The primary barrier to simulation, up until the last few years, has been the sheer amount of computing power it takes to handle the hundreds of thousands of calculations that must occur each second in order to simulate even a 2D particle system of a typical finale or other dense show segment. Now that processors have moved up to 2GHz and beyond, with special video accelerator boards being cheap and plentiful, the minimum requirements for a reasonably performing simulation workstation are within everyone's reach. Even a $600 low-end desktop computer fitted with a $70 Ge-Force or ATI based video accelerator card will give you a usable simulation workstation.**

Simulation Advantages

You may initially find yourself asking what advantages simulation based scripting offers over your existing methods. When I started work on ShowSim about three years ago, I thought the primary advantage would be the ability for show operators to visually proof read their shows before firing them.

This is of course one obvious use for simulation, but there are many other advantages that are not immediately apparent. For someone who is new to choreography or just getting into computer firing systems, the visual proof reading advantage really does help out a lot, allowing designers to shorten their learning curve by trying things out on their computer instead of burning through product on the field. Even seasoned choreography experts find the proofing aspect invaluable, which was the case with SeaWorld Orlando when using ShowSim to fine tune their latest pyro theatrical.

Because the intricate close proximity shows used in theme parks need very exact timing to align with the live acts that accompany them, simulation can save a great deal of trial and error with live product. Without simulation, these situations require the shows to be fully rehearsed with live material, video tapped, then adjusted accordingly after watching the tape.

ShowSim features a mock switch-board firing panel that allows operators to fire through a cue list by clicking switches one at a time just as they would when firing the real show. This allows an operator to practice pacing the firing rate before doing the actual show-a feature that new shooters will likely find more valuable than experienced shooters. This feature alone would make simulation a valuable learning tool for training new display operators, or operators who only fire one or two shows per year during the busy season.

The show designers working for Santore & Sons, who have done extensive work with ShowSim over the past year, also found an unexpected benefit of using simulation-which actually lowered the cost of their shows without lowering the quality. An experiment was done where a show was first scripting using their previous "blind" scripting methods with traditional software. The show was then imported into ShowSim and edited to remove overlapping effects that stepped on each other and any other effects that did not contribute anything visually. The result was a 10% reduction in product that could be pared out without any loss in the perceived intensity or overall quality of the show.

The killer feature of simulation, the one that display companies love the most, is the ability to help sell a show by giving the client a preview of what the proposed show will look like. The shows can be presented directly from a laptop running ShowSim, or more commonly from a DVD that has the animation burned onto it so the client can just view it on his own TV. All the descriptive hype in the world can not go up against being able to just watch the show as it will actually look, from start to finish complete with music and sound effects played against the backdrop of a photograph taken from the actual shoot site. When a client can see exactly what he is buying, it tips the hand heavily in your favor relative to other companies bidding for the same contract.

Scripting with Simulation

With simulation comes additional complexity, resulting in a higher learning curve for getting all the performance possible out of a simulation package compared with traditional scripting software. A considerable amount of design effort has gone into ShowSim in order to reduce this complexity as much as possible, but a certain amount of added complexity is unavoidable.

Things like firing positions and background images must be setup for each show, and the viewing distances of these must be set appropriately to get the perspective you want. If the effects you need to use don't exist in the stock library, then you have to learn how to work the effect editors so that you can make them yourself.

The scripting process itself is really no different when using simulation than without it. The familiar process of working with a graphic timeline, cue listing and effect listing are the same as what you may already be used to. You can drop cue markers onto the music graph in real time while listening to the soundtrack, then drag and drop effects from your inventory listing to load the cues. ShowSim goes beyond the limitations of most scripting applications by supporting timer chains and multiple different types or sizes of effects on a single cue. PyroClock users can actually script their timer chains exactly as they will be in the show, and the simulation will handle the timing. The printed reports also reflect the use of timer chains, with a separate report that gives a breakdown of each chain to use as a guide when constructing them.

