'Inspired 3D Short Film Production': Production Planning — Part 5

Be sure to check out Parts 1-4 of Production Planning to learn about basics such as the production pipeline, budget analysis and production planning.

What To Do if Your Financial Bid is Too High

If you run the numbers and your cost estimate is beyond what you can afford, don't throw in the towel just yet. There are several possible solutions to the problem of an over-inflated production budget.

• Raise more cash

• Simplify your intended cinematic vision

• Take more time

• Lower your production costs

Raise More Cash

Is it possible to increase the amount of money you can spend on your production? If you are a hobbyist working at home, perhaps a friend or relative might be willing to make a donation. Consider a part-time job. Take out a personal loan or borrow from a credit card. See whether any software or hardware companies might be willing to sponsor a filmmaker who is using one of their products. Such scenarios are rare, but they are possible. If your production costs fall under the category of art/film school tuition, can you get scholarships, financial aid or student loans?

Simplify Your Intended Cinematic Vision

As we've mentioned many times, simplicity in design and execution are worthwhile goals for a CG short. Minimizing the complexity of your proposed film idea can significantly lower your production costs by allowing you to reduce your team size, use less powerful (and less expensive) software and hardware, delete unnecessary production steps and even do without certain equipment that you had previously thought you would need. Simplifying your film idea will also shorten your production cycle, which will mean less money spent on potential teammate salaries, upgrades, repairs and technical support fees. If you are working on your short as a full-time endeavor between jobs, the shorter your production cycle, the sooner you can go back to work and replenish your savings account.

If you plan to use dialogue, carefully consider whether you can possibly tell your story silently instead. Doing so will save you money and time because you will no longer have to find or hire voice actors, schedule recording sessions, utilize audio equipment and software, model phoneme face shapes and perform lip-synch animation. If your story absolutely needs words, perhaps you could use narration instead of actual dialogue. This option will still require voice actors and audio equipment, but phoneme face-shape creation and lip-synching will not be necessary. Or, if appropriate for your story genre, consider silent-film-style dialogue cards or even comic-book-inspired word balloons.

Can you reduce the length of your story? Delete unnecessary scenes and make sure all existing story points are told efficiently. The shorter your film, the shorter your production cycle.

Can you tell your story using fewer characters? Each character in your film needs to be modeled, rigged and animated. Are you absolutely certain that you need to show every single animal species in your Noah's Ark story? Surely you can leave out a few dozen or so.

Will you be able to render all of your scene elements in single passes, thereby removing the need for a compositing stage?

Can you simplify your characters and background elements so they take less time to build, animate and render? Perhaps your goofy cartoon alien works just as well with three fingers instead of five. Maybe you don't need to actually model every single tree. Look into instancing or using 2D cards or background plates instead (see Chapters 13 and 20).

Take More Time

Although lengthening your production cycle might very well end up costing you more, giving yourself more time might actually save you some money. You might be able to get away with half as many teammates if you give yourself twice as much time to complete your film. Also consider that rendering times can eat up a substantial portion of your schedule. Although you can indeed decrease these times with RAM and CPU upgrades, doing so can be costly. On the other hand, if your schedule is long enough to accommodate slower rendering times, you might be able to get away with weaker and less expensive equipment. Giving yourself an extended deadline might also mean that you can spend fewer hours per day on your short, thus allowing you to simultaneously work a full- or part-time job to offset your production expenses.

Lower Your Production Costs

Don't buy more computers than you really need. A 3.0-GHz CPU is certainly faster than a 2.8. However, the speed difference will probably be barely perceptible, while the cost difference might be a bit more painful. It is certainly tempting to assemble the finest workstation money can buy, but even a package deal at a local chain store will be sufficient for most production needs these days. Do you absolutely need a $1,800, 23" flat screen LCD monitor, or will a 19" CRT for $300 do the trick? Look for secondhand equipment in classified ads and on eBay, as well as returned or discontinued and heavily discounted store items.

Carefully consider your software choices. Prices can range from free to several thousand dollars. In many cases, you do indeed get what you pay for; however, just because a particular piece of software costs 10 times as much as its competitor, it doesn't necessarily mean that it is 10 times as powerful. This holds true for sophisticated 3D CG packages, paint programs and editing software, as well as simple databases and renamers.

Also keep in mind that you can very likely get away with older versions of your chosen software packages. Often the absolute latest version of a particular program only has a few extra bells and whistles compared to last year's version, which you should be able to find for a significantly lower price.

Borrow or rent as much as possible, especially when it comes to seldom-used items. If you are only going to need a scanner once to convert your storyboard drawings into digital files, go to a local print shop rather than buying a scanner. See whether you can occasionally use such items at work or school if feasible and permitted.

Try to get multiple uses out of single items. A camcorder can double as a microphone, and a digital camera can double as a scanner.

Use talented friends and relatives instead of professional actors for dialogue/narration recording. If your teammates will be charging you by the hour, see whether you can possibly do a bit more of the work yourself and hire a smaller team.

