 60P but it shoots (captures) at the frame rate you select -- the extra frames are added after capture to add up to 60 for recording. So capture rate does affect exposure time per frame.

 

So does that mean that there would be a correlating amount of light gain when undercranking? A stop and a half gain @ 12p?

[David Mullen ASC]

Yes, shooting at 12 fps on the Varicam with a 1/24th shutter is one-stop more exposure than 24 fps with a 1/48th shutter.

[Will Novy]

Yes, shooting at 12 fps on the Varicam with a 1/24th shutter is one-stop more exposure than 24 fps with a 1/48th shutter.

So i take it david, 48 frames a second at 96 shutter is one stop less? Or am i way off?

[David Mullen ASC]

So i take it david, 48 frames a second at 96 shutter is one stop less? Or am i way off?

 

No, that's correct -- 1/96th is one-stop less exposure than 1/48th. As you can see, it's the SHUTTER SPEED that determines exposure, whereas the frame rate only matters in that it affects the shutter speed (i.e. at 24 fps with a 180 degree shutter angle, the shutter speed becomes 1/48th of a second.)

 

Now with most video cameras, you can turn off the shutter, thus at 48 fps, for example, you can have the frame exposed for that entire period, i.e. 1/48th of a second (it's not physically possible to take 48 pictures per second and have a per-frame exposure time longer than 1/48th, not without dropping the capture speed.) However, you get twice as much motion blur over the standard 180 degree shutter used by a film camera, so the motion looks smearier than is possible with a film camera at that frame rate.

[Will Novy]

No, that's correct -- 1/96th is one-stop less exposure than 1/48th. As you can see, it's the SHUTTER SPEED that determines exposure, whereas the frame rate only matters in that it affects the shutter speed (i.e. at 24 fps with a 180 degree shutter angle, the shutter speed becomes 1/48th of a second.)

 

Now with most video cameras, you can turn off the shutter, thus at 48 fps, for example, you can have the frame exposed for that entire period, i.e. 1/48th of a second (it's not physically possible to take 48 pictures per second and have a per-frame exposure time longer than 1/48th, not without dropping the capture speed.) However, you get twice as much motion blur over the standard 180 degree shutter used by a film camera, so the motion looks smearier than is possible with a film camera at that frame rate.

Interesting information. Ill have to remember that later on. I thought that both the shutter and frame rate determined the exposure. But since it seems the shutter determines it, its a interesting way to increase, lower exposure for tricky shots without messing too much with the source of light.

[David Mullen ASC]

Interesting information. Ill have to remember that later on. I thought that both the shutter and frame rate determined the exposure. But since it seems the shutter determines it, its a interesting way to increase, lower exposure for tricky shots without messing too much with the source of light.

 

Frame rate + shutter angle affect shutter speed (i.e. exposure time). But ultimately, it's the shutter speed that matters (and the f-stop, gain setting or film speed, etc.) In other words, if you run the video camera at 24 fps and a shutter speed of 1/48th and then switch to 48 fps but keep the shutter speed 1/48th, then the exposure time is the same.

 

With a film camera, you generally cannot have a bigger shutter angle than 180 degrees, or 200 degrees on the Panaflexes, so 48 fps with a 1/48th shutter speed is not possible.

[Lyle Norton Vincent]

My problem isn't with the Varicam, I just thought it might be too early to ask this question in the HVX forum. I had a greenscreen shoot with the HVX-200 at 60P. The director wanted a very contrasty look on the subjects. I rated the camera at 500 I used a CRT monitor and my light meter and both readings suggested to me I was exposing properly and yet the folks who were keying the image later complained that their desktop vectorscopes said the footage was a little underexposed. Not to any unusable degree but it was lower than they would have liked. I'm wondering if this is caused by lighting the subjects less evenly for a desired look? Did that throw off the editors?

 

If not I'm wondering what may have caused this. The monitor was calibrated properly and my meter is not off. I was assuming light loss from the 60P that didn't show up on the preview. If that's not the case than perhaps I rated the camera too fast for greenscreen but then why did it look good on the monitor live? I had the shutter off on the camera. Any ideas?

 

 

The underexposeure problem may be the result of rating the camera at 500 ASA. I recently shot a short film with this camera and rated my meters at 160 ASA. During prep I determinded this using a macbeth chart and a 18% gray card. The menus were basically zeroed out and the gama was set to dynamic, so I think that this is the highest ASA this camera is capable of with gain set to 0db. I found this to be a real drawback to this camera, especially for night int/ext. We usally use these type of cameras when the budget is low and therefor we don't have a lot of lights. Another big problem is the built in lens goes from 1.6 at its widest to a 2.8 when in full telephoto. So if I'm lighting a scene and we have puchins and CUs ect then I have to light the whole scene at a 2.8 unless I want to do relighting. When I shoot I like to move fast and not have to totally relight for every shot. Plus when working in these small formats I want to be on the long end of the lens anyway. I guess this low ASA is from craming so many pixels onto a 1/3" chip.

Edited May 11, 2006 by LVDP

[Paralax]

I realize this is a VariCam Forum, but as the HVX-200 is a small VariCam, I'm hoping the question is still relevant. I'm about to shoot gun-and-run circumstances (BRoll & Interviews) in the U.S. and the U.K. and am strongly considering using the P2 at 720/30P in order to have variable frame-rate options, but I'm concerned I'll run into a shutter speed issue with available practicals in the U.K. 720/30p does not allow for a shutter of 1/100, which is what I believe I require to avoid strobing in these circumstances. Any insight will be greatly appreciated.