Thursday, July 30, 2009

#0019 The Pipeline -Form Follows Function

All Saints Chapel in the Cathedral Basilica of St.Image via Wikipedia

Pipeline Design 101

Anyone who has taken any sort of introductory design class, whether in art, graphics, architecture or what have you, knows that "form follows function". When designing a CG pipeline one needs to consider the function of the pipeline before designing the structure let alone the mechanics.

In our last post, "#0018 Understanding Pipelines" we discussed that the term pipeline is broadly used for many different things, from software and hardware component design to include various types of CG work pipelines. In an epiphany of sorts, I came to realize that a CG pipeline can fall in any of three classes:
  1. the production pipeline: the labor processes and steps
  2. the material pipeline: the data assets acquired, produced and delivered
  3. the approval pipeline: the direction and redirection mechanism and loop
Another way of looking at this is process, data and metadata; to get totally geek and introduce some new jargon, we could speak of the primary, secondary and tertiary pipelines.

Case Study 1:
Small Shop, Big Job, Varied Jobs

I recall on

Travelstart Design Studio is officially 3 of u...

e day after working with a fellow for six months he stood up and declared to me, "we need a pipeline!"

At the time we were producing about 35 shots a week with a crew of about seven 3d artists and five compositors. Our production pipeline was deliberately simple. While we were all working on the same television series with a fairly defined theme, the content of each episode could vary widely, and hence the nature of demands on our small 3d team was constantly shifting.

Primary Pipeline: Production
I therefore preferred to staff with a number of generalists at different skill levels, but I included one generalist who was a strong modeler and one who was strong in dynamics. A 3d supervisor provided muscle to round out any gaps in skill sets, and I would step in to demonstrate "advanced" 3d techniques in anything from models to animation to shaders.

From our 3d unit the work would move to our compositing unit, where again we had little specialization, other than skill levels. Again the unit was lead by a highly experienced and capable comp supervisor. The use on our team of compositors at a wide range of skill levels, with a couple of old salts who knew green screen and shot aesthetics well, rounded out by a young motion graphic designer, allowed us to assign shots based on difficulty, design needs and so on. And again I could step in to advise or teach or do.

So, we had a simple production pipeline with a form suited to it's function: produce a large volume of material on a weekly basis with highly variant visual effects demands.

Secondary Pipeline: Data
Our coordinator managed the materials pipeline, which at our level of operation involved checking in assets from editorial and production and delivering product to editorial. I set-up a system using GOOGLE Docs spreadsheets to log assets needed, assets received and shots delivered and take status notes.

Between 3d and 2d we'd established a materials pipeline that wasn't very sophisticated: artists would log a shot ready for compositing and specify a database location code and compositors would pick up the work and mark it with a code indicating it had been picked up. One artist was assigned to A simple effective. And again this was communicated using the same GOOGLE docs database.

In addition, our "generalist-modeler" was tasked to collect and manage a database of re-usable models. At the same time, we tasked a couple of compositors to maintain our library of 2d assets.

Again I was intimately involved with this (and later we hired an I.T. Coordinator to help). So our data pipeline was in place.

Tertiary Pipeline: Direction and Approvals (or the data about the job)
Direction and approvals went through me. I set-up the GOOGLE docs system for tracking every show, and in this I logged all the shots, gave them a number and included the director's notes and script snippets. This was set-up so that a week or two ahead of need our coordinator would make a new show document and inform the director. After the first time working with us, about 90% of the directors were able to fill out this form before my first meeting, and would modify it during pre-production.

I would add my shot direction and my 3d and comp supervisors would add theirs. As shots were submitted, we used a tight naming system to code the internal approval level into the shot name. This system was designed so that the take delivered always referred back to a specific composition in a specific project file, and each 3d element referred back to the originating Maya file, to allow rapid identification of files for revision. Enforcing this naming convention was a continual effort of education and cajolery on my part assisted by the coordinators and other supervisors.

Change notes were communicated in the spreadsheet as well. While not foolproof, we made an effort to get all notes written down and communicated them orally as well. The pace and volume of work made formal dailies impossible, but on a daily basis shots were reviewed by the 3d supervisor, comp supervisor and myself. We encouraged directors to come in and would call and ask them to come in, but were generally too busy to chase them. Happily, almost no shot that made it past the other supervisors' and my eyes was ever bounced. So we had an approval pipeline that functioned.

Case Study 2:
Big shop, Long Job, Volume Business

Computer cubicles inside the Digital and Multi...

Another shop I worked for was very different. I want to get on to other things, so here is a summary:

Production Pipeline: Specialists in roto, wire removal, cleanup and tracking; specialists in modeling, animation, characters and dynamics; specialists in scripting and special programming needs; specialists in data management and film outs; specialists in green screen extraction, 2d tracking and generalist compositors and 3d animators. I forgot to mention specialists in storyboarding, design, character modeling, character rigging, texture UV layout, texture painting, researchers, lighting and shaders. Plus coordinators, CG supervisor, Roto supervisor, etc. About 25-30 job descriptions and about a staff of 150 working on two or three films.
This shop was geared up to handle about a dozen or two shot sequences, comprising anywhere from eight to 40 shots, with similar shot needs in each sequence. Production turn-around for a film with 300 or so shots ordered was about eight months.
Data Pipeline: tell a coordinator where things are
Approval Pipeline: dailies

The Two Pipelines
We need a pipeline!

An icon from the Crystal icon theme.Image via Wikipedia

Getting back to the story: so he says to me, "We need a pipeline!"
Of course I dismissed the suggestion, which wasn't the best way to handle it, but that's a story for another day. The point is, we already had a fully functional pipeline.

My shop #1 is engineered for a low budget operation with flexibility in mind. Another small shop that specializes in a particular kind of work might find a different pipeline more to their advantage.

For example, a shop producing over a four month period 500 or so shots with 3d vfx of WWII aircraft and warships in the Pacific theater will need and be able to get advantages out of a more specialized pipeline than we employed. The mission of shop #1 is to support a science variety show, and the range of topics is immense. One week they are animating black holes, another constellations and another depicting exploration of Mars. They have hypothetical animals one week, and volcanoes the next. The work needs were unpredictable; the shop making airplanes had predictable needs.

