Since March, 2003, Vista Systems of Phoenix, Ariz., hasn't sold a single CanvasMaster display system. Up until that time, CanvasMaster was Vista's flagship solution for creating widescreen presentations. But company president Clark Williams isn't worried about the product's slippage. That's because he understands the reason for the fall — the arrival of Vista's much-ballyhooed widescreen product, Montage. That product, not coincidentally, was introduced last February and has since rocketed to prominence in the staging world for a variety of reasons.

Montage's launch, according to Williams, has been the most successful product launch in Vista's history, therefore, meriting examination. Long before Montage arrived, widescreen presentations were already becoming more common, thanks to advances in projector technology and edge-blending image processors.

But Montage represented a major step forward in the art of widescreen presentations: a 10-bit video-processing and control platform with unique architecture that lets users manipulate unlimited video and graphics inputs in ways that were previously complicated or, in some cases, impossible. In addition, Montage's ability to produce hardware-generated, 3D, beveled borders and drop shadows lets users create more sophisticated presentations more quickly.

In less than 12 months, Montage has become the tool of choice for high-profile events and presentations sponsored by organizations ranging from Microsoft to ABC World News Tonight. Major vendors across the nation have raced to add Montage to their arsenals, despite its steeper price range in comparison to previous technologies in this category.

In the following interview, Clark Williams discusses the reasons for Montage's rapid success and about the future of widescreen presentation technology.

SRO: It seems like there has been an explosion in the use of widescreen displays in the last couple years. What do you attribute this to?

Williams: Widescreen isn't new. It was the way most shows were done back in the mid-1980s and early 1990s. It's just that they were done with slides. The use of widescreen started to trail off when video and PowerPoint became more dominant presentation mediums, because video projectors in their infancy weren't very powerful in terms of brightness. There were a lot of disadvantages to using them, so everybody stayed with this 4:3 aspect ratio screen format, since it was the standardized video format.

But the pent-up demand for widescreen has always been there. And as soon as the projectors started to become bright enough, people began to look at widescreen again.

Our first foray into widescreen video was a couple of products ago, back in the CanvasMaster and ScreenMaster days. They were basically enabling technologies that allowed people to use standard off-the-shelf hardware and software to achieve a widescreen look by letting them easily get an appropriate number of pixels up on multiple screens through the use of edge-blending.

The problem with those earlier products was that they weren't designed from the ground up to handle this kind of application. They were basically “bolt-ons” to our ScreenMaster line. ScreenMaster was originally designed as a multiple screen control system for live events. That allowed you to do some of what a widescreen system needs to do, enough to get people using it. But there were significant limitations.

In contrast, Montage is a product designed from the ground up to overcome all the inherent restrictions in the bolt-on offerings.

SRO: What makes Montage so different from other offerings on the market?

Williams: Without exception, all of those previous generation products, including ours and those of our competitors, have one thing in common: They all employ output-by-output processing schemes, which means that each projector has a processor, and that processor has access to one or two or three inputs. Those inputs can never leave that projector's pixel space. No more than three inputs can ever get into that projector and they can never touch any other projector.

So when you take this output-by-output processing scheme and use it to run three screens, you need three of these boxes, with each providing three little pixel spaces that are running at 1280×1024. And if each box has two inputs, then the maximum inputs for the whole screen is six, and the maximum inputs that can be used on any one screen at any one time is two. Those pixel spaces are never shared.

Another disadvantage of the output-by-output processing scheme is that it makes it difficult to record the whole image. The only place that whole image exists is on the screen. It doesn't exist in any one place electronically. So you can't record it unless you take a camera and point it at the screen.

SRO: And with Montage, you've done away with thisapproach?

Williams. That's right. We decided we wanted all the pixels in one space. The entire image lives in this shared pixel space inside the processor. That was an ambitious goal because it required us to move seven million pixels all to one place. Then we needed to be able to downscale that and record it out.

For the user, the big difference is that an input is no longer restricted to one projector output. It can traverse the entire pixel space. And if it happens to be touching five projectors, it's still just one input. With any other system, to get one image across five projectors, it takes five processor inputs. With Montage, your input can fly all the way across that pixel space no matter how many projectors it touches. It really doesn't care.