Where things get complex is when you can not find the effect you need. It is at this point that you must delve into the effect editors and build exactly what you want, and that's when you have to crack the user manual.

Because these editors are required to support a great deal of flexibility in order to handle creating the endless variety of effects that exist, there are quite a few parameters that must be understood. Building cake items can be particularly tedious, since you must first model each effect that is to be fired from the cake in the shell editor, then use a separate cake editor to bring them all together into a mini-script that fires with the proper timing and angles required.

As tedious as this may sound, I just want to point out that I created the 1200+ library of common effects that ship with ShowSim in about one week.

Once you get the hang of things, it does go quite fast. The key to speed is to find effects that are very close to what you want, then clone them and make the small change you need. This way you are not always starting from scratch.

Effect Libraries

In the ideal world of simulation scripting, effect libraries would be available from each manufacturer, with each effect they sell being modeled with the exact timing and effects. The reality of reaching this goal is a daunting task however. The architecture of ShowSim has been designed with this goal in mind, using compact effect definitions that can be packed by the hundreds into tiny text files that are easily distributed. An effect library manager allows manufacturer specific effect packs to be downloaded from a remote server, and the entire system is database driven to eliminate the huge headache that would result from having to manage hundreds of files that would result from a file-based approach to storing effect definitions.

Unfortunately, the reality of the fireworks world is that even if it were humanly possible to create effect libraries for all the cakes and shells for all the various manufacturers, they would quickly be rendered outdated on a yearly basis as the companies keep changing the performance of certain

products from one year to the next. Then you have the issue of inconsistency in timings between different lots of shells even from the same manufacturer within the same year! In the end, the serious choreography companies still have to periodically measure and record rise times, cake durations, gerb burn times and other parameters as new shipments come in, just as they do now.

So while the ultimate goal of having an accurate simulation available for every firework product on the market is an impossible undertaking, it is a noble one that should still be attempted to the degree it is possible. In some cases the manufacturers will have enough incentive to go through the work of creating the effect libraries themselves. This was the case with Santore & Sons, who are both a manufacturer of their own products and a user of ShowSim for scripting their commercial shows. Santore has already built a library representing their own line of products with over 800 effects so that their show designers could work more efficiently. By making this effect library available to all other ShowSim users, Santore is effectively distributing an animated product catalog for show designers to browse and draw from in a very convenient way. It is this win-win situation that I hope will eventually fuel a manufacturer based effort in tackling the never ending task of effect modeling.

From Simulation to Show

One question that may be lingering in the back of your mind is "how do I get my show out of ShowSim and into my hardware?" ShowSim falls into a new class of scripting software I like to call "universal" scripting, which is to say that it can work with a wide variety of hardware without being tied to one specific platform. This approach really makes life easier for the display operator by not tying their software investments into one specific type of hardware. It also makes shows scripted in ShowSim more portable, since the show could be sent to different display companies to be fired without worrying about what hardware they were using.

ShowSim does not communicate with any hardware to directly fire your shows-- your manufacturer specific firing software must handle that task. The way it works is that your show data is converted into one of several different formats that can be read by the firing software that drives your hardware.

FireOne users can output their shows directly to .FIR files without the need for Cue Maker or Script Maker. PyroMate users can create .CSV files that are then loaded into SmartShow to fire the show. In some cases, such as PyroDigital, the data can be fed directly from ShowSim into the hardware via a serial cable or FSK encoded sound file. Systems that ShowSim currently support include FireOne, PyroDigital, PyroMate, Firelinx, Q-Fire and Pyroleda. New formats are periodically added as customers request them, which are then distributed as online updates available to all users.

Note:

** Most notebook computers are still not up to the task of producing good simulations. The reason for this stems from the fact that typical notebook computers have only average video card performance with no way of upgrading to a video accelerator card. Without hardware based acceleration to take the graphics processing burden off of the main CPU, performance will take a big hit. Only the more expensive gaming notebooks are currently up to the task.