Likewise, shop #2, making feature films, was able to achieve specialization because the volume of work included a significant amount of essentilly repetitious shots. At the same time, the volume of work allowed them to have at least one and ususally several persons expert in almost any specialization. And, at the same time, the length of projects and the overlap of several films, allowed them to maintain a pool of talent with a large variety of skills and experience, providing indirectly generalization and directly labor economy. Further, the shop was big enough to have more coordinators than I could count, so the details of their material and approval pipelines were hidden from artists. (Today, more reliance is made on technology to coordinate these pipelines.)

Masai Giraffe

follows Function

Your job, as a CG Supervisor, is to understand the function first and design a pipeline to fit: remember form follows function. The idea that some people have that a pipeline comprised of specialists is not the best solution in every situation. Just as the giraffe has a long neck to reach the highest leaves, your CG pipeline must follow the form that gives you the most efficient and effective solution to the work you need to do. In some situations, the best solution is a team with diverse capabilities and a leader able to create ad hoc pipelines using those talents to get special problems solved. In other situations, an assembly line of specialists will be more effective.

One size does not fit all.

---Happy Pixels!

Tuesday, July 28, 2009

#0018 Understanding Pipelines

Trans-Alaska Oil Pipeline
Deconstructing the "pipeline"
Among your key responsibilities as a Computer Graphics Supervisor may be the design or implementation of a CG pipeline.  Even if you do not design one, you will manage one, and as the pipeline manager your concern will be to keep it running efficiently.  Further, even if you do not design the pipeline you manage, you will be accountable to management to insure the pipeline design and staff capabilities are most appropriate for current needs.
Your task will be compounded by a very simple problem: jargon.  The word "pipeline" is a rather meaningless term.  Like so many terms in computer graphics, it has been adopted from some other discipline and the word crafted to mean something for a specific situation.  You will soon find that there is little literature on the subject and that the word "pipeline" means different things within computer graphics to different people.
Case in point:  I'm sitting across the table discussing an employee's recommendation that we need a "pipeline", a recommendation he's made to myself and my boss, who wants to know why we don't have one.  As I sit there, I'm thinking about how anyone could think our crew of 10-20 artists could have spent the last 12 months making around 1800 vfx shots without a very functional pipeline.  Perhaps, I think, he wants a change in the pipeline he's been part of for 12 months?  So I ask him what he means, and he begins describing an asset management system....

Description unavailable

The term "pipeline" may seem to be a term grounded in the antiquity of computer graphics if you've only been around a few years, but it is a rather recent bit of jargon, introduced 10-12 years ago.  The word "pipeline" replaced "workflows", which in turn replaced the terms "cg process" and "cg production phases" around 20 years ago.  The change in terminology has both opened and constricted our thinking about visual effects production processes and the people who make them happen.

The term "pipe" was introduced in computer jargon with the development of UNIX.  A UNIX pipe is an input-output data structure that eliminates the need for program A to write a stream of data to a file that will be read and processed by program B.  Instead, the operating system command line language allowed the insertion of a pipe symbol, which meant, "connect the output of program A to program B", skipping writing the file to save time and disk space.  
A few years later, a young company making a computer optimized for graphics, Silicon Graphics, designed and sold graphics processors using a "pipeline" concept.  Essentially, the graphics engine developed by SGI encapsulated in hardware, at a great speed advantage, the UNIX concept of piping the data through a series of simply written modules making a series of small operations into a significant result.
Over time the concept of moving computer data through a series of software modules would get translated into a work process involving progressing the work product through a series of specialized workers (who happen to use node-based software and other technologies all based on the software pipe concept).  And so "CG process" became "CG workflows" and these became "CG pipelines"....

Back to my employee meeting.  His answer got me to think about pipelines differently.  For some time I had been thinking of the pipeline in a very narrow sense, as a type of CG  work-flow or description of the CG process.  Suddenly I saw that a pipeline could be the flow of any kind of data, not just the flow of our work product, and that the CG pipeline we used consisted of a multitude of different classes of pipelines that touched one another in places.  In an instant, concepts I was thoroughly familiar with became redefined and reorganized.  Like parallel universes, pipelines coexisted within my production environment in different dimensions of perspective.
    The Three Pipelines
Computer cubicles inside the Digital and Multi...

The Production Pipeline
The traditional way of thinking about a pipeline is the production pipeline.  This is essentially modeled after post-industrial assembly systems where workers specialize in a given task.  The assembly line is a such a system where the work product moves past a series of work stations at a fixed speed, forcing each specialty to be constrained to a fixed duration and narrow set of skills.   Computer graphics is unlikely to achieve the efficiency of Henry Ford's automated line, but the concept of specialization has been embraced. We can take advantage of the phased nature of CG production --model construction, motion construction, lighting, shading, rendering and compositing to break up the work.  The degree of specialization of a pipeline is a subject for some study: not all work missions will support the same production pipeline profile.
The Materials (Or Data) Pipeline
Every CG environment involves the delivery of a product, a shot, a sequence or a complete show.  Along the way assets are gathered and intermediate assets or work products are constructed.  Assets used as inputs may include footage, stills, and reference materials.  One person's output becomes the next person's input.  Hence the pipe-line.  Managing the flow of materials through a shop is an important sub-pipeline, in fact it's really a corollary pipeline.  
The Approval (Or Metadata) Pipeline
A key material in any pipeline comprises the aggregate of direction and instructions about a particular work product (again, a shot or element of a shot at the finest level).  Managing this data involves collecting, organizing and distributing the information to the right persons.  Because it involves a decision loop, the approval pipeline is itself a looped pipeline with an iterative nature.  
Other Pipelines
Adjacent to these may be other pipelines, but these are the essential three.  The first is labor; the second is the data used  or made by labor; and the third is the metadata, the data about the data.  Any other pipelines are special cases of one of these three.  For example, a film-out pipeline is a special case of a production pipeline.  A rendering pipeline is a part of the production pipeline. A material check-in procedure is a sub-set of a materials pipeline.  A sign-off procedure is a part of an approval pipeline. A payroll system is a meta-data pipeline that collects and processes information approving compensation for the work done.
The Integrated Pipeline
The degree that these three pipelines are brought together can determine how synergistic the entire operation is.  In many operations, these three pipelines exist more or less in isolation.  In some operations, the piepline is highly computerized.  In others, the data may be managed through slips of paper, whiteboards or your gray matter as supervisor.  While products exist to help manage the data and meta-data pipeline, these depend on a solid foundation of procedure and conventions.
The Best Pipeline
The question is, what is the best pipeline?  The answer is simple: the most appropriate one.  Some people hear a bit of buzz about the pipeline at shop ABC, and think that everyone should organize the same way.  Or they hear some marketing hype about a wonderful asset management system or project tracking application, and they are all hot for it.  Your job as a CG Supervisor is to help management sort through the noise and recommend the most appropriate pipeline to satisfy the short-term job at hand and the long-term realities of your company.  A pipeline that may be perfectly beautiful and elegant is worthless if it does not fit your company's needs.  Analyzing those needs and weighing your options is another story.
Thanks for joining us today.
Now go make some pretty pictures!