So Montage really uses an architecture that is vastly different from anything else on the market and offers serious capabilities beyond anything else on the market. Of course, to be fair, Montage is quite a bit more expensive than these other products. But people don't buy Montage to do what they were doing before. They buy Montage because it can do 50% to 100% more than these other products.

SRO: Besides creating a shared pixel space, what else did you have to do to give Montage more powerful capabilities?

Williams: A lot of the things that it can do, such as borders and drop shadows, are a function of our use of higher performance components. All of our designs are our own. It's our own intellectual property. We designed things based on exactly what our needs are.

A lot of the other products, especially the lower-cost products, use off-the-shelf hardware. We designed our own scaler and filtering and sync circuits. A lot of the other companies buy off-the-shelf hardware that is sold in huge volumes, so they can get it very cheap. For example, they might use a little scaling chip that was designed to go in an LCD panel that costs $17. Now, you can't adjust the registers on a frame-by-frame basis, so you can't do any smooth movement with them, because they weren't designed to do that. But they are $17.

Our scalers, in contrast, are totally in-house intellectual property, so that's what allows us to achieve a level of performance other companies can't. It's also the biggest reason for the price difference.

SRO: How difficult was the development process for Montage?

Williams: Any time you are developing something that's never been done before, it's difficult. And Montage represented a drastic change in the way things were done. We originally anticipated it would be an 18-month development time from kickoff to prototypes, and it ended up being 34 months.

SRO: What kind of customer feedback did you incorporate into the development process?

Williams: Our customers have always been extremely valuable in helping us be sure we are developing the right product. In Montage's case, they contributed a lot of the little details, but the major architecture was kept secret up until within a couple of weeks of product launch.

SRO: So did you do any Beta testing?

Williams: Essentially, what happened was that in its first two weeks, Montage was used in several high-profile events: Intel's Centrino launch and Upfront events for Nickelodeon, MTV, and Discovery Networks. And we had engineers on-site at every one of those events to help use the product, while gathering information to help refine the product.

Real world use is really the litmus test for our products. Our products really only get about 80% done in the laboratory. The rest has to come from the people out there using it. So we go on-site and take notes, and when problems arise, sometimes we fix it right there on-site, and in some cases, we come back to the office and consider what we saw out there and look to make any necessary improvements. We try and stay in tune with customers — understand what are they struggling with and what features aren't getting used because maybe they are too difficult. There might be things the system can do, but they don't actually make it to the screen because the users don't understand the feature or are afraid of the feature. So we try and figure out how to make that better.

SRO: Where does Montage's development go from here?

Williams: If we look at where the market it going, we believe that the new 2K projectors that are coming out soon from all the major projector manufacturers will create an even greater demand for windowing technology. People will want to be able to show not just one image, but five or six images on that pixel space. So what we plan to do is develop a lot of smaller products that are geared around these 2K projectors and do Montage derivative products that are smaller and more affordable.

The biggest thing we hear from our customers right now is that the producers on the big shows love Montage, but when those producers go back to a small show, they still want to put a window over here or do this and that, and you have to tell them you can't do that with [other, older tools], which are all the show budget allows for. So we have a tremendous demand right now for Montage technology in all shapes and sizes.

Our goal with Montage was to secure the high-profile, large-event market using our beachhead customers, and we've already accomplished that ahead of schedule. The next stage is to build smaller, more affordable products. Each one will be the premium product in its category, but they'll be affordable to a lot more people.

Also with 2K projectors, the mechanics of doing widescreen are going to change. What used to be a four-projector show will be a two-projector show. So this whole widescreen look will get easier to accomplish. Sometimes, you won't even need to use edge-blending if you're only using a single projector.

Right now, most people are connecting to their screen via analog. In the near future, we see that changing to almost all digital connections between the processor and the actual projector. We are working on things right now with Digital Projection Inc. [Kennesaw, Ga.] to allow a direct 2000×1080, 10-bit connection over 2 DVI links. That will put up an extremely high-quality image. Most of the projectors are capable of 10-bits and beyond, it's just that none of the outputs feeding them are outputting that

Stephen Porter is a freelance writer who has been covering video, graphics, and digital content creation technologies and applications for more than 15 years.