Does anyone know what the EI/ASA/ISO is for the HV20? Or how you would determine it?

Bonus round - and once you do know it how do you effectively use it when lighting a scene?

Thanks!

Rob Robinson

I found this...


1. Frame up a gray card.
2. Activate the camera's auto iris. It will adjust the iris to provide an exposure that's equivalent to middle gray. Some cameras allow you to add over or under exposure compensation to this function so make sure you normalize this setting first.
3. Set the appropriate shutter speed on your light meter. Hold the meter against the gray card and face the white sphere towards the camera. Read the light level.
4. Adjust the ASA setting on the meter until the iris reading on the meter matches the iris reading on the camera. This is your ASA.

So can someone (who has a gray card) do this?

The problem with the HV 20 and finding a ISO level is that the camera in TV or locked shutter mode, does not display F-stops. This makes it difficult to find the measurement in conjunction with a meter.

The problem with the HV 20 and finding a ISO level is that the camera in TV or locked shutter mode, does not display F-stops. This makes it difficult to find the measurement in conjunction with a meter.

Try half pressing the photo button. It should show on the display.

Try half pressing the photo button. It should show on the display.

Yes this will work for sure (I think I was the one who first discovered this):

http://www.dvxuser.com/V6/showpost.php?p=930794&postcount=139

I would expect to lock the shutter for this test (TV mode) to 1/48 and then use the PHOTO button and joystick to toggle the various iris settings to get the F-stop of the camera.

Thanks for the tip. I will try this soon.

That does work, but ONLY if you have a mini-SD card installed. With no card present, you get no shutter/iris display, it only shows a red "no card" icon. But if you have a mini-SD card in, half-pressing the photo button will show you the current shutter and iris. And yes, it does work after locking the EXP, so you can adjust the dial and check and re-check your current shutter & iris by doing this.

I have a miniSD card on the way. So did anyone find the EI of the cam yet?

I have a miniSD card on the way. So did anyone find the EI of the cam yet?

Thanks Brian, btw what light meter do you plan to use? I'm looking to get one. Also what "gray card" will you use (who makes it - product #, etc..?)

Thanks!

Rob Robinson

I purchased a Gossen DigiPro F light meter and created an 8.5" x 11" 18% gray card (from my printer with RGB of 119,119,119) and then lit a scene such that the aperture on the HV20 ( in TV mode and EXP at +0 ) read 2.8 with a shutter of 1/48. Then I adjusted the ISO setting on the light meter until it matched the iris/shutter of 2.8 and 1/48. The ISO setting that matched was 1600. Does this sound right?

No, that is definitely not right. I did a quick test with my Nikon D50 as a light meter and it looked like the number should be more like iso 200 for cine 24p at 1/48 f2.8. thats really really estimated though. My guess is somewhere between 100 and 300. Probably closer to 100. I've been meaning to do a real test but i havent had time nor do I have all the items I need to do an accurate test. I might be able to do it a bit more carefully but itll probably still just be in relation to my DSLR's iso ratings.

I'm clearly no expert on these matters, but wouldn't it be reasonable that the D50 has a different light sensitivity than the HV20? One of the claims of the HV20 is that it performs well in low-light situations. Why wouldn't 1600 be acceptable?

I may go and buy a "real" gray card tommorrow and try the test again. Although I can't imagine it making that much of a difference - but maybe...

EDIT: I just re-read your post and I think I understand now what you are saying. A "proper exposure for 1/48 2.8 *is* an ISO of 200 (100-300). So if I had to set the meter to 1600 then something must have been wrong with my setup?

Anybody else have a gray card and a light meter to test this?

Thanks!

Rob Robinson

The HV20 certainly doesn't have much sensitivity. When people talk about "low light", the question has to be based around not only sensitivity but also grain/noise. If you lock the HV20 at 0dB of gain, it looks like it's probably three stops darker than a DVX at 0dB of gain, which would put its ISO in the neighborhood of about 100.

If you just point the camera at a dark scene, recognize that the auto gain is going to kick in, and it can kick in ferociously. I've seen it at 27dB.

So, what's the definition of low light performance? Is it "brightness at any cost"? Or is it "brightness level at a rational noise level"? Or is it "brightness with 0dB of gain"? Once you narrow down what definition you are comfortable with, we can get a better answer. The definition I normally use is "brightness at 0dB", and by that measure my eyeball/yardstick guess is about 100.