Saturday, July 25, 2009

#0017 Back in the Blog Spot: Reader Feedback, Hot books

Cypress swampImage by Elizabeth Buie via Flickr
Hello readers!

Back from the Bayou
I'm back after a short trip to LOUISIANA to discuss vfx supervising some films there.

Louisiana wants to attract more film business and unlike my lovely and great home state it is backing it up with tax incentives and such. Sort of put me in mind of my blog #0014 Labor Issue: Part 3, The Satellite Office.

Great place. Climate was unseasonably cool and dry, but the weather began getting warmer and wetter the day I left. The crayfish gumbo was swee-eet.

Saw a really beautiful swamp, reminded me of the work I did at FLIGHT 33 for The Universe: Alien Faces (YouTube video at end of blog). Wish I'd seen it before. No time to go through the swamp on an air boat. Forgot my camera anyway. Next time.

Bull Shark - postcrossingImage by Alhazred via Flickr

By the way, I was told there are BIG sharks in the local bayou --and not the nice kind either. Aggressive Bull sharks. Maybe someone was having one on me, but it appears to be true. Plus your run of the mill gators, snakes and such.

A Reader Writes
This from reader CHAD FOX:
Your blog, the art of cg supervision, was passed on to me the other day and I just want to say thank you for writing it.
I am a Visual effects CG artist working towards becoming a CG supervisor and you have offered up the very thoughts, principles and knowledge that I have seen in so few of my own supervisors. One great point was made about not considering the Human factor!! I can't say enough about how true I think that is and have tried and failed to convey this idea into every supervisor I come across who does not take it into account. Morale and motivation HUGELY effects production at the artist level. A team can get two to ten times as much done if they are motivated and happy compared to when they are frustrated and unmotivated. I find frustration and lack of motivations comes to us when the same old problems continuously cropping up and no one above you seems to be learning from the mistakes.
Thanks Chad for your feedback. It really helps motivate me to know readers find a resonance here. Chad also flagged my broken comments link, which is now fixed, so I hope to hear from more readers soon.

Keeping Up With My Reading
At any given time I'm reading about a half dozen books. Which means to say I'm bouncing between books as my attention wanders. Right now I'm reading a few I hope to digest for you in the near future and make relevant to our work as cg supervisors:
  • The Art of Innovation - The IDEO design studio innovation methodology.
  • The Art of the Storyboard - Techniques and tipsfor making better boards and developing skills. Somewhat basic, but one of few I've found on the subject.
  • Tools for Teams - Alternative ways to organize the crew
What books are you getting into?
Drop a comment on this message
What are you reading that inspires you creatively?
What are you reading that will help you in your job?

My PipesImage by ifyr via Flickr
On My Mind
VFX PIPELINES are still very much on my mind. It will take about a 10 part series to cover the material, and needs to be all written together to make it flow, so bear with me please. I am still also reading a master's thesis on the subject that may effect my take, the first part has been interesting reading. If you have any thoughts on pipeline design in CG/VFX production, please email me at VFX pipelines .

I'm also looking at alternative ways we can organize and manage our staff to get the work done. If you have any thoughts or experience with non-heirarchical work teams in CG/VFX production, please email me at CG Work groups.

Thanks again for dropping in for a little reading
and remember, B Y O Java.

Friday, July 17, 2009

#0016 CG Supervisor: open to the creative spark

Hello readers. As I said a day or two ago, I've some ideas rattling around in my head to share about Computer Graphics Pipeline design and maintenance. By coincidence, I stumbled upon a masters thesis on the subject, and would like to compare notes. This bit of serendipity, or "luck" as some call it, just dropped in my lap, I cannot recall from where.

This put in my mind how computer graphics can be like this, and you will benefit if you can be open to the unforeseen. This means encouraging artists to show you the strange and unexpected results of tests gone "bad". You might not care to use the knowledge now, but it could help you later. For instance, having seen the result of the "bad" test or take before, you could help another artist diagnose and fix the problem today. Having seen it before, you could apply it as an artistic stylization or effect for a new project's design.

High Voltage Sparks Over and Through Red LED 3/3Image by SCholewiak via Flickr

This spirit of openness supports what I was saying in an earlier post about being flexible. Sometimes we can allow ourselves to feel overwhelmed or pressed by the schedule. At such times it becomes very difficult to look at off-track but interesting work; to listen to imaginative suggestions of your staff; to be open to fresh ideas. Perhaps this is when we need to be MOST diligent about keeping the lines of communication open; most careful to encourage the artistic expression --while keeping every eye focused on the prize. Thirty seconds of attention to an artist can save you a great deal in ways you may not see. A minute of time can make an artist feel better about your relationship, their job, and the work you do. That can be priceless motivation. Put another way, an artist seeking to engage you into their imagination is an artist engaged in his work and engaged with excitement for his job.

In your press for the goal keep your eye on the prize AND keep your eyes open for the unexpected, the fortuitous, the special spark of imagination. It could be your next inspiration.

Tuesday, July 14, 2009

#0015 Priority One

changed priorities

Today I thought I'd start a series on PIPELINES.  But, something else came up, and it got me to thinking about priorities.

As a CG supervisor, VFX supervisor, producer, Computer Graphics artist or compositing wizard; in any job one has to balance priorities.

