Panel Discussion: Lessons from Intel’s Open Pluggable Specification

Introduction

In a previous article, we looked at how Intel’s ATX specification continues to be a strong foundation for a multi-vendor PC ecosystem. In this article, we are going to take a look at another computing foundation that Intel established, the Open Pluggable Specification (OPS).

Before 2010, digital signage and other commercial displays could only be paired with their vendor’s compute unit, such as NEC’s Single Board Computer modules and Samsung’s Smart Signage Plug-in Module. The only flexible option for integrators would be to mount external media players or PCs to the backs of their displays and connect the two components with cables.

Responding to this fragmentation, Intel partnered with NEC and Microsoft to introduce OPS. Announced in November 2010, OPS presented a standardized interface between a compute module and a commercial display, giving the market a common foundation. Intel’s approach and goal are strikingly similar to ours. OpenSFF is creating open specifications for modular and scalable small form factor computers, as well as a complementary management software. The story of OPS provides us with useful lessons as we develop our standard.

What OPS defined

OPS defines a fully enclosed compute module measuring 180mm L x 119mm W x 30mm H, and has a single 80-pin JAE TX25 plug that connects to a compatible display’s JAE TX24 socket. DisplayPort, DVI-D/HDMI, audio, USB, UART, control signals, and DC power are all routed through that one interface. This blind-mate connector has the added benefit of rendering cabling unnecessary.

One of OPS’ defining characteristics is its deliberate agnosticism. Despite Intel and Microsoft’s backing, the specification does not require specific processors, operating systems, or software. OPS modules have indeed been powered by a variety of CPUs, from Intel’s own Atom, Core, and even Xeon lines, up to ARM SoCs. Modules that run on Linux or Android are just as common as ones that are based on Windows. This flexibility is one of the primary reasons for its wide adoption. All it requires from display manufacturers is to implement the module slot. Module vendors, system integrators, and customers can customize the compute layer themselves.

The success of OPS

Similar to ATX, OPS achieved a level of adoption that validates a hardware standard. Alongside Sharp NEC, practically the entire industry embraced OPS. Samsung, LG, Philips, and other manufacturers made or are still making OPS-compatible displays, and in many cases, their own OPS modules. Numerous third-party module vendors also emerged, including SMART Technologies, Axiomtek, Yatal Tech, Polywell, and NEXCOM.

A 2012 Intel case study highlights the benefits of OPS’ connector-based approach. Before switching to OPS-compatible products, a cinema chain’s branch at a Singaporean shopping mall employed a combination of physical signage and PCs wired to displays. The chain’s OPS-compatible solution reduced its total cost of ownership by 10% because they no longer needed “lengthy cable runs or HDMI extenders.” NEXCOM’s case study from 2021 about a bank deployment in North America praises OPS as an “ultra-convenient architecture”. The OPS module that NEXCOM deployed has a socketed CPU, making it more upgradeable and serviceable compared to an embedded or discrete solution. The case study also commended the standardized connector, saying that it eliminated the “complex wirings” seen in traditional deployments. A series of SMART Technologies customer testimonial videos from 2024 demonstrates how an OPS module can be more flexible and cost-efficient than a PC. In the videos, the IT manager for a school in Texas explains that their OPS module allows them to easily reposition their interactive whiteboard, not just within a classroom but to different areas of the school. Previously, they were limited by the physical connection between a laptop or desktop PC and displays or projectors that were mounted in place. The IT manager also said that the module was cheaper than a laptop, equating it to purchasing an additional computer at “a deep discount.”

As we will explore in this article, OPS has diverged into smaller branches, and it is not clear whether Intel will continue to maintain the standard or move on. Nevertheless, the OPS ecosystem continues to expand. In 2025, JWIPC announced the S124, an OPS module powered by Intel’s Meteor Lake processors. This year, LG launched an OPS module running Chrome OS, albeit strictly for its interactive whiteboard line. Just last month, Alibaba published a buyer’s guide to help customers decide between purchasing prebuilt OPS modules or going the DIY route.

Beyond its entrenchment, OPS’ sustained success ultimately comes down to the fact that it largely established interoperability between commercial displays and compute modules. The full picture is not as straightforward as customers would prefer, but OPS’ broad and lasting presence is a significant achievement nonetheless.

Why OPS did not completely prevent fragmentation

OPS’ history has its fair share of complications. Some stem from the specification itself, while others are from external factors and parties. These are challenges that anyone developing a modular hardware standard must carefully consider.

Power and thermal budgets

The OPS specification states that a module should receive 12V to 19V DC at up to 8A through its connector. In its OPS primer, module vendor Yatal Tech points out the consequences of the specification’s language. The absence of a power target means there are displays that cannot provide enough power to a number of OPS modules, such as those that have graphics cards. Interestingly, Alibaba’s buyer’s guide explains that there are also interactive flat panel displays that expect to receive power through the OPS module, instead of the other way around.

