The Single Point of Failure:
Why Cinema's Future Rests on One Chip

Imagine walking into a multiplex, popcorn in hand, ready for the latest blockbuster. The lights dim, the sound system hums to life, and then... nothing. The screen stays dark. Why? Because a single chip manufacturer, thousands of miles away, decided the cinema market just wasn't profitable enough anymore.

It sounds like the plot of a bad disaster movie, but for those of us in the projection booth and systems integration world, it is a very real, lurking nightmare.

The global cinema industry is currently resting its entire visual future on the shoulders of one company and one technology: Texas Instruments' Digital Micromirror Device (DMD) chip, the core component of their DLP® technology (formally known as Digital Light Processing).^[1] If Texas Instruments (TI) were to pull the plug due to declining cinema demand, the global exhibition industry would be left scrambling in the dark.

And before you say, "That could never happen — an entire industry can't be taken down by one supplier," let me take you back to 2012, when the global automotive industry learned exactly what happens when you put all your eggs in one highly obscure, highly flammable basket.

Section 01The TI Monopoly: The Only Game in Town

Let's get straight to the technical reality. Texas Instruments is the sole inventor and manufacturer of commercial DMDs.^[2] Every single DLP projector — whether it is a Barco, Christie, or Sharp/NEC sitting in your projection booth — relies on TI silicon. There is no second source. There is no backup supplier for reflective or transmissive display technology for cinema screens. TI holds every patent, owns every production line, and controls every DMD that has ever been installed in a commercial cinema anywhere on earth.

According to Texas Instruments' own published figures, nine out of ten cinemas worldwide use DLP technology.^[1] More striking still, 100% of all digital IMAX theaters run on DLP — because every one of them is powered by Barco projectors built around TI's cinema-grade DMD chip. That IMAX-Barco partnership has been exclusive since 2012, and in 2021 IMAX signed a further agreement with Barco to convert over 1,000 of its global systems to laser projection.^[3] Every single one of those systems contains a TI DMD at its core.

We love DLP. It gives us pixel-level sharpness, incredible reliability, and the kind of color stability that keeps audiences coming back. But here is the rub: TI is a massive semiconductor company, and cinema is just one small slice of their pie. With global box office revenues still struggling to return to pre-pandemic levels,^[4] and the broader projection market facing long-term structural challenges, what happens if TI looks at their spreadsheet and decides the R&D and production lines for cinema-grade DMDs are no longer worth the investment?

Section 02The Evonik Crisis: When the Cars Stopped Rolling

To understand the sheer panic of a single-component dependency, look at the Evonik Nylon 12 crisis of 2012. It is perhaps the tightest historical parallel to our current DLP situation, and it is a story that should be required reading for every cinema executive on the planet.

In the automotive world, fuel lines and brake lines are made from a highly specialized, incredibly tough plastic called Nylon 12, or PA-12.^[5] It resists fuel, water, and road salt better than any alternative. Almost every automaker in the world — Ford, General Motors, Toyota, Volkswagen, BMW — depended on it for safety-critical components in virtually every vehicle they built.^[6]

But Nylon 12 is manufactured from an obscure chemical precursor called cyclododecatriene, or CDT.^[5] And in 2012, one company — Evonik Industries of Germany — was the only Nylon 12 producer in the world that made its own CDT. More importantly, Evonik was simultaneously the primary supplier of CDT to every other Nylon 12 manufacturer on earth, including its chief rival Arkema. In other words, Evonik did not just control one link in the chain — it controlled the link that all other links depended upon.

On March 31, 2012, Evonik's CDT plant in Marl, Germany, exploded.^[5]

Instantly, roughly half the world's Nylon 12 production capacity was affected. Because testing and validating new brake and fuel line materials requires months or years of rigorous safety certification, the industry had no backup plan. There was no drop-in replacement waiting on a shelf. There was no second factory that could absorb the volume.

"There was a very strong likelihood at the time that a big portion of this industry would shut down."

— Jay Phillion, Chief Quality and Purchasing Officer, TI Automotive ^[5]

The global automotive industry was thrown into absolute crisis. More than 200 executives from the world's biggest automakers and their suppliers convened an emergency summit in Troy, Michigan, to scour the earth for every last kilogram of Nylon 12 and desperately fast-track alternatives through the approval process.^[6] Through extraordinary coordination, they managed to avoid losing a single vehicle of production. It took Evonik approximately nine months to rebuild the CDT plant and restore the supply chain to normal.^[7]

Section 03Why the Cinema Problem Is Unprecedented

The Evonik explosion is a terrifying case study in supply chain fragility. But the Texas Instruments DLP monopoly is structurally worse in at least two critical ways. Understanding those differences is what makes the cinema situation not just serious, but potentially unprecedented in modern industrial history.