All the planning in the world and good intentions can be wonderful, but at some point it comes down to what gets done today, and in what order.  As an artist using any tool, be it Maya, Softimage, Houdini, Nuke, Shake, After Effects or whatever, you may map out your tactics for getting your assigned task done.  Along comes the supervisor with a bombshell: you need to drop what you're doing on task A and get task C (which you thought you could defer working on) out as quickly as possible.  Likewise, as a supervisor, your plan for the day may be to get certain shots moved forward a step closer to completion, others started, and still others wrapped up and archived.  But along comes your boss, or some crisis or some other unexpected interruption --and there goes your priority list for the day.

Apollo 11 Launch The American flag heralds the...
 Every project, whether it is completing 50 shots before the staff leaves on holiday or landing a man on the moon, involves establishing a plan of action; designing processes (aka workflows aka pipelines) to get the job done; and effort to work the plan and use the process.  At every stage there are priorities.  While these may seem like they should be cast in lead and baked in stone, priorities can be fluid and changing because human beings, serendipity and tide and wind are often contrary to our expectations.

The inevitability of this is so powerful that among the personal skills we all must develop, and supervisors to a higher degree, is the ability to be flexible and adapt to changing situations.  The less flexible and adapting to the environment a supervisor is, the more difficult it will be to get jobs done.

All this may seem fairly obvious truth, but I tell you some people in the workplace are rigid and inflexible.  Things must be done in a certain way and in a certain order.  In a boss, this may be annoying or aggravating, but the boss is the boss for a reason, and making these decisions goes with the job description.  But in a co-worker of equal authority, this becomes the source of friction and can lead to personal disagreements if neither side is comfortable with being flexible.  Even worse, in a subordinate, this can be disastrous, because the rigidity of a worker can make your job as a supervisor so difficult you have no choice but to move the subordinate on to a new employer.

A discussion about PIPELINES, which is something near to my heart and vital to our work, would best take place after looking at how we deal with issues about setting and adapting our priorities.  As a supervisor, you need to look at how you communicate your priorities and help your subordinates understand them and set their priorities to your company's best advantage.  As an employee and colleague, you need to look at how well you understand your bosses' priorities and how flexible and adaptable you can be.

This, it seems clear to me, is one of the top priorities in being a Computer Graphics Supervisor.

Monday, July 13, 2009

#0014 CG labor issue: Part 3, the Satellite office

If your VFX or CG production company is in a large city, like Los Angeles, finding ways to increase capacity without a large increase in cost should be a concern, and satellite facilities may be an option.  

In post #0011, we looked at the possible capacity problem:
"Some leaders of the visual effects business, both at vfx shops and at studios, are warning there could be a shortage of vfx capacity within a year -- a shortage that could drive up costs and even threaten release dates." 
--David S. Cohen VFXTalk
At some point in the near future, the economy will revive for visual effects and other CG work, and the question execs will face is whether or not production capacity can be restored or expanded quickly enough while still recovering from reduced cash flow.  Even though, when the time comes, there may be a great deal of unemployed talent available to be snapped up, some will be lost to career changes. When this economy breaks good VFX production in the US could take another hit simply because we don't have the production capacity and filmmakers will outsource even more production.

Outsourcing Costs Will Rise
While opening a shop in India, or buying services in Morocco may be a perfectly great way to get work done, prices for overseas outsourcing will gradually rise.  I read in the Los Angeles Times in March 2006 that outsourcing contracts tended to move to less expensive countries every few years because talented labor brought rising wages. While this is great for world-wide capacity, it has the negative effect of building infrastructure and talent that will at some point rise to prominence and begin replacing the Hollywood filmmakers and decision makers themselves.  Keeping as much core competency local as possible is good for US filmmakers.

Buy Low, Sell High

Now is actually a good time to look at expanding facilities, or draw up plans to expand facilities, using a 21st century model with a core facility in the urban center and subsidiary facilities in outlying areas nearby.   While this may seem expensive, it will be less expensive than expanding capacity within high density and high rent urban centers.
Opening satellite facilities in low-rent, comfortable, affordable, suburban cities outside the main hub helps solve the problem location brings to attracting a workforce and will allow companies to double in size without doubling costs.  Many artists, seeking affordable housing, live an hour or more from jobs. Because this is often a quality of life issue, workers might be more willing to relocate to a similar bedroom community.  For example, a worker might consider a relocation from a $200,000 home in one outlying city to a similar home in another outlying city before moving into the core, where an affordable home is out of the question.
A satellite facility would be located between 90 and 150 minutes from the main facility.   Ninety minutes is approaching a major threshold for commuters, and a practical one for employers wanting workers fresh and alert. It's far enough away to keep everyone from asking to go there, and close enough that an executive could make a visit and not be out of the office overnight.  Finally, it's far enough away to get very reasonable rents.
Some Benefits of A Satellite Operation
  • access to an experienced and familiar labor pool
  • workers close by for collaboration
  • supervisors and management are close enough to meet
  • main offices close enough that workers could participate in company training
  • labor can be kept inside company morale-building programs
  • offices outside high-rent zones
  • lower city and property taxes
  • ease of expansion in contraction through short term leases
  • attract workers who don't want to commute to city core
  • increased worker productivity due to shorter commutes
  • civic participation in traffic abatement programs; may qualify as a remediation
Not a new idea
Some may ask why not just move to Nevada, Arizona or New Mexico.  Sony Pictures Imageworks did just that in 2007, in part, opening a facility in Albuquerque, New Mexico.  I'm sure this move accomplished much of what I am talking about.  Likewise, Imageworks has tapped remote resources by opening facilities in Novata, California and Chenai, India.

The downside is that these facilities are far enough away to stretch communications to the limit and require an overnight (or extremely long day) to get any face time with workers in Culver City.  The idea of the satellite office is to attract and keep talent in the greater metropolitan area while allowing the highest degree of collaboration possible.

Other companies have multi-national operations, with offices in Los Angeles, New York, Canada, England, etc.  So the concept is not entirely new.

Keeping Your Base Strong
No matter where yourbase of operations is, you want to keep it strong.  For many companies, the base is Southern California, which still has perhaps the single largest labor force in visual effects, games, multi-media and animation.

