feature article
Subscribe Now

FPGAs Control Graphics and Video in Embedded Systems

As I sit at my desk, there are six LCD displays within one meter of my chair: cell phone, desk phone, thermometer, calculator, laptop and external monitor.  Not all devices with LCD displays need much intelligence to display information, such as my desk phone, but the number of systems that display complex graphics and video continues to increase.

Declining display prices and rising user expectations for additional product features and functionality fuel this growth.  Companies in the transportation, automotive, information, automation, medical, industrial and consumer markets use displays in a vast array of products.

The personal computer and consumer markets are the largest users of display systems.  However, the embedded market, consisting of all other electronic market segments that use display systems, has several specific requirements not found in the PC and consumer markets.  These include long product manufacturing life, generic system bus interface and flexibility to adapt to new standards and display types.

Typical Graphics System

The block diagram shown in Figure 1 is an example of a typical graphics or video controller system.  On the left side of the diagram are some of the different input signals that can drive a graphics system.  This example shows a 7:1 LVDS or Channel Link signals, a Society of Motion Picture and Television Engineers  (SMPTE) input signal, a type of interface bus and an NTSC/PAL video decoder.

20071113_lattice_fig1.jpg

Figure 1. Typical Graphics System Architecture

Once the signals enter the system, they are processed by a general-purpose processor or, depending on the architecture, sent directly to the graphics processor.  Additionally, there is Flash memory for program storage and SDRAM for storage of page and video information.

The graphics processor may be an Application Specific Standard Processor (ASSP), a custom Application Specific Integrated Circuit (ASIC) or a Field Programmable Gate Array (FPGA) device.  Depending on the system, there may be multiple displays in the system (as shown), which requires additional logic to manage the signals to support each display.

Developers encounter several barriers when attempting to match an ASSP graphics controller into their embedded design.  These barriers include:

  1. Support for new or derivative video and imager standards in legacy systems cannot be supported in hardwired ASSP or ASIC devices.
  2. Hardwired systems also have difficulty adapting to different display types, due to varying resolutions, aspect ratios and display signal interfaces.
  3. PC and consumer products typically have a very short life cycle; in contrast, the embedded markets have very long production and support requirements that cannot tolerate changes forced by an end-of-life situation.
  4. Most ASSP devices have standard PC bus interfaces such as AGP, PCI and PCIexpress.  Unfortunately for embedded designers, many of the systems they develop do not have these bus interfaces.

While ASICs have an attractively low per piece price, they are very expensive to develop.  The mask charges alone can run in the millions of dollars and if the design is changed to support new standards, the mask charges are incurred yet again.

FPGA Graphics System

The versatile nature of FPGA devices, plus commercially available Intellectual Property (IP), allows for the integration of almost all the graphics system functions.  The video controller, RISC processor, display interfaces, bus standards and different video input standards can be included in one device. 

Figure 2 is an example of the high level of integration possible using FPGAs.  The Channel Link and SMPTE interfaces are processed in the I/O structures of the FPGA.   The RISC processor is included in the form of an IP-based, soft-core 32-bit processor core.  IP for the graphics / video core completes the design with built-in support for touch panel inputs, LCD backlight control, memory controller and multiple displays.  This design greatly reduces the number of parts in the design by integrating all the functions into the FPGA.  By using an FPGA with a built-in single-chip boot Flash, another device (boot PROM) can be eliminated from the design.

20071113_lattice_fig2.jpg

Figure 2. Integrated Graphics Solution

An embedded graphics design using an FPGA device mitigates the issues associated with ASSP and ASIC devices.  FPGAs make it easy to develop a modular design that offers the flexibility to integrate different IP blocks, depending on the cost and feature requirements of the system.  FPGAs enable design integration that typically results in a reduction of board space and parts inventory count.  The long life commitments for FPGA devices ensure they will be available for the operational life of the product.  Most FPGAs are field upgradeable (some devices even while they are operating), which allows for easy upgrades to support new standards and options.

Standard video and graphics IP for FPGAs offer tested and verified designs that are easy to integrate into a graphics solution.  Using packaged IP solutions also speeds system development, enabling engineers to concentrate their efforts on the application rather then the low-level interfaces and graphics engines.  Modular graphics IP cores allow the development and inclusion of custom graphics accelerators within the video controller.  In this way, the graphics performance can be specifically scaled and optimized for the target system.

