feature article
Subscribe Now

FPGAs Enabling Consumer Electronics – A Growing Trend

It was in the late 1990s that FPGA vendors first started exploring the low-cost, high-volume consumer market – a market that generally was ceded to standard, fixed-function devices. Exploring this market was made possible by the development of FPGAs with the right balance of features and the migration to advanced process nodes which helped to drastically reduce FPGA unit costs.

Historically for high-volume, price-sensitive applications, FPGAs were not the lowest-cost solution. However, as custom logic and FPGAs started to get pad-limited vs. core-limited, this started to change. Pad limitation occurs when the size of the die is determined solely by the number of required I/O pads and not by the amount of logic in the core. Many custom devices such as ASSPs and ASICs had been pad limited for some time. At process geometries below 0.5µ, FPGAs started to get pad limited for the first time. Price is ultimately dictated by die size – when devices get pad limited, the pricing differential between a custom and programmable product begins to narrow.

Today low-cost FPGAs are at the forefront of the process curve with architectures implemented on 90nm process technology. The relentless march down the process curve, coupled with increasing yields on larger wafer sizes (e.g. 12”), has resulted in a dramatic decrease in FPGA costs. For example, the cost per 1000 logic cells for Xilinx 90nm Spartan Series FPGAs has fallen by a factor of 30 since 1998.

FPGA Use Model in Consumer Electronics

The penetration of FPGAs into consumer systems can be broken down into three distinct use models: 1) Glue/Control logic use model; 2) Interconnect logic use model and 3) Core data processing use model.

Figure 1 illustrates how the usage model has expanded over time, as the cost per functionality has gone down.

The sharp drop in cost sparked the use of low-cost FPGAs primarily as glue logic in many high-end consumer systems. Since glue-logic devices are not in the data-path, performance requirements are usually very low. Most of the low-cost FPGAs used for this purpose are slower in speed-grade.

As low-cost FPGAs developed more I/O capabilities – including support for different I/O standards, multiple and independent I/O banks, and multi-voltage support – they were deployed to not only implement the glue/control logic of the system, but also the interconnect logic. Used in this way, the I/O capabilities of FPGAs are effectively used to connect multiple ASICs/ASSPs, memories, processors and other devices – which often have mismatched voltages and I/O standards.

Flat-panel displays are an ideal example of low-cost FPGAs used for interconnect logic as shown in Figure 2. Support for low-swing differential I/O standards (such as reduced swing differential signaling ‘RSDS’) as well as other single-ended standards (such as SSTL), enables the FPGA to interface with LVDS clocks, implement single-ended control signals and directly drive large LCD/plasma screen using RSDS signals.

Glue/control and interconnect logic is by and large the model of FPGA usage in most high-end consumer systems today.

FPGAs as Core Logic – a NEW Use Model

The latest low-cost 90nm FPGAs developed have incorporated platform features that allow the integration of complex DSP, memory-control, complex clocking and embedded processing functions with minimal overhead. These platform features coupled with extremely low price-points, have enabled designers to use FPGAs to implement core system and data processing functions for the first time.

The price points of low cost FPGAs can be measured both in terms of unit costs as well the cost of an implemented function. If the entire FPGA is consumed by a function, then the function cost equals the cost of the FPGA. More often however, the function occupies only a portion of the FPGA, leaving some amount of unused logic for other functions or customization of the design. This type of analysis is especially useful when considering the relative cost of a standalone device such as a soft microprocessor or a complex DSP function (e.g. a JPEG encoder).

For example, the recently announced Xilinx Spartan-3E family was optimized to support embedded processing and digital signal processing functions. In particular, a 32-bit embedded soft processor can be delivered at an effective cost of $0.48 and an 8-bit microcontroller for less than $0.10. In the signal processing realm, Spartan-3E FPGAs deliver a DSP cost/performance ratio of less than $1/GMAC/s.

Some examples of FPGAs increasingly used for data processing in conjunction with a main system controller/DSP device, include:

• An additional soft processor in the FPGA implementing routine data processing, frees the main system controller to focus on other compute intensive jobs.

• Higher performance DSP functions offloaded onto the FPGA (utilizing the embedded multipliers in parallel), permits the use of lower cost DSP to reduce overall system costs.

• DDR memory controller implemented in the FPGA eliminates the need to have an additional standard part, again reducing overall system costs.

In the flat-panel design example, the FPGA can also be used to implement common image processing algorithm, such as:

• SD/HD color space conversion

• 4:4:4 to 4:2:2 down-sampling

• Digital RGB-to-USB card reader function

• Image compression/decompression

Peering Into the Future

There is an emerging class of high-end consumer systems that have comparatively low run rates and/or short market windows. These systems can afford the FPGA price premium – which has shrunk considerably in the last few years – but cannot afford the minimum order quantity (MOQ) or the lead time associated with custom chips.

These systems tend to use FPGAs for the bulk of their core processing and DSP requirements. Today, these systems are only a small percentage of consumer systems. With the consumer electronics industry favoring programmability and field upgradeability to differentiate end products – FPGAs clearly are the ideal choice.

Consumer product manufacturers are forced to experiment with a number of system options before they find a “hit product.” Time to commoditization is also getting shorter as competition is vicious in this field. The programmability of FPGAs reduce these inherent risks and speed time-to-market.

Both these trends indicate that we will see increased use of FPGAs in consumer electronics – as systems manufacturers scramble to introduce a variety of products to the market faster than ever before and as FPGAs deliver more capability at even lower unit costs.

Leave a Reply

featured blogs
Nov 22, 2024
We're providing every session and keynote from Works With 2024 on-demand. It's the only place wireless IoT developers can access hands-on training for free....
Nov 22, 2024
I just saw a video on YouTube'”it's a few very funny minutes from a show by an engineer who transitioned into being a comedian...

featured video

Introducing FPGAi – Innovations Unlocked by AI-enabled FPGAs

Sponsored by Intel

Altera Innovators Day presentation by Ilya Ganusov showing the advantages of FPGAs for implementing AI-based Systems. See additional videos on AI and other Altera Innovators Day in Altera’s YouTube channel playlists.

Learn more about FPGAs for Artificial Intelligence here

featured paper

Quantized Neural Networks for FPGA Inference

Sponsored by Intel

Implementing a low precision network in FPGA hardware for efficient inferencing provides numerous advantages when it comes to meeting demanding specifications. The increased flexibility allows optimization of throughput, overall power consumption, resource usage, device size, TOPs/watt, and deterministic latency. These are important benefits where scaling and efficiency are inherent requirements of the application.

Click to read more

featured chalk talk

Accelerating Tapeouts with Synopsys Cloud and AI
Sponsored by Synopsys
In this episode of Chalk Talk, Amelia Dalton and Vikram Bhatia from Synopsys explore how you can accelerate your next tapeout with Synopsys Cloud and AI. They also discuss new enhancements and customer use cases that leverage AI with hybrid cloud deployment scenarios, and how this platform can help CAD managers and engineers reduce licensing overheads and seamlessly run complex EDA design flows through Synopsys Cloud.
Jul 8, 2024
37,310 views