So, what's the definition of low light performance? Is it "brightness at any cost"? Or is it "brightness level at a rational noise level"? Or is it "brightness with 0dB of gain"? Once you narrow down what definition you are comfortable with, we can get a better answer. The definition I normally use is "brightness at 0dB", and by that measure my eyeball/yardstick guess is about 100.

I like "brightness at 0db" definition the best - seems more objective. My goal in finding the ISO of the HV20 is so that I can then set the light meter to then check light settings for a scene knowing that I want to use a fixed shutter with an aperture sufficient enough so that I don't need any gain from the camera. For example what EV do I need from the meter in order to have a shutter of 1/48 and an aperture of 2.2 (mid-zoom) and no gain. If I find that the EV on the meter is too low then I'll add more light, etc... But in order to any of this reliably I need to know what the ISO of the camera is first. And finding that seems to be elusive at the moment...

Rob,

The meter I use is a Minolta Autometer III, which has been properly calibrated by a meter company. The gray card I use is made by Kodak and is 18%.

Once my sd card arrives I will do a test with the HV20. Hopefully I can test this weekend.

Your rating does sound a bit high at 1600. It may be that your meter is off a bit.

The last camera I rated was a standard def JVC DV300. This cam has 3 1/3" chips. The rating I found was between 800 and 1000 ASA or EI. I found this cam to be as sensitive or even a bit more than the PD-150 which is known to be a sensitive cam.

I do not expect the HV20 to be this high considering its single chip and high res.

Will keep you posted.

The ISO setting that matched was 1600. Does this sound right?


No way. My guess is ~200ASA. I shot some color chart, gray card and resolution charts the other day.

This wasn't an 'official' test as I didn't have my light meter with me and I was shooting them under available room lighting. However, I was shooting them in a room that I'm familiar with. It has no windows and I've always gotten around a 2.8 at 200ASA in this room with its available lighting. When shooting these tests with the HV20, the f-stop results were similar at 1/48th.

I'm going to do a shoot out between the XL2 and the HV20, just for kicks. At that point I'm going to also do 'real' exposure and latitude tests in multiple modes with a variety of contrast settings. However, due to my work schedule these tests are up in the air right now, so feel free to start without me :)

I just performed an ISO test with the HV 20. At 24p 1/48 F2.0, the camera is achieving 120 ISO. This seems right considering its small chip and high sensor count.

I just performed an ISO test with the HV 20. At 24p 1/48 F2.0, the camera is achieving 120 ISO. This seems right considering its small chip and high sensor count.

Thanks Brian! I broke down and bought a Delta 18% gray card (and threw my homemade card in the the trash). I was careful to use a room with no windows and a single light source (100W bulb). I framed the card with a 50mm lens at a few feet from the wall. I then used your 120 ISO setting and selected the 1/48 shutter speed as well on the meter and on the camera. Low and behold the aperture on the camera read 4.0 and so did the meter! Excellent! I then tried ISO 100 and ISO 160,200 and I could tell that 120 was the indeed the sweet spot. After that I looked at the EV on the meter and I was reading anywhere between 9.0 (for 2.8) and 10.0 (for 4.0) so I think now I have something that I can work with for lightling up a scene. Bascially I'll light the scene, then run the meter around and check for EV values within range. Very cool!

Thanks again!

Rob Robinson

No problem. It's nice to know we can now use this finding to properly set a scene. It's just too bad that we now have to compensate for the low sensitivity by using so much light. When running the test I had to place my 20w flourescent source 1.5' from the gray card to get 2.0. Looks like it's back to tungsten for me.

No problem. It's nice to know we can now use this finding to properly set a scene. It's just too bad that we now have to compensate for the low sensitivity by using so much light. When running the test I had to place my 20w flourescent source 1.5' from the gray card to get 2.0. Looks like it's back to tungsten for me.

Now that we're armed with the ISO/ASO rating for the camera (120), and knowing that in order to avoid adding gain we need a minimum aperture of 1.8 - 2.8 (depending on the amount of zoom), can we determine what a minimum lighting setup would be? Looking at this table:

http://www.cinematography.net/Pages%20GB/exposure.html

It would appear that with an ASA of 120 and a T-stop (likely close enough to F-stops) of 1.8 - 2.8 would require 32 - 80 foot candles of light. So the next question is how many and what configuration of lights would get us to that range? I lit a scene last night with a lowel 500W omni (key) and a lowel tota 750W (fill) and I had to bring the lights within 5-6 feet of the subject in order to get the EV to 9.0. What I'm wondering is knowing what options I have (changing key/fill configs) and what types of lights I can use (higher wattages for example) to get the job done. What would be good to know is how many foot candles do these (and other lights) generate and what are the real-word parameters needed to know in order to calculate these setups (distance, beam widths, etc...) in advance.

