Hv20 Asa

[lordtangent]

I thought this might interest some of you. I borrowed a light meter from a friend for an upcoming shoot. Of course, a light meter is useless you know the ASA of th camera you are working with. So I rated my HV20.

I used the technique I learned in cinematography class for assigning an ASA to a video camera. This is how you do it:

1. Take an 18% "middle gray" card and point the camera at it, filling the frame with the card. (light the card of course)

2. Let the cameras auto exposure expose for the card. Lock the exposure. The camera should have exposed the card to look middle gray (that is the way all auto exposures work)

3. Now read off the aperture and shutter speed from the camera.
4. Use a light meter to figure out how much light there is.
5. Calculate the ASA using the info from the meter along with the aperture and shutter speed you got from the camera. (the scales on the meter will help you with that.)

There are only three variables to exposure: Aperture, shutter speed, and ASA. All you need to know once you have the aperture and shutter speed is how much light you were actually working with and you can figure out the ASA. This is the part where a light meter comes in handy. (or at least a digital camera with manual controls over all three variables... it has a light meter internally after all) I used a proper light meter. It's easier because it's kind of like a calculator that figures out the the missing variable for you. And actually, in this case the more low tech manual/analog type meter is better because the spinning scales are easier to read and adjust. Once you have a meter in your hands the procedure for figuring out the ASA with the shutter speed and aperture should be pretty obvious to you (it's going to be a little different from light-meter to light-meter, so I can't give a step by step on that part. It will have to remain an exercise for the reader...)

Anyway, the results are: I rated my HV20 at about 80 ASA in "Cinemode". It is not a very fast camera by any any stretch of the imagination. To keep it at 1/48th and f2.8 with no gain (exposing for the gray card) you need around 160 foot candles. That is actually a lot of light. Forget about using house lights to get a good result. Unless your actors are a foot or two away from them it's not going to be enough. You'll need movie lights to really get the lighting up to where you have enough to be useful. I was testing with a 1k broad today, and it was just enough to get a good exposure with the light 7-8 feet away from the talent. (Wide open at 1/48th sec)

It could be worse. At least you can replicate that on the cheap with "shop lights" They are about 500w each and they sell those 2 headed ones that come up to about 1k total.

My FLO lights are just barely powerful enough to be useful. They need to be really close to the talent (4 feet or so) to be enough to provide a proper exposure. I need to build some bigger flo lights!


Has any one else rated their HV20? I'm interested to hear your results!

[24Peter]

Why did you use F 2.8? The camera has a max apeture of F1.8. That's an additional stop + 1/3. I find normal 100 watt houselights plenty for my family stuff (though for my filmmaking efforts I use additional lights.)

[lordtangent]

Based on the way the test is done, any aperture will work as long as there is no gain or ND confusing the test. The camera would is still be 80 ASA. I used f2.8 to be sure I was not getting any gain and also, that is what the auto exposure settled on for the grey card. (I guess I could have continued to adjust the light down until I hit f1.8, but there is no point to that really. You just need a shutter speed and an aperture to figure out the ASA.)

I could open up more and shoot in less light. That is true. My 160 foot candle number is off in that case, because I forgot to take the extra stops into account when calculating the minimum foot candles. Taking the extra stops into account you only need about 60 foot candles before you hit gain. And of course, moving down the 1/24th of a second in with f1.8 you would need even less. Say maybe 30 foot candles. That is actually pretty low light. So yeah, the camera can still work in low light (with no gain) in spite of its slow rating. Still, it is slow. There are certain effects you can't achieve without using a lot of gain as a result.

[Duke]

Great information. Thanks,
Duke

[f-stop]

I did an ASA test a few weeks ago and the rating that I got was 64 ASA. Pretty close to your 80 ASA with only 1/3 stop of difference compared to your result.

Here's my reading with HV20 at 60i Tv mode:

Tv=1/15, f4

Incident light meter's reading:

ISO 64, Tv=1/15, f4

With the HV20 at 24p and Tv mode:

Tv=1/12, f4.8

[lordtangent]

Thanks f-stop!

Is there anyone else?

[hepabst]

I did a test today with my HV20 and I ended up with an ASA-25 outdoors, cloudy day. Camera 1/48 f5.2, meter 1/48 f5.2, asa 25.

[vince]

I had similar results as far as needing a lot of light with CFL's to obtain decent exposure and avoid gain. I was using 14 * 40 watts bulbs from about 6-8 feet and still seeing some gain (Tv at 60, Av ranging between 1.8.-2.4). These bulbs give soft light but don't throw far even with a fairly good reflector. Using an egg crate did not help put more light on the talent. Doubling the amount of watts only gives a few stops improvement. It is just enough for my interview settings but I'm curious how much light is needed to actually to have some room to play with?

[lordtangent]

I have used a 1k bulb in a china ball to light a room and that gave me just enough room for a little play on exposure.

Keep a fire extinguisher handy!

[hepabst]

I wonder why my asa test is so much lower than the others. Does the HV20's CMOS chip have a fixed exposure the way film does? How is a digital chip rated in this regard? The reason I've been looking into this is because I use a letus mini(test done without mini) and I want to understand the over all f-stop. Plus I want to start using a meter to understand my lighting setup without having to use the camera all the time.