feature article
Subscribe Now

Low Cost Leapfrog

New FPGAs Jump into the High Volume Arena

Imagine watching a professional sports contest (let’s say a basketball game), starting sometime in the middle. In this particular game, there is no visible scoreboard. You watch one team score, then the other. Each seems to be making progress and amassing points, but without a scorecard, you really have no way to tell who’s winning or losing.

There were two exciting announcements in the emerging low-cost, high-volume FPGA market this week. The key issue in this contest is, of course, price. Unfortunately, amidst the flurry of features, claims and specifications, price is the one thing that’s almost impossible to nail down precisely. Each company explains details of the new features, capabilities, and design elements, with strong assurances that these will all add up to more of the kind of gates you need for less money than their competitors charge.

Now, back at our basketball game, suppose you took matters into your own hands and went down to ask one of the teams how they were progressing. Then suppose they gave you an answer like “By halftime in this game, we’re pretty sure we’ll have about 50 points.” When you went to ask the other team how they were doing, they’d reply “…well, as of 10 minutes ago we had 21 points…” While both of these tidbits may be informative and accurate, they really give you no more insight as to who’s winning the game.

As long as both teams are scoring, though, you know that points are being racked up for somebody. In the case of low-cost FPGAs, that can only be good news for the customers. No matter who’s ahead, the contest is heating up and competition improves the breed. This week, the competitive spotlight is on Lattice and Altera with their new lines of low-cost programmable penny-pinchers.

First up (and first with samples available) is Lattice Semiconductor with their new Lattice-EC and Lattice ECP-DSP families. Delivering on the expectations they set a couple of months ago with their newly-announced Fujitsu fab agreement, this is clearly a breakthrough announcement for Lattice. These are highly-competitive products targeted at a new, rapid growth market where the playing field is much more level than in the traditional FPGA space. While Lattice has worked to make their presence known in FPGA before, they have primarily succeeded in the CPLD market. This announcement represents their best effort by far to field a competitive FPGA offering.

Low-cost market segments like consumer electronics have not traditionally made extensive use of FPGAs. They have, however, used large volumes of CPLDs, so Lattice’s experience selling CPLDs to that market will help them with their new FPGA line. While other vendors have ground their teeth deciding whether or not to put high-performance hard-wired multipliers on their low-cost devices, Lattice decided to offer both flavors and let the customer choose. The ECP-DSP line (“Economy Plus DSP”) includes DSP blocks that can implement up to 36X36 multipliers and 18X18MACs, as well as summation and pipelining for those wanting DSP functions in their low-cost FPGA. The EC (“Economy”) line is cost-optimized without the multipliers for those who just want the most gates for their buck.

The new lines are based on Fujitsu’s Low-k, copper, 130nm CMOS process and use a 1.2V power supply. The devices are optimized for small die-size which should give them very competitive pricing, even compared with 90nm offerings (which are still suffering from shrinking pains as the kinks are ironed out of the process).

In the spirit of reducing total system cost, Lattice has designed the new family to be configurable by low-cost SPI flash devices. Lattice claims that this can offer 4 times lower cost per bit than proprietary boot PROMs. The density range of the new family is from 1.5K to 41K (4-input) LUTs. The devices are offered in low-cost packages including TQFP, PQFP and fpBGA in pin counts from 67 to 576.

Lattice’s press release gives specific pricing information, but take note – different vendors give pricing based on different volumes and different delivery dates. (Remember our basketball game?) Don’t compare prices for 250K volume in mid-2005 with Lattice’s 1K volume in mid-2004. Pricing varies with both volume and date as new processes come online and yields improve, so be sure you’re comparing apples to apples.

Next up this week Altera is also raising their bid in the low-cost arena. Altera is announcing a 90nm sequel to their wildly successful “Cyclone” line. “Cyclone II” offers higher density and lower cost than its predecessor and keeps Altera in a strong position in the econo-race as well. Altera claims the new family will be 30% lower cost than Cyclone and will offer 3X the density.

