OK, let me get a few things straight:

[Arbel Rom]

I have a few questions about the EX-3:

1. What is the ISO equivalent of the camera. At -3 dB gain, and at 0 dB?
2. Does anyone know of a good way to calibrate the brightness and contrast of the viewfinder so
they will represent what is recorded by the camera?
3. On what F stop is the lens sharpest?

Thanks,
Arbel.

[Olof Ekbergh]

In my testing 0DB is about ASA800 and -3DB about 600, others may give you closer values, but those are with my Gossen Luma Pro testing.

To Calibrate the VF or LCD, turn on SMPTE Bars and calibrate brightness so of the black bars on the bottom row only the lightest one shows a little. And make the color of the bars about right (from experience) a little extra saturation will show if skin tones are off. You have to tweak this as lighting conditions change, just like any monitor. It will be very different shooting a sunset or middle of a sunny day, or in a dark theatre.

f2.8-4 seems best, I try always to shoot between 1.9 and 5.6. f8 and up can get ugly.

[Arbel Rom]

Wow... ASA 600... this is very sensitive.
I was sure its around 400 or something...

Thanks!!

[Craig Seeman]

Don't forget sensitivity changes relative to shooting format.
1080i is the most sensitive.
1080p24, 1080p30, 720p60 (1080p25, 720p50 for PAL compatible thinking) are all different given frame rate and 180° shutter angle.

[Olof Ekbergh]

Good point Craig.

I used 1080 30p. My standard mode these days.

I think 1080i is about 1/2 to 2/3 f-stop faster.

Your PP will make an apparent difference as well. It is hard to come up with an exact way to use a light meter to set the EXcams. I find the histogram and zebra to be the best way. It is fun to check though.

[Andrew Stone]

Hold on a sec before someone ruins a shoot. There is no way the EX3 is equivalent to 800 ASA. Maybe if you did some crazy tweaking in the picture profiles you might be able to get it to feel 800.

I just did a quick internet search to see what various people have thought the EX3's equivalent ASA is and here's what I found:

I just figured it out, if shutter is off in 30P then shutter is 1/30. You can also set shutter as angle, I remember 11° was magic for sports in film (24p) I think it is about 1/800 of a second, you can go all the way up to 1/2000 of a sec. ASA on EX3 is 400 (or 320 depending on who you believe) at 0db gain. The 500 is 800 at 0db so EX3 is 1 stop or so slower...

-Olof Ekbergh

OK have a good laugh and I'll come back in a few minutes....




The last shoot I did, I took out my (pretty decent) light meter and took readings around the area I was shooting. The closest ASA range I got that was comparable on the light meter to the readings I was getting "in camera" was 250 ASA. I had a home brewed Picture Profile going so that would skew things.

I have read probably about 6 times over the past couple of years people pontificating on the ASA of the EX series cameras. The consensus from my recollection is around 320 ASA.

Just poking fun Olof. The internet is a cruel mistress.

-Andrew

[Craig Seeman]

I don't like linking off forum for answers but Adam Wilt, who does copious tests, has that information in an article he wrote.

ProVideo Coalition.com: Camera Log by Adam Wilt | Founder | Pro Cameras, HDV Camera, HD Camera, Sony, Panasonic, JVC, RED, Video Camera Reviews

and the relevant quote:
____________

Sensitivity

To measure sensitivity, I set the exposure of an 18% gray card at 50% on the waveform monitor with the camera in its default Standard Gamma 3. I metered the card with a Gossen Starlite and incident light with a Spectra Pro IV, and varied the ISO settings until the shutter speeds and apertures matched the camera (both meters agreed within 1/10 stop). I determined the sensitivity of the camera to be:

• ISO 400 in 1080p modes

• ISO 800 in 1080i modes (just as you’d expect: with dual-row summation, you get twice the sensitivity), and

• ISO 500 in 720p modes.

Apparently Sony is picking up some gain during downsampling to 720p, analogous to what happens in dual-row summation.

The camera can run at –3dB, useful if you want the lowest noise possible. The highest gain, +18dB, is noisy but not unusable; it’s comparable to the Z1 at +18dB or the HVX200 at +12dB. The noise is best characterized as colorful, fine grain: it’s perhaps a bit more colorful than noise from some other Sonys, but not annoyingly so.

[Perrone Ford]

Again,

Take into account shooting modes.

