T351 Summer - Week 4

Lecture & lab today:

• Advanced editing concepts
• Cables and connectors (WooHoo!)
• Digitizing / Color Sampling / Codecs
• We'll finish editing the Interview/Feature story projects and review them.

Reality Check

• Final Projects: Your initial Final Project proposal and treatments are due this week. I want to meet with everyone about their projects on Thursday. You need to have met with me about your project and have turned in your pre-production materials before you begin shooting. Revised treatments and scripts must be turned in this week. You have until 5 pm Friday (emailed Word docs, PDF, or RTF are fine) to turn them in. (You can also leave them in the T351 box in the Production Lab, or in my mailbox on the 2nd floor.)
  • Kinsey project - If someone wants to take on a short informational piece on the Kinsey Institute, contact Jenny Bass
• Drama/Storytelling projects: Each team must turn in a working treatment and script by tomorrow. If you don't have it yet, then you need to meet with your partner and write them. Remember, your projects don't have to be long- just have a clear story. And remember the #1 ingredient for a story is conflict followed by ultimate resolution. It's important to show motivation of characters and believable action.
  • Each team needs one pre-production packet. They will edit the projects together and turn in separate critiques. If someone is having "creative differences" with their partner, they can make their own edit.
  • Finished treatments are due by the end of lab today or by noon tomorrow along with your script.
• Wednesday - No lab. You have all day to work on Drama/Storytelling projects
• Thursday - Lab devoted to completing and reviewing Drama/Storytelling projects

Misc Notes:

Logs: Some don't understand what a log is. Footage logs correspond to reels. They identify where the shots, what they consist of, and if they're any good. On any sizeable project they are worth every second of time you invest in making them. Here is an example of one from my Elkinsville project. (doc & pdf) Here's a blank log in a Word doc form that you can modify.

Procedures for shooting

• Rule #1: Always select the proper filter (For cameras that have filter wheels)
• Then either white balance or use the preset

Do you have a new tape?

• Set timecode accordingly
• Record bars for 30-60 seconds
• Remember to shoot at least 5 seconds of preroll and post roll.

Are you partway through a tape?

• Position tape where it needs to be (This is why we always record 15 seconds or so of nothing (black, etc.) when finished shooting and only partway through a tape.)
• Make sure timecode to set to regen
• Play the video and it should pick up the timecode

No matter what – always turn on color bars and look at your viewfinder or monitor before shooting. If it’s off then adjust it accordingly.

Production Values - Each of you should know what good video looks like and what good audio sounds like. It is expected that you make your remaining projects look and sound as good as possible. Good video calls for more than good focus and proper exposure. it is selecting dynamic shots that tell your story in the right order.

Lighting - Unfortunately we can't usually rely on natural light. When shooting indoors, it is almost always necessary to place some sort of soft key at a low angle. Overhead fixtures produce drab-looking results. Sometimes reflectors are enough to add a bit of punch.

TIP: As in making graphics, squint your eyes so the image blurs a little. WHere are the highest points of contrast? Ideally, these should be your subject.

Iris - If really unsure about setting the aperture, press the auto iris button.

Continuity / Blocking - Remember the 180 degree rule, editing on action, avoiding jump cuts, etc. Whenever you're shooting a scene people expect continuity. You can't cut on action unless you plan for it (put it in the script) and shoot it from two angles.

Editing 102

Readings:

• Review cybercollege unit 57 (Understanding timecode)
• Review cybercollege unit 58 (Off-line and on-line editing)
• Ken Stone's Final Cut Pro pages (optional - great collection of tips)
• Collected video tips from Stephen Schleicher(optional - some good tips)

The concept of advanced editing implies an inherent need to work efficiently. Post-production editing suites can cost a thousand dollars an hour to rent. Both producer and editor are under the gun to work quickly and efficiently.

The producer shouldn't be in an edit session wondering what shot to use. Most major decisions can be planned in advance and made outside of costly studios or editing suites.

Remember: It's totally possible- and highly recommended- to plan every shot in advance. This is what scripts and storyboards are for.

Outside of planning, we can save money and time with proper project management techniques.

Project Management

Pre-production - Keep the objective and target audience in mind when developing the treatment. Start with and refine the treatment. Once you have the story flowing properly in the treamtent, then you can move to writing a script.