But this could erode if work continues flowing outside.  While satellite operations in California will not stop outsourcing to other states and other nations, nor should it, it will help to preserve the talent base and keep production capacity high.  The labor market here is big enough and scattered enough, that one or two satellite operations a 90 minute drive or so from the main office could reap benefits for local companies.
With all the talent we have here in SoCal, we should be looking at ways to reduce overhead and wages here. Workers not faced with half million dollar mortgages and long, costly commutes will and take and can be offered lower wages. Overhead can be brought way down in outlying areas.

I'm not just talking about Los Angeles. This could be said for London, New York, Melbourne and dozens of other cities.  At some point cities become too big, and dispersing the work force makes some sense.

It seems to me that a studio with several facilities in the same time zone with highly talented pros could be a good thing.
Perhaps you have a thought on this?  Please jump into the discussion, comment on this article or write me at
Thank you for dropping by! 

Sunday, July 12, 2009

#0013 CG labor issue: Part 2, Location

I recall spending several weeks winnowing through resumes and hiring a talented 3d artist, another few weeks bringing the artist up to speed within our breakneck production pipeline, only to have her suddenly resign because she found a job with a 20 minute commute instead of a 75 minute one.  

Finding good talent at a good rate with good work ethics and attitude challenges CG Supervisors and Producers again and again when production labor needs rise.  A key factor in this equation, often ignored or disregarded, is the company location. 

There are many reason why a company locates where it does, and for most of us, this above our pay-grade. Even so, you may someday be in a position to make or influence or advise on the decision on where to locate offices.  Today, I want to look at not where a company locates its primary offices, but rather, how it can attract more experienced employees through the use of satellite expansion offices.

Among the problems that may not be unique to the United States workforce is our aging population.  This effects where many workers choose to live in a peculiar way.  Young people are attracted to dense urban environments with singles-oriented activities.  Mature workers, with families, often want a more suburban environment with good schools and quiet neighborhoods.  The upshot, as we see in Los Angeles, is a labor force often composed of young college graduates living fairly close in and experienced workers commuting long hours, which carries a cost. 

Some companies solve this problem by allowing very experienced and valued artists to telecommute.  However, for large companies, the pace of collaboration can make this difficult.  The company is located where it is to be close to clients, the industry hub, the urban vibe, or whatever; artists are needed at this location to get the job done.

The thinking currently is to have a core group of talent local, who can design effects and collaborate more closely with VFX supervisors and directors, and to outsource overseas for massive labor projects.  This rational makes sense, but it depends on maintaining local production capacity and attracting a maturing workforce.  

In Part 3 An alternative to local expansion or outsourcing

Perhaps you have a thought on this?  Please add your comment on this.
Thank you for dropping by!

Saturday, July 11, 2009

#0012 CG labor issue: Part 1, Capacity

The U.S. and California economies,  coupled with last year's uncertainties about labor negotiations with various guilds, have slowed production, causing some CG visual effects studios to close or radically downsize.
David S. Cohen warns in Variety (May 15, 2009 ) that the economic recession coupled with a production slowdown has slammed VFX suppliers, forcing some studio closures and staff reductions. He reports: 
"Some leaders of the visual effects business, both at vfx shops and at studios, are warning there could be a shortage of vfx capacity within a year -- a shortage that could drive up costs and even threaten release dates."

At the SIGGRAPH 2009 website validation of the slowdown can be seen.  SIGGRAPH, recognizing that large numbers of potential attendees are unemployed, is giving unemployed artists discount to encourage attendance.

Likewise AUTODESK is also addressing the problem.  The company is helping unemployed artists by offering them a free 90 day student license to encourage learning new skills. Training and certification is also being offered at a discount.

This is not a surprise, with official unemployment in the United States pushing 10%, the visual effects and computer graphics industries are bound to be effected.
Visual effects production capacity could be insufficient in USA  

The Challenge: Keep Capacity High and Costs Low
The challenge studios face in today's economy is how to keep costs low and productivity high.  Addressing this problem is related to maintaining or increasing domestic US capacity and productivity in visual effects.  While rents and wages may not have risen due to the economic slowdown, the economy and demands for labor will eventually change.  When it does, rapid expansion could be costly and increase quality control issues.

Film production companies have long solved the over-capacity problem for the vfx studios to some extent by shopping up the film to half a dozen or more companies.  A primary company will usually take on the lion's share of shots, but because they are at capacity with this and other projects, the load gets divided.  VFX studios are hampered by an ability or reluctance to enlarge facilities, a somewhat long-term move.  Physical unavailability of office space can be a factor in the high-rent districts many of these companies occupy; long term commitments can to a labor force 30% larger can be a concern as well.

Boutiques, often solve the need for expansion capacity through a virtual facility composed of freelancers working out of their homes.  The advantage to the boutique is the reduction in overhead, access to best available workers, access to a reliable pool of familiar workers, and easy labor size adjustment.
Now scale that up to the company with 100,200 or 300 workers working on team projects and divided into specialty functional units.  Now the philosophy has become to do as much work as possible at a primary facility and outsource work as needed to subsidiary facilities abroad.  Some companies will employ a few freelancers, but with large collaborative projects working with off-site freelancers is often more hassle than it's worth.  This strategy more or less works: core competencies and interaction with directors is kept local, labor intensive work is got at low foreign wages.

Outsourcing overseas has been a solution, but has it's difficulties as well.  There is another way.... One solution is to expand the circle of available workers to include those located 75-150 miles away from currrent production offices.

Perhaps you have a thought on this?  Please jump into the discussion, comment on this article or write me at
Thank you for your feedback! 

Friday, July 10, 2009

#0011 Another hour -- the CG Supervisor's budget nightmare

It's happened to us all:  working hard on a project and someone from accounting comes in with the news of doom --the project is over budget. 

