My First HD Feature
So . . .
I just finished a 24p feature in Phoenix, Az. (wow! working in my home state again was great - although I consider Phoenix a seperate entity from Arizona). Anyway, I thought I'd share some of my experience with the project from a newbies pov. This will probably take a few emails.
Being that I was an HD virgin and in fact have only shot high end video on a few occassions, this was all quite new to me. I was brought in literally about two weeks before production and the Director (who was also Exec. Producing as is often the case) was quite inaccessible to talk prep and style and so on, so I was a bit on my own to come up with a 'look'. I checked out the F900 package in Burbank (individual owner sort of working through Band Pro) and got a run through of how to set it up and do the daily maintenance stuff and how to go through the menus.
We shot 23.98p
We started out at table 5
Largely defaultish settings that I also went over with our very own Mr. Fairburn.
So with everybodies talk of what HD can't handle compared to film, and how to protect it, I came into the project with a very flat image in my head that I couldn't shake. I knew that wasn't what I wanted but . . . So we shot our first location and the footage all looked nice (on the 9"HD monitor and 13" NTSC with AJA down converter). An awards presentation in a theater - spot light, lots of backlighting on a dim audience, etc. I tried to keep everything within 5 stops and underexposed somewhere between 1/2 and 1 stop. Whatever.
On the way to the second location, still a bit confused by where I was headed with it all, it finally hit me to think, how would I want this next scene to look on film? "A stressful photo-shoot at a racetrack in the heat of the day in Phoenix, Arizona." I saw lots of contrast, saturated colors (lots of colors to feast on there with the racetrack ads, the cars themselves, the wardrobe etc. - why not punch it?) etc. and I thought - there's got to be a way to push that with this camera. I made a couple of phone calls and learned about Gamma and Black Gamma, etc. I lowered the Black Master Gamma to like -75 as well as some other settings and changed to table 3 (Mr. Fairburn's favorite :-) Thanks again Sean!) Voila - I had me some interesting looking footage.
The Director went nuts. All the talent were sweating to death (it was still fairly cool so it was only around 110 in the sun - fortunately the camera and I were under a carport where it was even coolor ;-} and the colors and contrast REALLY worked well with the scene. The highlights were blooming of course, and my instincts were telling me I wasn't protecting the video enough, but I though - gee this is too much fun and looks too appropriate to think conservatively so I underexposed around the same as before and went with it.
From that point on I spent the rest of the shoot going into each scene knowing where it fell in the narrative and where the characters were and of course what the location was and adjusted much of the settings to accomodate - including again the Gamma, Black Gamma and table (often #3).
When shooting video, I also like to manipulate the color with 'the old white balance through your swatchbook' trick, which I often relied on to specifically warm up or cool down a scene a little. In some cases, outside in the desert when the scene called for a warmer look, I went with an 81EF. I often used a 1/2 BPM (until it got broken on the second to the last day :_-( ) For some dreamy sequences I went with a Diffusion #4.
During the last week of the shoot, we all went and watched Ep II digitally projected . . .
Overall, it was definitely a FUN experience. I should have some stills if not video clips on my site in the next month and I'll post that when the time comes.
Sadly all I had to watch really was that small HD monitor and more often than not I still felt like I was just watching REALLY nice video, but I'm hoping I'll feel better about it when I can sit back and watch it on a normal size monitor - or maybe the 3:2 pulldown will give it a different feel for me - I don't know. My biggest frustration was the low-res B&W viewfinder. I operated this feature which had a LOT of camera movement (dolly, jibarm, handheld) so often times, all I could do was put the camera in a couple of the known positions, run over to the monitor and do my best to judge the whole scene. I've gotten spoiled to seeing everything through lens WHILE I'm operating.
Also, I know there was some discussion a few weeks ago about the necessity of an extension viewfinder. Many people commented that it was unnecessary and only a toy or an image object, so I went into this (without one) open minded. I must say - I definitely prefer to have one around. Maybe I'm doing something wrong, but I often find myself operating shots - with dolly/jib moves especially - where the camera and I end up looking like a pretzel for some portion of the shot and that extension viewfinder really comes in handy then. I definitely want one when possible.
I have more, but my daughter and her 5 sleepover friends are stirring now, so breakfast is in order - I'll revisit later.
One final note. I'm talking to some producers about another HD feature in the next few weeks and we are all talking about the Panasonic 27V. I would love to have the experience with both cameras! We'll see.
I am nearly finished with a feature shot in 24 p with the panasonic 27V. Shot one last year with the F900. Sort of like comparing a PD 150 and a canon XL1: I don't like either of them but I like the PD 150 better. In the case of HD i Like the 27 V better. Doesn't handle the highlights as well as the F900 without the Cine gamma mod but it does lots of other things very well, easier to hand hold and significantly better light sensitivity. I can do a better report after this is over.