A strong script and a well-developed preproduction plan is your key to a sucessful production. When working on any sizable video project (especially for paying clients), always focus on making a bulletproof script. This is a script that has been carefully scrutinized and has been checked off on again and again. If it isn't in the script it isn't in the video! This way any changes after the script has been approved are billable.

It's also important to develop detailed production schedules. Consider the most efficient way to shoot a production. It's typically not in the order of the script. It also help to take the script and all of the various locations and develop shot sheets. These tell you what shots you need at any given location.

Production

• Take care to properly set the time code at the start of each reel and record at least 30 seconds of color bars.
• Make sure there are no breaks in either the video or in the time code.
• Be sure to always shoot at least 5 seconds of B-roll before cueing action/talent and allow for post-roll at the end of each shot.
• Carefully and consistently label your reels as you shoot.

Pre-production - After the shoot, the video producer often needs copies of the source reels so that the producer (or their PA) can log the footage outside of the costly studio environment. This way he or she can select the right takes without taking up an edit room.

Window Dub / Timecode print

One of the first things you should be able to make for your producer is a window dub or timecode print. This is a copy of a videotape which has the timecode numbers superimposed on the screen. (I like to give my producers a VHS window dub corresponding to each reel)

The best way to make this is with the VTR. This way undesirable footage never has to be brought into the Avid or FCP editing system. You need a professional VTR to do this. Most consumer decks don't have the ability to display timecode numbers on the screen. If you have a professional VTR with this feature, simply make a dub onto another tape (eg. VHS) with the time code display turned on.

You can also do this in FCP. But all of the footage must first be digitized into the system.

To make a window dub in FCP:

• Place the source reel footage into a new sequence.
• Under Effects, select Video Filters, Video, then Timecode print
• By default, the effect create a TC generator, you need to go to the "filters" tab and change the generator selection to reader.
• With a fast enough system, you shouldn't have to render this effect
• Also note that you can use this filter to output a timecode print of your master sequence as well.

Graphics - It also helps to have a graphic style sheet that you will use when making graphics. Just as you don't want to search for lost shots while editing, you don't want to spend time wondering what color the graphics should be or what font to use. Test and try fonts and other visual design elements ahead of time. Prepare graphics in advance. (I usually make a few mock-ups of different graphic templates to run by my client before editing)

Ideally when you go into the first edit session you should have:

• Target audience, objective, tratment and other pre-production materials close at hand
• Script
• All reels clearly labeled
• Footage logs - one log for each reel
• Clear vision of graphic design

I usually have a folder containing all of this which lives in the box of tapes for any particular project in the edit suite. This way the other editors or PAs have access to it at all times. My master project folder containing all of this plus the signed releases and anything else gets filed under my client's name in my file cabinet

I also maintain an electronic folder that contains the proposal, treatment, script(s), footage logs, text for lower thirds and credits, computer graphics, animation files, and any other files associated with the project. This gets backed up onto disc and saved somewhere safe.

Assuming our project management is under control, we can move on to specific techniques. (We can review the shortcuts for these during lab this week.)

Logging and digitizing footage:

Identifying the reel in the editing application is key to succesful, worry free editing. If you follow the proper procedures while logging, accidents such as crashed hard drives will turn into minor annoyances, not major catastrophes.

(The hard drives in WTIU crashed right after I had digitized most of my Elkinsville footage and after I'd spent a week editing. We were able to rebuild the project in a single day.)

When logging or digitzing be sure you carefully name your reel- it'll correspond to what's printed on the reel, right?

• Elkinsville reel 1
• Elkinsville reel 2
• Etc.

The basic options in FCP are log clip or capture clip, capture now or batch capture. When working on longer-form programs, it's often easier to batch capture clips after they've been logged. You can either import third party log data or create it from withing Avid or FCP.

To do this in FCP, mark the in and out points of a clip, name it, and then move forward to the next clip. This will create clips in your browser without the associated media. When you are done logging, select the clips in the browser and then "batch capture". If you do this you can capture them all while you get a cup of coffee.

This will not work if you have breaks in your time code.

Importing media

Always be sure to copy any media you want to use in your project into your master media folder. Do this before importing the media. If you import directly over the network, or from a mounted ZIP or CD, FCP will expect that drive or disk to be mounted in order to access your media.