You've worked hard to make sure everyone has the tools they require and the information they need.  You've busted your tail and so have your CG coordinators to make sure the materials pipeline is flowing, command and control pipeline is productive, and every artist and tech is passing the torch as planned.
Yet, there's this budget thing.  Over.  Too high.  Bad news.
It's happened to us all...  
the project is over budget.  
Take a moment and consider this scenario:  a company with a staff of 40 artists cranking away on the project.  One by one, as the weeks go by, without anyone noticing, they are all missing their production goals by a measly one hour per week.  Your best compositor needs an extra hour to bring his shot into the sweet zone the vfx supervisor's looking for.  Your modeling supervisor isn't satisfied with the detail on The Big Bad Ugly Monster, so the organic modeling specialist is going to need an hour to fix things up.  UV's are off.  An hour.  The matte painting still looks fake.  An hour.  The camera and layout were not as ordered.  An hour.  The transition from running to jumping when your hero character comes to the rescue is awkward.  An hour.  An hour. An hour. An hour.
"I just need an hour to fix this."  We've all heard it.
Now, our crew of 40 artists and techies has each spent an extra hour to get the work done, so now, your either a week over budget (or should we count that as time  and a half?) or a week behind schedule.  Behind schedule is not an option.
An hour per person per week is, and must be within your budget tolerances.  That's 2.5% of your labor costs, and that should be in your safety margin.  But suppose it's an hour per person outside that safety margin?  You need an extra worker to cover that hour (or, should I say 97.5% of it) per week.
Put another way, each artist who misses production goals by 2 hours a week, adds 2 weeks production cost to a forty week schedule.  Each artist.  Multiply 2 hours of lost productivity by your entire staff and by the average wages paid, and you've got some serious money fast.  It's the death of a 1000 tiny cuts.
 Four hours extra a week = 10%  over 
As a CG supervisor, you need to allow for inefficiency in the workplace, and your company has to decide how much it can tolerate.  This is a subjective and artistic endeavor, after all, so plans can and will go awry.  That's why your producer is doubling any number you give; that's why the Exec Producer should be adding something on top of that.
The issue cannot be solved just with budget padding.  At some point it becomes your problem, a problem you need to face with imagination and diligent watchful management.

Think about it.
I'd like to hear some of the ways you solve this problem.
Please leave a comment 
Best success on your projects!
Thanks for tuning in to The Art of CG Supervision

Thursday, July 9, 2009

#0010 CG Supervisor's Shot Breakdown: Part 4 - The form

Ultimately, the format of your breakdown depends on what information you include or are required to present. What information you should put in your spreadsheet depends entirely on what type of breakdown you are preparing and what method you are using for arriving at a cost estimate. You will need to consider the methods of cost estimation explained in post #0007 and the types of breakdowns discussed in post, #0008. Post #0009 can help you with the issues you should keep in mind when making your choices.

Here we can cover the essentials and offer a few suggestions applicable to most breakdowns; you can adapt this to suit your needs.

The Form of Breakdown Presentation
Use a Spreadsheet not a Text Document
Sometimes I've seen a breakdown presented as a text document, but this is a rare and inconvenient form for others to work with. Even if the text document contains tables, it is clumsy for others to add their own data, tabulate results, project what-if scenarios and the like. So, unless you are using a database or some other software for preparing estimates, you should use a spreadsheet. If you are running a small shop and using Quickbooks or similar software to prepare customer estimates, you should still develop your breakdown in a spreadsheet, for your benefit if not your customers'.

The Essential Data
Minimally, you should have a shot ID #, a description of the shot, a description of the effects work and a labor or cost estimate.

The shot ID number is often taken from the script or assigned to you by a vfx supervisor or producer who has pre-selected what shots you should consider. Unless this ID# is formatted in an unworkable way for your approval system or file naming conventions, you should use it as provided.

You will need a shot description. I usually give every shot both a descriptive phrase of my own and the full description provided. (If you are working with a script, it's on you to figure this out.) The descriptive phrase consists of three to five words, something easy to remember and something that could be written on a white board or in a small spreadsheet cell. The full shot description, if provided by a director or other person, may contain information both about what the shot is about and what you need to do. However, I usually break that into separate data fields in my spreadsheet, because one is about story and one is direction on the work required.

If necessary I will rephrase direction, but I usually prefer to keep the original words of the director intact so I have them when needed later. For this reason, I usually put the director's requests in one column, mine in another, and may allow additional columns if subordinate supervisors are helping prepare the breakdown.

Keeping an eye on the story is important, because it helps you make sure you're meeting shot objectives and continuity between shots. To help the descriptive field, I often have a field with the script segment or transcription of the audio sound bite. Again this has to do with story. However, I would not take the time to transcript the audio from a tape or video clip, this should be provided in a document or you should just leave it blank, it's not a good use of your time. (A coordinator could add this for you later, but it's better if you just can cut and paste it from a script or preliminary breakdown.)

I've experimented with listing characters, props, backgrounds and other scene elements. I've experimented with giving camera direction, lensing and lighting direction. I've experimented with breaking down what tasks are required.

When doing a widget-counting breakdown, I've found that it works best to put the widgets in rows. This way I can collapse and expand rows using subtotal functions to hide data details or drill into the estimate as needed. This leaves columns for each scene, shot or shot shot options.

Use Formulas to Make Your Breakdown Flexible
Put in formulas where needed. When you do this, define variables (elsewhere on the same sheet or make a variables sheet). This will help you play with scenarios when you're in a competitive bidding situation. For example, you could make your labor costs, pads, and so forth variables.

Don't forget to add administrative overhead -- you need to capture the labor costs of coordinators, supervisors, clerks and IT personnel. Custom programming should be anticipated as a line item if you're counting widgets --the point is to be thorough.

If using classification methods you can set-up cells to restrict input to certain phrases or codes and then use formulas to convert those into labor and cost values.

Always pad. I like to pad at a low level, for example, I add a pad to each area of labor: modeling, animation, painting and so on. In this way I've pre-padded my labor. (However, if you tell artists this number, there goes your pad!) I've done it the other way also, a global pad. Sometimes I put in a bit at both levels.

Make it Read
Always make your spreadsheet neat. You need to help your readers quickly comprehend the data and focus their attention on what's important.

For presentation, you don't need to print or display every column and every row. Build it well and by clicking the collapse and open symbols, you can have a single document that allows a summary presentation, a detailed presentation, and a full (and confidential) work presentation. When I use subtotals, I put them above the details (automatic subtotalling puts it below) because then when I collapse the data below, the subtotal remains as a sort of index tab.