I know this is a complex question but if anybody has pointers as to where to start looking for answers they would be appreciated!

Thanks!

Rob Robinson

To recap, the (un)official ISO rating for the Canon HV20 is 120 IS0? Has anyone else discovered anything different (between 100-200 ISO)?

And, Rob, any new news? :D

I have actually found mine to be 80 ISO.

The meter in the HV20 agrees very closely with a hand-held meter set to ISO 100. It also agrees with my SLR's meter, set to ISO 100.

However, the "normal" exposure of the HV20 is a tad overexposed for my taste. So for my purposes, I'd rate it at ISO 80. If you don't mind the way the metering works on the HV20, then you'd rate it at 100 exactly.

Ben, if it's overexposed normally, which corresponds to iso100, wouldnt you want to rate it higher to get correct exposure? like iso120? meaning less light would be necessary to get the correctly exposed image?

I actually shot a commercial this weekend and shot the first few takes with my HV20, and the last few with 16mm kodak vision2 film. Maybe im not as used to film as id like to be, but as the built in light meter of the canon scoopic camera wasnt working and i wasnt able to borrow a spot meter in time for the shoot, i exposed the film assuming about an iso120 of my hv20. basically i shot the commercial wide open both on the hv20 and iso200 and iso500 16mm film. im a bit new to film, but even though my hv20 was underexposed by about a stop (which led me to assume the 200 should be exposed about right). yeah, not a science at all in this case, but i didnt have much time or the resources to avoid guessing. anyway, i was surprised how well the hdv from the hv20 held up against film. I mean there was no comparison in terms of depth of field and color rendering, but i was a bit surprised by latitude and resolution.

I always wonder about this stuff, i mean if these cameras have similar gamma curves, but film has more overexposure latitude and digital formats have more underexposure latitude, is it really right to be applying film sensitivity speeds to digital cameras? in this situation, the iso500 film was quite grainy, especially in the dark areas, whereas the hv20 was a bit "underexposed", but could be pushed up very easily and remained clean in the dark areas. it's likely the graininess of the film was from underexposure, but if film handles underexposure so poorly, maybe its not right to be rating iso speeds of digital cameras based on whatever the autoexposure function of the camera says is alright... if film stocks had their iso rated for only like 3-4 stops of overexposure, we'd be rating what is now iso500 film closer to iso64 (yeah just a guess, but you get the idea). i admit i dont know that much about film so perhaps im missing something, but i thought id share. i also did some low light tests with my hv20 and it didnt do too poorly in very low darkness and getting pushed up a few stops. it does amaze me how little noise it renders in dark areas.

Whoops! You're right -- I rate it at 120, not 80.

You're also right on when you question ISO ratings... The truth is that for both film and digital, ISO ratings are determined by engineers, and that the ratings themselves are compromises between numerous factors.

In the film world, slide/E6 ISO ratings are the most objective, because a small under or over exposure has disastrous results. Yet still, some people underexpose their E6 materials by as much as a stop, to retain highlight details. But negative/C41 materials are much more flexible, because an additional step is needed to interpret the negative into a positive... So you find photographers rating their C41 faster or slower (usually slower, to get more shadow detail and smoother midtones), based on the look they want. The ISO on the box becomes a timid suggestion.

Digital is more complex, because it acts like a slide (you can see it in positive, immediately), but it holds information more like a negative. Really, the ISO rating of any digital camera is based on some agreements that a bunch of engineers made last century. They decided how much highlight clipping and shadow noise is acceptable, and came up with a very generic transfer function (~gamma 2.2) to get middle gray where they wanted it.

If you decide you want to hold on to more highlights, you can rate your digital camera slower. However, this usually (but not always) means you're pushing your blacks harder, to give you more shadow information. This translates directly to more noise... But if you've worked with C41 before, you're probably used to some noise/grain in the shadows... :) You just have to be able to look at the LCD and realize that you're not stuck with that overly dark interpretation...