So if your client hands you a CD containing their new logo, what will you do with it?

Using vectorscopes, waveform monitors and TBCs:

Video signals can be broken down into two components: luminance and chrominance. Luminance is the brightness component & chrominance is the color component.

A good editing suite will have a vectorscope and waveform monitor set up, so that the video levels and color can be objectively monitored. It's easy to make graphics in Photoshop too bright, but if you keep your eyes on the waveform monitor, you can tell when the signal reaches 100 IRE.

Waveform Monitor - A device used to examine the luminance portion of the video signal and its synchronizing pulses. The scale starts at –40 – goes to 0 then up to 120 IRE Units (IRE = Institute of Radio Engineers). One f-stop translates into about 20 IRE units The major setting to be aware of are:

• Whites shouldn't be any hotter than 100 IRE on a waveform monitor.
• Blacks (Analog or NTSC) black should register 7.5 IRE on a waveform monitor
• Digital (DTV) black should register at 0 IRE
  

Vectorscope - A vector display measuring device that allows visual checking of the phase and amplitude of the color components of a video signal. They are especially useful when used with color bars, as the display face has targets that show both proper phase and saturation.

NOTE: You can't adjust or manipulate a video signal with just a waveform monitor and vectorscope. They simply let you examine the signal. You must use a TBC, a camera control unit or other device to modify the signal.

TBC (time base corrector) - A piece of equipment used to correct instabilities in analog video signals, provide synchronization between video signals, and adjust phase differences in signals to correct color or make them consistent with other signals. TBCs usually have a "proc amp" which lets you "tweak" or adjust the video's brightness, hue, saturation and setup.

• Basic proc amp adjustments include
• Chroma (amount of color)
• Phase / Hue (actual color)
• Brightness (amount of gain or brightness)
• Contrast (on some)
• Setup (aka pedestal) A signal elevating the black level and all other portions of the video signal

If you have a copy-protected VHS tape or DVD that you need to dub, you can run the video through a TBC. It strips the old sync, which has been modified to create dubbing problems, and replace it with new sync.

FCP and Avid provide computer-generated waveform monitors and vectorscopes. This provides an excellent way to check levels for graphics and when applying video effects (these are often too bright for legal video).

Note that when capturing DV footage in FCP's log and capture tool, you can't control the proc amp settings (color, brightness, setup, etc). You'll find that the "clip settings" are greyed out. This is because you are transferring footage that's already been digitized. However, when you are wokring with a third party capture card (Like an AJA or Cinewave) you can modify the video signal through the proc amp settings.

If your miniDV footage is too dark or needs color correction, you must do this by either applying color correction or a filter.

Color Bars are electronic reference signals generated by cameras or post-production equipment. They should always be recorded at the head of a videotape to provide a consistent reference in post production. They can be used for matching the output of two cameras in a multi-camera shoot and to set up video monitors. In general there are two types of bars full field and SMPTE (split). The SMPTE bars are more useful.

When digitizing source footage, it's always a good idea to capture some of the bars from the beginning of each reel. This lets you check the digitized footage to ensure color accuracy.

You need two things:

• Understanding of the process & tools. This allows you to focus on having fun and being creative. How to get more familiar with the tools? Do the tutorials (again) Edit a lot in your spare time. Read the on-line manuals. Go to creativecow.net and read the forums. Spend a lot of time doing it and you'll become proficient. If you don't understand the process and tools you're likely to get stuck on basic things like "media offline".
• Your ducks all in a row! Namely start with a good treatment and then a detailed script. Also have your footage logs, graphics, and music etc. Minimize the time you spend in an edit session trying to figure out what shot comes next by planning in advance. This is what should be done in pre-production or at some corner cafe with a mug of your favorite beverage. When you edit, you need a plan, a concrete idea as to what you’re doing or you are wasting your time or someone else's money.

Time code

Time code is an electronic numerical signal recorded or embedded into the signal, which allows videotape and multitrack audio machines to be synchronized with frame accuracy. With time code, each frame or location on a tape is assigned a unique number. This allows us to access that specific frame or location on the tape precisely, again, again, and again with frame accuracy.

Here in the US with our NTSC standard, we’ve been taught that video runs at 30 frames per second- actually it’s 29.97. While we count it on a 30 frames per second basis, video runs at 29.97 frames per second.