First, the nature of the threat is fundamentally different. The Evonik plant exploded. That was a catastrophe, but it was also a recoverable one. A factory can be rebuilt. Engineers can be rehired. Production can resume. The automotive industry scrambled for nine months, held its breath, and survived.

If Texas Instruments makes a business decision to discontinue cinema-grade DMD production, there is no explosion to clean up. There is no concrete to pour. There is a press release, an end-of-life (EOL) notice, and a date on a calendar after which no more chips will be made.

Second, the concentration of dependency is arguably more severe. In the automotive case, Nylon 12 was critical but it was one material among thousands used in a vehicle. The cinema industry has no such flexibility. The DMD chip is not one component among many in a projector. It is the projector. Without it, there is no image.

Section 04The "Zombie Fab" Problem: Where DMDs Are Actually Made

If you want to understand exactly why a TI exit is a plausible near-term risk rather than a theoretical fear, you have to look at how and where the DMD is manufactured.

The DMD is a Micro-Electro-Mechanical System (MEMS) device. It is not just an electronic circuit; it is a physical machine with millions of microscopic moving mirrors. To build it, TI leverages standard semiconductor infrastructure but adds unique, highly complex MEMS mechanical fabrication steps on top of the silicon.^[8]

The DMD production chain is entirely domestic — and entirely concentrated in a single metropolitan area. Wafer fabrication occurs at TI's DMOS5 facility at 13121 TI Boulevard, Dallas, Texas. Once the wafers leave DMOS5, they travel approximately 30 miles west to TI's DLP Products division in Fort Worth, Texas, where the chips are assembled, hermetically packaged, and tested.^[8]

The hermetic ceramic package is not a standard plastic semiconductor package — it is a specialized, complex enclosure that must be vacuum-sealed to protect the millions of microscopic moving mirrors inside from contamination, moisture, and particulates. This specialized packaging is why DMD back-end manufacturing cannot be shipped to TI's offshore assembly and test facilities in Malaysia or the Philippines, which handle standard analog chips in conventional plastic packages.

The DMOS5 plant was built in 1993. It is a 200-millimeter (200mm) wafer fab running on legacy equipment.^[9] In semiconductor industry parlance, DMOS5 is what is known as a "zombie fab."^[10]

A zombie fab is a facility operating on 200mm or older equipment that is no longer available for sale. Because the equipment cannot be purchased new, the fab cannot be expanded. When a machine breaks down, repairs often require cannibalizing parts from other machines, slowly reducing the number of functioning tools available.

Texas Instruments has a documented, ruthless pattern of shutting down these aging facilities as their new, massive 300mm fabs come online. Between 2020 and 2025, TI systematically closed its older 150mm fabs in North Texas, laying off hundreds of workers.^[11][12] The company is currently executing a $60 billion investment strategy to build massive new 300mm fabs in Sherman and Richardson, Texas.^[13] Those new fabs are built to produce standard analog and embedded processing chips — not MEMS devices like the DMD.

The entire DMD production chain lives in two buildings in the Dallas-Fort Worth metroplex. A single business decision, a single natural disaster, or a single labor action affecting that corridor would eliminate 100% of the world's cinema DMD supply simultaneously.

Would TI invest hundreds of millions to transfer that MEMS process to a new 300mm fab? Let's look at their balance sheet.

The warning signs are flashing red. In the fourth quarter of 2025 alone, the operating profit for TI's "Other" segment collapsed by a staggering 91% year-over-year — plummeting from $82 million to just $7 million.^[16] When a company looks at a legacy fab running legacy equipment to produce a highly complex chip for a segment whose profits are evaporating, the writing is on the wall.

Section 05The Broader Pattern: The Entire Projection Industry Is Contracting

It is crucial to understand that cinema is not facing this crisis in a vacuum. The entire global projection market has been in structural decline for a decade, squeezed by the relentless advance of direct-view LCD and LED panels into corporate, education, and large-venue spaces.