The intelligent home thermostat shown in Figure 3 provides an example of the use of graphics in a “Smart Home” application.  The Flash-based FPGA has standard IP for the graphics controller, as well as wired and wireless interfaces for communicating with other devices in the home.  The LCD display can be used to display temperature trending information and temperature from remote indoor and outdoor thermometers, as well as video feeds from networked security cameras.

20071113_lattice_fig3.jpg

Figure 3.  Intelligent Home Thermostat

FPGAs offer the versatility to include the optimal bus and interface standards.  This allows the system architect to design with the best interfaces for the entire system, rather than having to design around a specific graphics processor interface or bus.  Designs using FPGAs allow the system designer complete control over the system interface: the entire design can be optimized to support real-time control and display systems that process video streams without any performance degradation.

In addition, FPGA manufactures support density migration within the same sized package.  This allows more logic to be added for enhanced features, or less logic capacity for reduced functionality, without affecting the PCB layout.  Density migration results in cost savings in development, production, servicing and logistics.

FPGAs Enable Flexibility For The Future

The use of FPGA devices to build video and graphics controllers is enabling the embedded market to add graphics display technology in more and more applications.  By using non-PC related video products, the embedded market can rely on having an extended product life cycle well beyond the typical 2- to 3-year period.  One concern for designers has been that the prices of FPGAs were too high to be embedded in systems.  Today, however, FPGAs are designed for low-cost with enhanced features that makes the per piece cost of FPGAs very competitive with other graphics solutions.

I am excited by the additional functionality that LCD displays add to many embedded systems.  The ability to receive additional feedback from a system and have custom interfaces tailored for the individual user is the next step in the evolution of embedded technology.

Leave a Reply

featured blogs
Oct 23, 2020
Processing a component onto a PCB used to be fairly straightforward. Through hole products, a single or double row surface mount with a larger center-line rarely offer unique challenges obtaining a proper solder joint. However, as electronics continue to get smaller and conne...
Oct 23, 2020
[From the last episode: We noted that some inventions, like in-memory compute, aren'€™t intuitive, being driven instead by the math.] We have one more addition to add to our in-memory compute system. Remember that, when we use a regular memory, what goes in is an address '...
Oct 23, 2020
Any suggestions for a 4x4 keypad in which the keys aren'€™t wobbly and you don'€™t have to strike a key dead center for it to make contact?...
Oct 23, 2020
At 11:10am Korean time this morning, Cadence's Elias Fallon delivered one of the keynotes at ISOCC (International System On Chip Conference). It was titled EDA and Machine Learning: The Next Leap... [[ Click on the title to access the full blog on the Cadence Community ...

featured video

Better PPA with Innovus Mixed Placer Technology – Gigaplace XL

Sponsored by Cadence Design Systems

With the increase of on-chip storage elements, it has become extremely time consuming to come up with an optimized floorplan with manual methods. Innovus Implementation’s advanced multi-objective placement technology, GigaPlace XL, provides automation to optimize at scale, concurrent placement of macros, and standard cells for multiple objectives like timing, wirelength, congestion, and power. This technology provides an innovative way to address design productivity along with design quality improvements reducing weeks of manual floorplan time down to a few hours.

Click here for more information about Innovus Implementation System

featured paper

Designing highly efficient, powerful and fast EV charging stations

Sponsored by Texas Instruments

Scaling the necessary power for fast EV charging stations can be challenging. One solution is to use modular power converters stacked in parallel. Learn more in our technical article.

Click here to download the technical article

Featured Chalk Talk

PiezoListen: A New Kind of Speaker for New Applications

Sponsored by Mouser Electronics and TDK

Until recently, putting speakers into extremely space-constrained designs was a daunting challenge. Now, however, advances in piezo speakers bring remarkable performance to ultra-small ultra-thin speakers. In this episode of Chalk Talk, Amelia Dalton chats with Matt Reynolds of TDK about PiezoListen - a whole new kind of high-performance multilayer piezo speaker.

Click here for more information about TDK PiezoListen™ Ultra-Thin Piezo Speakers