-- Mark Smith
Oh, Seven films Inc.
143 Grand St
Jersey City, NJ 07302
That's because its Gained up +6 db when marked at 0 db
B. Sean Fairburn
That's not correct. Remember in 'Spinal Tap' when the guitarist was showing his amp. THE 'difference' in his amp was that it went to eleven (an actual Marshall amp). The director asked him why not just use an amp that goes to ten. He said "Because this one goes to eleven'. The eleven was nothing more than the designation for the amp; a reference. Sony's reference of sensitivity is not identical to Panasonic's reference, so to say '0' on one is equal to '+3' one another or whatever is not entirely accurate to say since both use a different scale, let alone different electronics to make that picture.
Producer, Director, Cinematographer
Hellgate Pictures, Inc.
444 E. 82 Street New York, NY 10028
That was one of my favorite scenes in Spinal Tap!
But you are correct. The only apples-to-apples way to compare sensitivity is to compare it at an equivalent signal to noise ratio and bandwidth. Since the bandwidth of the 27V is less than the F900 (pre tape, at least), this would mean using the bandwidth of the Panasonic for both. Another factor to consider is that the 27V is a 720 line camera and the F900 is a 1080 line camera; everything else being equal, the higher frame rate camera will always have less sensitivity...
Sean is referring to what many other people have noticed, including some of the rental houses that own both: that the Panasonic 27V at 0 db produces noisier images than the Sony F900 at 0 db, so one solution is to use the -3db or -6 db settings on the Panasonic.
This suggests that from a PRACTICAL standpoint, the difference in working sensitivity between the Panasonic and the Sony is not that different IF you want similar noise levels.
Cinematographer / L.A.
I thought that a db was a db. Electrical gain is electrical gain a matter how you look >at it.
It is but let's not confuse the term as used with sound. Here is a definitiont that might help:
In electronics and communications, the decibel (abbreviated as dB, and also as db and DB) is a logarithmic expression of the ratio between two signal power, voltage, or current levels. In acoustics, the decibel is used as an absolute indicator of sound power per unit area. A decibel is one-tenth of a Bel, a seldom-used unit named for Alexander Graham Bell, inventor of the telephone.
Suppose a signal has a power of P1 watts, and a second signal has a power of P2 watts. Then the power amplitude difference in decibels, symbolized SdBP, is :
SdBP = 10 log10 (P2 / P1)
Decibels can be calculated in terms of the effective voltage if the load impedance remains constant. Suppose a signal has an rms (root-mean-square) voltage of V1 across a load, and a second signal has an rms voltage of V2 across another load having the same impedance. Then the voltage amplitude difference in decibels, symbolized SdBV, is :
SdBV = 20 log10 (V2 / V1)
Decibels can also be calculated in terms of the effective current (amperage) if the impedance remains constant. Suppose a signal delivers an rms (root-mean-square) amperage of A1 through a load, and a second signal delivers an rms amperage of A2 through another load having the same impedance. Then the current amplitude difference in decibels, symbolized SdBA, is :
SdBA = 20 log10 (A2 / A1)
When a decibel figure is positive, then the second signal is stronger than the first signal. When a decibel figure is negative, then the second signal is weaker than the first signal. In amplifiers, the gain, also called the amplification factor, is often expressed in decibels. A circuit amplifies only if the decibel figure for the output-to-input power ratio (SdBP) is positive.
In sound, decibels are defined in terms of power per unit surface area on a scale from the threshold of human hearing, 0 dB, upward towards the threshold of pain, about 120-140 dB. As examples: the sound level in the average residential home is about 40 dB, average conversation is about 60 dB, typical home music listening levels are about 85 dB, a loud rock band about 110 dB, and a jet engine close up is 150dB.
Decibel units are commonly used in audio equalizers, both the hardware kind and the software kind, as a convenient reference point while editing. Boosting an equalizer band whose center point is 1000 by 3 dB means that you have raised the volume level of that frequency band by 3 dB as it relates to the other frequencies in the sound. A typical equalizer has a range for boosting or diminishing a sound level of +/-18 dB.
No, there are a lot of variables besides just gain... What signal to noise ratio is the camera manufacturer willing to accept, what is the pixel depth of the format (number of pixels), what bandwidth are the video channels, what are the CCD characteristics, how wide are the skirts on the prism (and how much matrix is required to correct for said skirts), what are the cutoff characteristics of the optical input filter; to mention a few. All of these will affect the signal to noise ratio and sensitivity and, directly or indirectly, what the manufacturer defines as the unity gain (0 Db) setting. For example, I can change the VA settings on a Sony F900/F950 and dramatically change the sensitivity and signal to noise even though the camera is still set at "0"... That is why the EI rating of a given video camera is much more transitory than that of a given film stock.