This contraction is forcing suppliers to exit. Sony Semiconductor Solutions has long been the primary manufacturer of High Temperature p-Si (HTPS) BrightEra LCD panels^[17] — the transmissive LCD panels that power 3LCD projectors used by Epson, the world's largest projector brand. But Sony is steadily retreating from the display business. In early 2025, Sony halted all home theater and installation projector sales in Europe.^[18] By January 2026, Sony announced it was divesting its entire TV and display business to a joint venture with TCL.^[19]

Sony's withdrawal from LCD panel supply is the non-cinema equivalent of TI walking away from the DMD. It confirms that the single-supplier dependency problem is industry-wide. As the market shrinks, the component suppliers exit first, and the downstream manufacturers are left stranded.

Even Epson, the undisputed volume king of the projector world, is feeling the pressure. In their FY2025 financial results, Epson reported an approximately 10% year-on-year revenue decline in their projector segment, explicitly citing "a slowdown in the overall market."^[20] Today, TI and Epson are effectively the only two companies supplying projection imaging technology at scale. That duopoly is incredibly fragile.

Section 06The Shrinking TAM: Returning to 1998

To understand why TI might walk away, we have to look at the Total Addressable Market (TAM) for cinema projectors.

There are currently around 220,000 to 240,000 commercial cinema screens operating worldwide.^[21] Approximately 25% of those have already been converted to laser projection. Less than 1% have converted to direct-view LED. That leaves roughly 150,000 screens remaining to be upgraded.

However, the post-pandemic reality is that the total screen count is not going to remain static. As circuits consolidate and underperforming locations close, the global screen count is likely to drop by 10% to 40% through this conversion cycle. This leaves a final addressable market of roughly 100,000 to 120,000 screens.

If projectors capture 80% of that remaining market and LED takes 20%, that leaves approximately 80,000 to 96,000 screens that will require new projectors over the next decade.

Section 07The Laser Illusion and the LED Premium

You might assume that the industry's massive transition from xenon lamps to laser illumination has somehow solved this problem. The uncomfortable truth is exactly the opposite. Every laser projector from Barco, Christie, and Sharp/NEC still uses a TI DMD — the laser is merely the light source. Even Barco's revolutionary new HDR Light Steering technology still relies on the DMD at the core of its optical path.^[22]

Furthermore, the original equipment manufacturers (OEMs) are contractually tethered to TI. Under the terms of the TI DLP Cinema® license agreement, Barco, Christie, and Sharp/NEC are heavily restricted to using the DMD as their display technology for DCI-compliant cinema projectors.^[26] This contractual restriction is a primary reason why these specific OEMs have not aggressively introduced their own alternative LED cinema solutions to date. If TI exits the business, those OEMs would suddenly be free to introduce their own LED solutions — but that transition would take years.

What About Direct-View LED?

Direct-view LED is the only technology that genuinely escapes the TI dependency, because it uses no projector at all. Samsung's Onyx is DCI-certified, supports true HDR, and recently expanded to a 14-meter (46 ft) format.^[23]

However, direct-view LED is a decade-long transition story, not a near-term drop-in replacement. A standard 10-meter Samsung Onyx LED cinema screen typically costs between $400,000 and $800,000.^[24] A comparable laser projector system costs roughly $80,000 to $150,000. That is a 5x to 10x price premium.

Furthermore, LED prices are not dropping rapidly enough to change this math in the near term,^[25] and the audio challenge — solid panels mean no center channel directly behind the screen — remains unresolved at scale. Without a massive influx of HDR content that justifies spending half a million dollars per screen, most circuits will carefully stretch their projection investments.

Projection will win the near-term battle against LED. But that does not help TI. A shrinking market where projection retains 80% share is still a shrinking market.

Section 08The Clock Is Ticking

That is not a supply chain risk. That is a countdown.

There is a counter-argument to this gloomy outlook. Some industry insiders suggest that if TI decides to exit, they would offer a "lifetime buy" — a one-time purchase window allowing OEMs to stockpile enough DMDs to build projectors for the next decade. There are also credible signals that TI is already rationalizing its cinema DMD lineup, consolidating the number of chip variants it offers and eliminating lower-volume, lower-yield models. This kind of product line consolidation is a classic precursor to a managed wind-down — it helps the bottom line in the short term while quietly narrowing the road ahead.

But even a lifetime buy is just a stay of execution. It means the industry would be running on a finite, depleting stockpile of chips with no future roadmap.

When the automotive industry faced the Evonik crisis, they held an emergency summit of 200 executives and organized multiple technical working groups within days. The cinema industry has no equivalent body, no contingency plan, and no active conversation about what happens when the DMD stockpile runs out.

If we wait for the press release announcing the end of DMD production, the screen will have already gone dark.
And this time, there won't be a bulb to replace.