During recording, a videotape recorder (capable of recording timecode) assigns a unique timecode number to each frame of video, which is recorded along with the video information.

Time code is formatted like a 24 hour clock HH:MM:SS:FF.

"Hours" range from 00 to 23, "minutes" range from 00 to 59, "seconds" range from 00 to 59, "frames" ranges from 00-29.

There are two ways to count or number timecode (which can usually be selected on the VTR):

• Basic, (Non-drop) frame
• Drop frame.
  

Basic (non-drop) timecode: each new frame of video is assigned the next higher number (06:01:00:29 becomes 06:01:01:00). The problem with basic non-drop timecode is that the frame numbers drift from the actual elapsed time of a program.

Imagine you've been asked to assemble a two-hour program for a TV station. You could set your timecode display to start at 0, then assemble your programming. When you got to 2 hours you'd be done, right? Wrong- unless you are using drop frame timecode.

Let's assume a 30 frame per second rate as our basic timecode readout leads us to believe and look at a day:

We need 2 hours of programming:
2 hours x 60 minutes = We need 120 total minutes.
120 minutes x 60 seconds = We need 7,200 total seconds.
7200 seconds x 30 frames = We need 216,000 total frames.

But video actually runs at 29.97 frames per second. That's a 3/100ths of a second difference from 30 frames per second.

We really need to take our total seconds and multiply by 29.97 instead of 30.
So 7,200 x 29.97 = 215,784 total frames.

There's 216 frames difference - or about 7.2 seconds.

These problems are avoided if you use drop frame timecode.

Drop frame time code is harder to calculate, but provides a numbering system that is more accurate, timewise. It will not drift as much over time andf can be locked to it with great accuracy. Most producers and networks request this for both field tapes and edited programs.

In drop frame time code, the frame numbers 0 and 1 are removed from each minute except for every tenth minute (starting from the first). That is, minute 00, 10, 20, 30 and so on, do not have any frame numbers dropped, but all other minutes do. (06;01;00;29 becomes 06;01;01;02)

You can tell when something is drop frame because the time code display has semicolons (;)

PS: There are several good shareware time code calculators out on the net.

On-line v Off-line editing

• Off-line is not intended for broadcast. You can create a rough draft and/or an EDL
• On-line produces the broadcast master

Cables and Connectors

How does one get video from one device to another? Typically through a cable and a connector. For analog video, the choices include:

• Composite (single cable) This is the "lowest common denominator." Every video recording & playback unit has this. Try to avoid as the composite video signal is prone to a variety of artifacts. Connectors: RCA & BNC
• Y/C (a.k.a. S-Video) The idea is to keep the signal broken down into the luminance & chrominance components. This is much better quality than composite. You can find Y/C ports on everything from consumer camcorders to high-end digital VTRs. Connectors: single, multi-pin connector on each end.
• Color difference The signal is split into three components: Y, R-Y, B-Y.; YUV; or Y'Pb'Pr'. The Y is for luminance, the U is for the blue color difference, the V is for the red color difference. Most high-end VTRs (Beta, MII, DV and digi beta, etc) have these connectors on them. This is the norm for getting Beta SP footage into an editor. It's a tad better than the Y/C system. Connectors: Usually three BNC connectors on each end. RCA connectors are often used on DVD players and projectors.
• RGB (true component) The Red, Green & Blue signals are kept separate. Not as common as the color difference system. Connectors: BNC & custom multipin connectors (triax).

For digital video you can use:

• Firewire (a.k.a. IEEE-1394) - a computer bus which has been accepted by most of the AV equipment manufacturers. You can squeeze DV down a firewire conection.
• SDI (Serial Digital Interface) Found on high-end digital video devices. Can include embedded audio along with the video.
• HD-SDI - The high-definition version of the SDI digital interface.

Digitizing & Color Sampling

Digitizing is the process of converting an analog signal into digital form. We do this to create digital video. Digital video is video that has been digitized and is now represented by binary code- 1s and 0s.

When we digitize video, we have to store the data somewhere- onto tape, disk, or even a memory chip. (Digital still cameras can capture small amounts of video directly onto a chip) When digitizing, compression is often used, in order to fit the most information into the smallest space. specialized file formats or data storage methods, called Codecs are used.