OPS’ thermal specifications also leave a lot to vendors and users. It simply states that a compatible panel should provide “sufficient airflow” to an installed module, claiming that it is difficult to prescribe a display panel design that can guarantee sufficient cooling for both the module and the display panel itself.

Management platforms

Remote management is essential to digital signage and interactive flat panel displays, as they are often deployed in multiple locations across wide distances. OPS standardized the physical and electrical interface between the module and the display, but not the management layer. This is where OPS’ agnosticism proved to be a roadblock to interoperability.

Naturally, Intel cited its Active Management Technology (AMT) as an out-of-band solution. One of its case studies from 2011 involved a large infrastructure for an appliance vendor in Thailand. OPS allowed the vendor to feed high quality HD videos to its TVs that were on display in retail chains. The infrastructure covered over 800 locations and 9,000 screens. By using OPS modules that were compatible with AMT, the digital signage vendor was able to remotely manage and troubleshoot the installations from a central location.

But OPS itself is architecture-agnostic; without a cross-platform management standard, major OEMs had sufficient justification to develop their own solutions. Examples include Sharp’s NaViSet Administrator, Philips’ CMND, LG’s ConnectedCare, and Samsung’s MagicINFO. These are functional tools with a variety of additional capabilities, such as asset management, content management, and convenient access to official support. They are also lock-in mechanisms that force customers to pair these vendors’ displays only with their respective modules.

Market-specific variants

Only four years after OPS was announced, Intel and JWIPC released OPS-C, an OPS variant created specifically for China’s education and business markets. OPS-C defines a larger module, allowing manufacturers to more easily use more powerful and cost-efficient desktop CPUs instead of being limited to mobile processors.

A year later, in 2015, Intel introduced OPS-C+, which retained OPS-C’s dimensions but supplemented the JAE TX25 with the 40-pin variant of Hirose’s HRS FX18 connector. The additional interface allows OPS-C+ modules to support 4K touchscreen displays.

Yatal Tech’s OPS guide notes that while the OPS-C slot in compatible displays could theoretically accommodate a standard OPS module because it was larger and used the same plug, some panels do not account for the module's mounting brackets. Customers would have to saw off those brackets for the OPS module to fit in the OPS-C slot. The article also states that only a few third-party vendors have adopted OPS-C+. As such, they advise customers with OPS-C+ displays to source a matching module from the display manufacturer—a safe but expensive option, and one that brings customers back to the proprietary landscape that OPS was created to address.

In 2017, Intel released OPS+. It reverts to the module dimensions defined in the original OPS specification, but adds a 60-pin variant of the HRS FX18 connector. As with OPS-C+, OPS+ was created to address modern demands by providing more performance and connectivity, such as support for DisplayPort 1.4 and HDMI 2.0. OPS+ modules can be equipped with Intel Xeon CPUs and drive a single 8K or up to three 4K displays simultaneously. Multiple OPS+ modules can also be interconnected as a server cluster.

The question of succession

It was also in 2017 that Intel introduced Smart Display Module (SDM), a distinct standard aimed at thin or embedded displays. SDM defines small and large variants that are both smaller than an OPS module, and does not specify a module housing. Many display manufacturers and module vendors have adopted SDM for their newer products, including Raspberry Pi, which recently announced an SDM adapter board for the Compute Module 5.

In 2024, EEWorld reported that Intel, CVTE, and Deshengda released OPS 2.0. The new version reportedly enables more performance and a better design compared to the original specification, and integrates OPS-C into OPS. Based on Hirose’s product documentation, OPS 2.0 uses the 120-pin variant of the HRS FX18, which would make it incompatible with original OPS products.

Build with OpenSFF

Like ATX before it, OPS proves that a modular, cross-vendor hardware standard can be genuinely valuable for every party in the ecosystem. Display manufacturers gained flexibility without needing to overhaul their designs as compute improves. Third-party vendors gained a stable entry point, and customers gained a generally interoperable and serviceable solution.

At the same time, OPS’ history shows that ambiguity within the specification can leave customers with compatibility issues that may not be readily apparent. It also tells us that, for certain use cases, standardizing node management is as necessary as defining physical and electrical interfaces. OPS’ pace and direction also reflects the priorities of a sole corporate author, where commercial considerations may play a part in the specification’s evolution.

The story of OPS serves as a mature reference with documented outcomes. It raises important questions for our specification developers: What needs to be strictly defined, and what can be safely left to adopters? How do we build a foundation that balances stability and headroom? How should we structure governance and communicate with members to ensure long-term neutrality? Our work is ongoing, and we are taking our time.

We encourage you to read our specifications, and we would be grateful if you spread the word about OpenSFF. For technical clarifications, partnerships, and other inquiries, reach out to our development team at [email protected].