Use color and fonts to focus or reduce attention. Use alignment to make text flow better. FOr example, I turn on word wrap for cells with descriptions and instructions, set all cells to align at the top, and the results look great.

Avoid leaving empty rows to create space in your spreadsheet. Instead, control the placement of text vertically and increase the row height. Empty rows make editing the spreadsheet very cumbersome.

Keep Some Things Confidential
Finally remember that an estimate should not reveal too much about your operation. So unless you're presenting the breakdown internally, I suggest you hide most of the details about how you come up with your estimate, giving just enough detail to show you've been competent and build confidence. Print it or export it as a PDF, and send that out to people. If they really require a spreadsheet, copy the data to a new sheet, but lose anything confidential. How you arrive at a bid is part of your company's proprietary strategic method of operation. Do you want your client sharing it with competitive bidders?

Remember, it's YOUR work
Think about what other information is important to you; be flexible and experiment. Err toward too much information. If done properly, your breakdown becomes your database source for managing your show's direction and approval pipeline. You can hand a good breakdown to a coordinator and have that entered into your company's collaboration system and be 90% there.

Keep it useful, keep it need and remember this is a presentation to your client about how well your company functions.

As a final note, remember your breakdown is an intention, an informed guess. If you can tactfully convey this idea to others you will take some pressure off yourself. Because at the end of the day, your breakdown will never match with production reality.

Wednesday, July 8, 2009

CG Supervision Reader Poll: What do you do and who supervises you?

Hello readers.  I'd like to know something about you so I can write for you a little better.

Will you tell me what roles you currently play or played in your most recent job? 
And will can tell me who supervises you or supervised you in the job?

Poll closes in July 18th. 

#0009 CG Supervisor's Shot Breakdown, part 3 - Choosing what's best

As a Computer Graphics supervisor, you'll find choosing a method of cost estimation and a breakdown approach is constrained by various factors, including the information given, project scale, preparation time allowed, your breakdown audience, your objectives and whether your breakdown is given freely or for pay.

In previous posts we examined some approaches to cost estimation: widget counting, difficulty ranking, attribute classification and difficulty ranked attribute classification (#0007) and the types of breakdowns (#0008). Today we will examine the factors that would influence your choice of estimation method and breakdown type. Tomorrow we wrap it up by looking at the presentation of your breakdown.

Considering Limiting Factors
Quality of Information
The major limiting factor in preparing a breakdown will always be the quality of the information available. You may receive the information in the form of a list in a text or spreadsheet. If you get it in any form other than a spreadsheet, start by converting it to a spreadsheet.

Often the list may not contain all the information you need, so you may have to interview the client or director or whomever you need to to get the correct direction. Sometimes all you will have is your interview with the director as you go through the script together.

If you are unlucky enough to be handed a raw script and nothing more, you should go through it with a high lighter, mark any shots you think should get effects or graphics, and mark or tag each page. If you mark it, I suggest a dot in the upper right corner. Small POST-IT flags work well also. Go through it twice if time permits, then take a third pass and enter the data in your spreadsheet.

Of course, if you are lucky enough to be handed a raw script and nothing more, you may discover shots that you could help with effects that someone else may miss. Either these shots will not get visual effects support, will be done a more expensive way, or could be sent to someone else. They could also end up on your lap a week before computer graphics shots are due because someone discovered them in the editing process.

Be especially wary of oral direction at this stage. Do your best to get everything written down. It's a good idea to have a coordinator or producer back you up in when reviewing shots orally so two people can reconcile notes later. If someone else is writing notes it also gives you more time to think about the shots.

Ultimately, your breakdown will only be as good as the information you have. As the saying goes,"garbage in...."

Scale of Project and Time Constraints
As we discussed in post #8, CG Supervisors Shot Breakdown, part 2, the scale of the project may dictate how detailed a breakdown you do and what method you use for estimating costs.  A huge project, or one with more shots than you have time to consider, could compel you use sampling or shot classification methods.  Always the time allowed is a factor.

The audience - who are you giving the breakdown to?

Your audience determines how you structure your breakdown,what methods you can consider using for cost analysis, and how you present the information.  An estimate for an external client may look very different than one you prepare for an internal client, meaning a producer in-house or your boss.
It also will influence how candid you may be about revealing your methodology, how much you limit the scope of the project, what caveats and exceptions you would flag, and so on.

Keep your breakdown objective in mind

Your reason for preparing the breakdown may determine what kind of breakdown you must provide and hence ultimately how much effort you must expend.  If your objective is to win a new client, there may be a deadline for submitting the bid, and even if one is not given there is a de facto business understanding that you need to give an answer quickly.  I use 24 hours as a standard with simple bids and may ask a week on a big project.  Whatever time constraint you have, bear in mind that you don't want to miss the boat.  Ideally, in a competitive situation you don't want to come to the game too late.

Likewise, why you're making the breakdown will determine again how candid you are, how much information you provide, and how careful you are in preparing the cost estimates.  If you are bidding a season of work or a major project, you most certainly need to prepare a labor or resource analysis as well.

A final consideration - who is paying for your time?

You may not think of this, but preparing a good breakdown takes a great deal of time.  If you are working on your own it's part of the cost of doing business, but can you justify spending a week on a bid?  Maybe you need to pare down your breakdown work to make sure you're doing other work that needs to get done.  If you work for a studio, your boss's outlay may be something to consider.  Do you take the time to do a detailed budget when your boss wants a ballpark?  It's his dime, give him what he wants.  You can advise him that a more complete study may turn up some costs you haven't considered - or some savings.

Don't overthink it

All this explanation on how to prepare a breakdown is something you want to absorb but not allow to absorb you.  Ultimately, you have a big job to do as a Computer Graphics Supervisor and while preparing a breakdown is a huge responsibility and can be of vital importance to properly estimating costs and resource needs, ultimately it is a plan of action given what you know at a particular instance in time.  Once work begins, a whole new set of realities will arise that will demand your best skills as a supervisor.

Thanks again for dropping by.

Tuesday, July 7, 2009

#0008 CG Supervisor's Shot Breakdown, part 2 - Types of breakdowns

When breaking down a CG project, the CG Supervisor has several types of breakdowns to choose from, including the ballpark estimate, sampled estimate, production budget, and production plan and resource analysis. Preparing the right kind of breakdown for the situation can reflect greatly on your critical judgment and performance as a CG supervisor.