At 1080/24p I rate the camera at ISO400. Confirmed with over 75 shoots, and I meter EVERYTHING I shoot first

At 1080/60i I rate the camera at ISO 800.

At 720/60p I rate the camera at ISO 320.

All these at 0 gain. I tend to shoot at -3 most of the time so I adjust accordingly. I also have custom profiles I use, so that will have a bearing as well.

I have lately been comparing my EX1 to my T2i at the same F-Stops and relative ISOs, just as a means of comparison. My numbers hold up well enough for me to rely on them, and have for over 1.5 years.

YMMV...

[Olof Ekbergh]

My original 2 year old 400ASA reading was when I was comparing it with my DSR500, it was known to have ASA 800 rating (from Sony tech), I also used different PP's back then.

I just went and tested again with my Luma Pro light meter at -3DB.

And I get ASA 400 at 1080 30P.

I get ASA 800 at 1080 60i.

This also agrees with my shooting with my 5DmkII and 7D, I do that along with the EX3 and EX1R a lot these days.

But I would caution using just a light meter for exposure, I do find the best is a good eval monitor and scopes, or in camera the histogram and Zebra.

In the days of film a light meter is all we had. Now we have far more tools, but also the camera settings will vary the light sensitivities greatly.

[Perrone Ford]

Agreed Olaf. Fortunately film (at least print film) allowed more latitude than we have in these digital cameras. And frankly, I prefer to be under half a stop or so with these cameras than blow the highlights. I do find it comforting that your numbers and mine agree to within half a stop. The difference could easily be in profile choices. There is also variance in the F-Stops as there is quite a bit of play in the numbers before the electronic number changes. You can move the iris quite some ways and watch the screen get brighter and darker before you'll note any change on the viewfinder of F-Stop.

[Craig Seeman]

Perrone, your 720 number differs from Adam Wilt's. Why do you think? Adam specifically notes when comparing 1080p to 720p than there's an apparent sensitivity gain when the camera downconverts to 720p. Your number shows a loss of sensitivity.

[Perrone Ford]

Originally Posted by Craig Seeman

Perrone, your 720 number differs from Adam Wilt's. Why do you think? Adam specifically notes when comparing 1080p to 720p than there's an apparent sensitivity gain when the camera downconverts to 720p. Your number shows a loss of sensitivity.

Two things I can think of:

1. Are you talking about 720/24p or 720/60p like I mentioned?
2. I use a different picture profile on 720/60p than I do for 1080/24p

[Craig Seeman]

Originally Posted by Perrone Ford

Two things I can think of:

1. Are you talking about 720/24p or 720/60p like I mentioned?
2. I use a different picture profile on 720/60p than I do for 1080/24p

Adam says
• ISO 500 in 720p modes.
So his use of the plural seems to indicate that either he found no difference due to frame rate or he averaged.

In order to do his measurements he said he used the default STD3 gamma.

Note he mentions details about his testing procedure.

[Perrone Ford]

Well,

720/24p would be using a shutter of approx 1/48 at 180 degrees. While 720/60p would be using a 1/120 shutter. CLEARLY there is going to be a difference. My guess is that he tested 720/24p which is why he saw a sensitivity increase from 1080/24p. Same light, fewer sensors to excite.

I also do not use the STD Gammas. I use the Cinegammas. So that's another variance between his numbers and mine. And then my Cinegammas are tweaked.

I remember reading his article back around the time I was going to get my EX1. His review did a lot to move me along the path to getting the camera, as did Jody Eldred's work. Best camera decision I've ever made. And relegating the DVX100 to the second best camera decision I've made.

Getting this T2i may take the crown now though.. Canon is practically giving them away!

Originally Posted by Craig Seeman

Adam says
• ISO 500 in 720p modes.
So his use of the plural seems to indicate that either he found no difference due to frame rate or he averaged.

In order to do his measurements he said he used the default STD3 gamma.

Note he mentions details about his testing procedure.

[Craig Seeman]

That's for clearing that up. It's important for people to know how pixels as well as shutter speed impacts results. Yes is obvious to veterans but there are certainly newbies about and they wander off and start quoting posts . . . .

Canon's having an exclusive trade show in NYC in a couple of weeks, apparently sponsored by B&H. I'm curious about the T2i but we have all these old Nikon lenses.

I'm also curious to see where Sony goes with the NEX-VG10 and the inevitable "pro" version. Alas that's OT though.

My gut is telling me to wait for Sony given, as you allude to, how much Sony got right with the EX.