Codec is an acronym that stands for "compressor/decompressor. A codec is a method for compressing and decompressing digital information. It can use specialized hardware, software or a combination of both.

DV is one type of Codec. JPEG, MPEG-2 (used for DVD video) are a few others.

Visit http://www.adamwilt.com/DV-FAQ-tech.html#colorSampling for a more detailed explanantion of this.

How It Works

A video signal consists of luminance (black and white) and chrominance (color) information. While the luminance and chrominance are combined to create a TV display, the two signals are treated differently. TV works sort of like a coloring book. The luminance draws the outlines (define darks and lights) and then the color is applied.

You can see the luminance portion of the signal on a TV monitor by turning the color (chrominance) all the way down.

Most of the important information is in the luminance portion of the signal.

Sampling (frequency) & Quantizing (bit-depth):

When we digitize video, we sample it. We take a digital snapshot and convert it into 0s and 1s. This is true for audio, video or a combination of the two.

• Sample rate is how many times per second we take a picture.
• Quantizing or bit depth is how good/detailed the picture is.

Here are a few different ways audio can be digitized:

• 8 bit at 22KHz (low end, computer alert sounds)
• 16 bit at 44.1 kHz (DAT, CD, MP3 at high quality)
• 16-bit at 48 kHz (DV, DAT/mini disc)
• 24 bit at 48 kHz (High-end DATs & workstations)

The higher (faster) the sampling rate, the better the quality. The larger the bit depth, the better the quality is.

The digitizing process:

1. Capture the original signal from an anaolog source (tape or live)

2. Sample the input signal. This simulates the analogue signal in the digital domain

3. Quantize the signal. This gives each sample a numeric value.

4. Compress the signal. The overall amount of data is reduced to a more reasonable size.

5. Record the signal. Once digitized, the signal may be recorded on a videocassette or a computer’s hard disk drive (HDD).

Color difference signals are one way to break down the information in a video signal. (Other ways include composite video, Y/C or S-Video, and RGB.) The color difference signals can be expressed as R-Y, B-Y or Cr, Cb or sometimes U, V. This color difference signals are used in the digitizing process.

Color difference signals: We all know by now that TV uses an additive color system based on RGB as the primary colors. Mix them all together & you get white, right? Well if the RGB information were stored as three separate signals it would take a lot of room to store all the information, since each signal needs equal bandwidth. Fortunately great technical minds have figured out a way to get the same information into a smaller box- they've devised a way so that it wouldn't take up as much room.

This is done by transforming the RGB signals into two new video signals that can be band-limited with minimal loss of perceived picture quality. These are known as color difference signals. What the heck is a color difference signal?

Combining the RGB signals according to the original NTSC broadcast system standards creates a monochrome luminance signal (Y). So you can basically pull out the blue and red signals and subtract them from the total luminance to get the green info.

4:4:4 vs. 4:2:2 vs. 4:1:1

Today’s digital technology provides us with several ways to digitize video, mainly 4:2:2 and 4:1:1. What do they refer to?

Quite simply, they refer to the ratio of the number of luminance (Y) samples to the samples of each of the two color difference signals.

In the video signal, the most important component is the luminance as it gives us all the detail absolutely necessary in the picture. As a result, we must sample luminance at a very high rate, 13.5 Megahertz (million times per second).

Given that the luminance portion is sampled at 13.5 MHz. Let's apply the beforementioned ratios: 4:2:2 and 4:1:1. In a 4:1:1 component digital sample, the color information is sampled at 1/4 the luminance rate: 3.375MHz. In a 4:2:2 system, the color is sampled at 1/2 the rate of the luminance or 6.75MHz.

What does this mean?

Quite simply, the color depth of a 4:2:2 component digital signal is twice that of a 4:1:1 signal and, from the standpoint of color bandwidth, is twice that of today’s popular component analogue formats. This means better color performance, particularly in areas such as special effects, chromakeying, alpha keying (transparencies) and computer generated graphics.

Quantizing / Bit Depth

The next step in the digitizing process is quantizing. In digital video we do not record video as we do in the analogue world, but rather a series of numbers which give us a reference as to what the initial analog video signal was.

• 8 bit: DV, DVCAM, DVCPRO, Digital S
• 10 bit: Digital Betacam, D1

 

 

Back to Jim Krause's Summer T351 Home Page