Time Estimate

Now that you've analyzed the financial feasibility of your production, the next step is to create an approximate overall time estimate, which will not only determine whether your deadline is reasonable, but will also act as a preliminary production schedule.

To do this, you will need to determine how much time it will take to complete each of the three phases of your project cycle pre-production, shot production and post-production. You can accomplish this by analyzing each individual production step within each phase.

If you don't feel that you have sufficient experience to generate fairly accurate time estimates for any of the individual tasks, seek advice from friends, colleagues, fellow students or books.

For bidding and budgeting purposes, a producer at a professional studio will need to make his estimate extremely accurate and will often crunch the numbers down to the hour. For your purposes it is probably not crucial for your estimate to be quite so precise, so think in terms of days rather than hours. When you are making these estimates, consider the concept of a man-day. We define a man-day as an eight-to-10-hour block of time. A man-day might differ from your actual workday, depending on how many hours per day you will be able to spend on your film. If you will be working on your film part time, say four hours per day, then it will take you two workdays to fulfill one man-day. So if you estimate that it will take five man-days to complete a specified task, realize that it will actually take you 10 part-time days to finish, not five. Same total hours, but different end dates.

Similarly, if you want to think in terms of weeks, consider what a week means to you. If you're working full time but taking weekends off, then a week would be five days. If you are a bona fide workaholic, then a week to you is probably seven days. If you estimate that a certain production task will take 35 man-days to complete and a workweek for you is Monday through Friday, then your estimate in weeks would be seven. If you plan to work seven-day weeks, then this task will take five weeks. Remember, the ultimate goal for your overall time estimate is to determine your completion date. The actual hours spent getting there are of secondary importance at this stage.

Once you have estimated the total amount of days that each production phase will take, you will need to divide these tallies by the number of members on your team and adjust for the actual contributions that each teammate will deliver. Also factor in any periods during which you might not be working at full capacity, such as holidays or vacations.

Pre-Production Time Estimate

We have defined pre-production as the stage in which the digital building blocks of your film are initially created and organized. Make a list of these assets and then try to roughly estimate how long it will take to create each one.

Vocal tracks and other necessary audio files. If your film will require dialogue, consider that you will need to find appropriate actors, stage recording sessions and process your audio files. This can take anywhere from a few hours to a few weeks depending on the number of lines being spoken per character. If sound effects or a music track will dictate the action of your film, it will be necessary to locate or create such audio files at the pre-production stage. Writing and performing original music will obviously take a lot more time than licensing an existing song or purchasing a piece of copyright-free music on a CD or from an Internet store.

CG character models. Obviously, the time it will take to create these assets will depend on the number of characters in your film and their overall complexity. A simple character model might take a day or two, while a complicated model might take a few weeks. Also consider that depending on your facial animation process and the possible existence of vocal tracks in your film, you might need to sculpt expressions and phoneme shapes for your characters.

CG backgrounds and props.

In general, background and prop models should take less time to build than character models. However, this does not mean that they are trivial. Like everything else, a time estimate for such assets will depend on their quantity and complexity.

Texture maps.

Using procedural texture maps within your CG package will certainly take less time than painting your own original textures; however, a fair amount of trial and error will be required to determine the right parameter settings. Try to come up with a list of all the texture maps you will need and then estimate how long it will take to paint each one. Using photographed maps might save you some time, but cleaning them up and making them tile properly (if necessary) will also take a bit of work.

Background paintings. If your shots will require such assets, make a list of how many you will need and then how long it will take to paint (or perhaps photograph) each one.

Character setups. Rigging even the simplest of characters can be rather time consuming depending on your software of choice and your skill level. A very simple character with disjointed geometry pieces might take a few days to set up effectively, while a complex character with organic forms, multiple limbs, hair and clothing will probably take several weeks.

Go through the above categories and jot down an appropriate time estimate for each one, based on the needs of your film. Then sum up the total to come up with your own pre-production time estimate.

To initiate a running example, let's say you came up with an estimate of 50 man-days. Suppose you'll be working full time, meaning one man-day equals one workday, but you'll be taking weekends off. Your pre-production phase will therefore reach its finish line in 10 weeks.

Shot Production Estimate

Before diving into actual shot production, we recommend that you create a 3D animatic of your entire film. If you plan to take this step, you will need to make a rough estimate of how long it will take to produce your 3D animatic. The time estimate will depend on the number of shots in your film, the size of your team, and the speed of each contributor. A rough but fairly standard estimate would be approximately two shots per man-day. So if your film has 60 shots and you have two teammates of equal speed, creating a 3D animatic should take 15 days. That would be three weeks if you are taking weekends off, or two weeks plus one day if you are working seven-day weeks.

Now that you have your rough estimates for pre-production and 3D animatic creation, it's time to look at actual shot production.

Draw a chart with nine columns, like the one in Figure 17. Use a spreadsheet package if you have one available, so the software can sum up the totals for you. Otherwise, use pen and graph paper or a dry-erase board and do the math yourself.