In my last post we examined some approaches to cost estimation: widget counting, difficulty ranking, attribute classification and difficulty ranked attribute classification (#0007). Today we will examine
the types of breakdowns (#0008). In our next post we will look at how to choose what's best (#0009). In Part 4, (#0010), we will conclude with how to format the breakdown presentation.

What kind of breakdown?
A few good choices
As a CG Supervisor you have several basic choices in your breakdown, including the ballpark estimate, which makes a loose commitment of time and money; the sampled estimate, which uses a portion of the shots as a sample and extrapolates costs from that; the production budget, which could look solely at labor but may include other costs; the production plan, which leaves out cost and focuses only on time; and the resource analysis, which subordinates most details to focus on resource needs. These are a few ways to classify the types of breakdowns; the boundaries may be fuzzy or arbitrary, but it allows us to discuss breakdowns in terms of your objectives and audience.

Essentials of any estimate
No matter what kind of breakdown you are preparing, there are certain essentials that must always be present.

  • Job related specifics: who, what, when due, when work must start (to avoid additional fees), and contact information (yours and theirs). Put your name on it. (Prepared by)
  • Validity period: most estimates expire after 30 or 60 days. Give the current date and the expiration date or period.
  • Scope of work description. It is vital to specify and limit the scope of work.

The ballpark estimate

Most supervisors are familiar with the ballpark. In its simplest form, it's a quick review of the available information and a brief oral statement or email with a number. At this level it's usually the least responsible and least reliable of estimates. What we want to consider is the written ballpark, sometimes called the estimate or budget estimate.

The important thing to consider when preparing a ballpark is whether your audience will view it as a ballpark and respect it as such. It does you no good at all if your audience considers it a firm number. Because the tendency for producers and others to do so will be high, you should remember to clearly state on the document itself and in accompanying documents that this is a rough estimate or ballpark and as such not binding.

However, most producers want something a bit binding they can take to their boss, so you might soften it or quantify just how binding it is. For example, you might say that the estimate is "contingent on material review", "contingent on final shot order", "contingent on footage shot", and so forth; the idea is to leave a back door out. Sometimes you may have a contingency amount which is the amount over the bid you can go given certain circumstances.

When preparing the ballpark, time is usually a factor. The widget counting approach is not advisable. The use of difficulty ranking may be sufficient, but you may want the added assurance you get by using difficulty ranked shot attributes to estimate costs.

The sampled estimate
Really large productions, involving hundreds of shots, will initially give a ballpark based on a sample of shots. I became aware of this when I was bidding a small vfx indy film with a list of 28 shots. Thinking they were ready for the big time, the producers took it to a major vfx post production studio, who met with them and dutifully reviewed the shots. At the end of the meeting the studio remarked that the information provided was sufficient for them to estimate costs, they just needed a total shot count to extrapolate the estimate from the sample of 28 shots.

It's an amusing anecdote, but it revealed to me a method of estimation I had never considered. Sampling.

Essentially one needs a few representative samples of shots that need to be done. It could be as few as one of each type, but three or four is probably better, because you want a range of difficulty. You can then employ whatever method you like, even widget counting, to arrive at a typical shot count. Multiply that by the number of shots with each given effect, and there you have it! I can think of no way to estimate the cost on a major vfx project such as an entire feature film or a television series.

I used this myself when estimating visual effects for season 2 of the UNIVERSE at Flight 33. Given a single episode and a brief discussion about the number of new shots and total minutes of new visuals desired per episode, I was able to work up a fairly reliable estimate for building a CGFX studio and producing 13 episodes.

The production budget
A budget is usually considered a bit more firm than an estimate, but to some degree it's just semantics. An estimate might be given to a client whereas a budget might be an internal presentation.

When preparing a production CG budget, you need to find out what other than labor you are expected to include. For example, it may be you will only include hourly personnel, with salaried personnel being handled separately. I think that's a bit unusual, but if you're not responsible for coordinators and other staff, you might not include them because the producer will be adding them. However, your subordinate supervisors and shot leads might be administrative overhead you need to consider.

Other costs you might not think to include are facilities rent and utilities; burden, an accounting term for payroll taxes and benefits, craft services and so forth. Do you need to include new equipment expenses or depreciation?

When preparing a production budget, you would want to work with the production accountant or controller to determine what you need to include, unless you have a producer telling you. It may be you actually need to get hard numbers from these people. However, unless you're responsible for a facility, you are unlikely to need to do a full CG production budget.

The production plan
A stripped down form of the production budget could be called the CG labor budget or production plan. In this you might only be evaluating the breakdown in terms of labor time. You would then send your spreadsheet on to a producer or production accountant who would then apply hourly rates, salaries, studio overhead and whatever else. This is likely to happen if your supervising a portion of a show or even an entire show at a large studio. In this case the studio wants to keep it's cost and compensation data confidential from you. So just provide what they need, a plan or estimate of labor time.

The resource analysis
A resource analysis differs from a budget in a subtle way, it considers both how long things will take to get done and how long you have and allows you to focus on what personnel and software and machines will be needed to get the work done on schedule. Often, this is an important step in budgeting, especially if you think you may need more resources or overtime to get the job done.

A good resource analysis should breakdown labor by job function. You need X number of modelers for X weeks. Or more precisely, between date A and date B. And so forth.

With a resource analysis, you have a tool to go to management with to request more help as needed, avoid layoffs that might cripple your operation, or restructure your staff needs. It's an important decision making tool.

Other resources
Some organizations publish bidding guidelines that producers and directors in these fields rely on. For example see the Association of Independent Commercial Producers (AICP) guidelines at

This series of articles should prepare the novice CG supervisor to make good breakdowns and could help an experienced supervisor make better ones. Eventually, much depends on your general experience in shot construction and specific understanding of methods and techniques you will use (or rather, your crew will use) to get the job done. remember to consult your experts when you get the slightest bit outside your comfort zone.

If you want further information on estimates, I'll just point out that the search terms, "how to prepare an estimate" returned nearly 12 million hits. It's a vast subject and their's plenty of help available.