These columns represent the level of completion of each shot and follow a typical digital pipeline of the steps required for individual shot production, described as follows:

• Blocking. This is where the basic animations of your characters and objects are established. Blocking involves providing just enough animation refinement so the story of each shot is clear, but without any details.

• Animation. This is where you refine your blocking passes and add details until your characters and objects are moving convincingly. Don't forget to include sufficient days for any facial animation and lip-synching that might be required.

• Lighting. This is obviously where you place lights appropriately in each shot so your stories will be told with the intended clarity and mood.

• Rendering. This is where your software creates a series of snapshot images of each frame of the current shot based on the placement, materials and textures of your characters and objects, as well as the placement, setting and movements of your lights and cameras.

• VFX. This is where you add any appropriate secondary or particle effects, such as cloth, hair, feathers, snow, rain, smoke, fire or water.

• Compositing. This is where, if applicable, you assemble the separate image layers of your shots into final collages. If you render all of the elements of each scene in a single pass, then compositing is not necessary.

• Audio. This is where any shot-specific sound effects or music are synched up to your rendered imagery.

Now, fill in a man-day estimate for every step of each shot, based on your individual experience and advice from your friends, colleagues, teachers or teammates. Consider that rendering may or may not require any man-days. If you set up your shots and then render them overnight or while you are away, then this step will not cost you any actual man-days.

Keep in mind that not all shots are the same. Some will be very short and simple, while others will be very long and complex. Some will require particle and audio FX, while others will not. Analyze each shot based on its length and complexity and try to come up with an appropriate number for each step.

Figure 17 represents an example chart for a very short eight-shot production, assuming a 3D animatic was already created and renders will run while the team sleeps.

Now add up all the columns and rows to get a total shot production estimate. In our example, the total number of man-days is 100. Working half time at five days per week, which would translate to about 40 weeks. In a full-time, five-day-week scenario, we'd be looking at about 20 weeks. A seven-day-per-week workaholic would finish in about 14 weeks, assuming he didn't burn out halfway through.

The addition of a second team member may or may not cut these numbers exactly in half, depending on his speed and the actual skills he will bring to the table. If you and your teammate are both generalists and can handle all stages of each shot, then you should each be able to do four shots in their entirety, and the time estimate can be approximately halved. However, if you will concentrate on animation while your teammate focuses on lighting, vfx and compositing, it will be a bit more complicated to come up with an accurate estimate because your teammate will finish shot number 2 and then have to sit idle while you spend your eight days on shot 3. He might be able to start lighting after your second animation day, but then a lot of overlapping will take place, and you'll need a much more detailed schedule to come up with an accurate time estimate for shot production. Regrettably, if your team is made up of artists with different skill sets, you'll have to simply make an educated guess. In the described scenario where one teammate animates while the other does the rest of the work, you can probably cut the total number by about a third.

Post-Production Estimate

For our running example, with a lone filmmaker working five full-time days per week, we are looking at 10 weeks of pre-production, plus three weeks for the 3D animatic, plus 20 weeks of shot production, equaling 33 weeks so far. Post-production tasks need to be factored in as well. These might includefinal editing, film grain filters, color correction, post-camera moves, titles and credits. Hopefully this will be the shortest of the three stages. Approximate the time it will take to add all of the finishing touches to your film. For our example, let's say 12 days for final edits, three days for a flying 3D title and some simple text credits, and then two weeks to burn copies and send out festival entries. That's another five weeks, so adding it all up so far, you have 38 weeks.

Consider, however, that many things can go wrong in the midst of a production. Computers can crash, files can get overwritten, and teammates can quit. Because of such possible mishaps, we recommend padding your initial estimate by 20%, just to be safe. So with padding, now you're looking at 46 weeks just under 11 months.

What was your total number? Was it more than you expected? If so, is your deadline somewhat flexible, so you're still in good shape and ready to begin production? Or does the disparity between your time estimate and your fixed deadline represent a major problem?

To get a copy of the book, check out Inspired 3D Short Film Production by Jeremy Cantor and Pepe Valencia; series edited by Kyle Clark and Michael Ford: Premier Press, 2004. 470 pages with illustrations. ISBN 1-59200-117-3 ($59.99). Read more about the Inspired series and check back to VFXWorld frequently to read new excerpts.

Jeremy Cantor, animation supervisor at Sony Pictures Imageworks, has been working far too many hours a week as a character/creature animator and supervisor in the feature film industry for the past decade or so at both Imageworks and Tippett Studio in Berkeley, California. His film credits include Harry Potter, Evolution, Hollow Man, My Favorite Martian and Starship Troopers. For more information, go to www.zayatz.com.

Pepe Valencia has been at Sony Pictures Imageworks since 1996. In addition to working as an animation supervisor on the feature film Peter Pan, his credits include Early Bloomer, Charlie's Angels: Full Throttle, Stuart Little 2, Harry Potter and the Sorcerer's Stone, Stuart Little, Hollow Man, Godzilla and Starship Troopers. For more information, go to www.pepe3d.com.