Altera prefaces the announcement of the new family with a history of the commercial successes of the original Cyclone line, and the list is impressive. With examples such as prosumer video cameras, GPS, fish-finders, wireless networking, digital TV, and consumer PC video products, Altera is showing that the storm of FPGA technology into the consumer arena has definitely begun, and Altera products are in it.

While everyone knew that price/density curves would drop to the point that overall time-to-market advantages and NRE savings would make FPGAs compelling options for consumer applications, these devices are making innovative use of reprogrammability in ways that render them clearly superior to ASIC implementations for these products. Pinnacle Systems effectively triples the density of the FPGA by reconfiguring it with one of 3 different bitstreams depending on the operating mode of their MovieBox Deluxe. In addition, they configure the device from a bitstream embedded in Windows drivers, so they are almost immune to hardware design errors as they can offer hardware patches and upgrades to existing customers via a simple download.

The Cyclone II family, scheduled for sampling in early 2005, is designed to surf the coming tsunami of high-volume, low-cost FPGA sockets with maximum efficiency. In addition to the size and cost advantages, Cyclone II will offer 4X the memory of Cyclone and includes up to 150 embedded 18X18 multipliers. The devices also include dedicated DDR2 and QDRII interface circuitry to connect to state-of-the-art external memory with 668-Mbps performance. Internal clock speeds are rated at 250MHz which is more than enough for most mainstream applications. In addition, the new device supports Altera’s Nios II soft-core processor making Cyclone II a super-low-cost embedded system design platform.

Unlike rival Xilinx (who started their 90nm rollout with their low-cost Spartan 3 family), Altera went to production first with its flagship Stratix II line, then based their low-cost offering on the same process. This should mean fewer skinned knees and bruised elbows as they fight their way into production with the new line. Also, by releasing their low-cost 90nm offering after the competition, they had a known target, so they are able to differentiate themselves without guessing.

Cyclone II, which is based in TSMC’s 90nm, low-k process, is scheduled for first delivery in February 2005 with the EP2C35 device. Additional family members are supposed to be in place by the second half of 2005.

The addition of these latest Lattice and Altera lines heat up an already hotly contested low-cost logic derby being fought by Xilinx’s Spartan 3, Altera’s Cyclone, and Actel’s ProASIC Plus families. This is a serious business with serious stakes, and we should watch for more entries in the coming months as vendors plan their strategies in this emerging and lucrative logic market.

Who’s going to win? Well, a scoreboard would certainly be nice right now, wouldn’t it? Unfortunately, with the primary indicator, (price for a given feature mix at a given volume on a given date,) still up in the air, it will be hard to say for awhile. For now, you’ll have to carefully evaluate your needs including feature set, delivery dates, and production volume, against the available offerings and make your decision from there. It will also pay to re-evaluate your options often as this market is destined to change rapidly, and most of the changes will be in your favor.

Leave a Reply

featured blogs
Apr 14, 2021
You put your design through a multitude of tools for various transformations. Going back to formal verification in between every change to rely on your simulation tools can be a rigorous approach,... [[ Click on the title to access the full blog on the Cadence Community site...
Apr 14, 2021
Hybrid Cloud architecture enables innovation in AI chip design; learn how our partnership with IBM combines the best in EDA & HPC to improve AI performance. The post Synopsys and IBM Research: Driving Real Progress in Large-Scale AI Silicon and Implementing a Hybrid Clou...
Apr 13, 2021
The human brain is very good at understanding the world around us.  An everyday example can be found when driving a car.  An experienced driver will be able to judge how large their car is, and how close they can approach an obstacle.  The driver does not need ...
Apr 12, 2021
The Semiconductor Ecosystem- It is the definition of 'High Tech', but it isn't just about… The post Calibre and the Semiconductor Ecosystem appeared first on Design with Calibre....

featured video

Meeting Cloud Data Bandwidth Requirements with HPC IP

Sponsored by Synopsys

As people continue to work remotely, demands on cloud data centers have never been higher. Chip designers for high-performance computing (HPC) SoCs are looking to new and innovative IP to meet their bandwidth, capacity, and security needs.

Click here for more information

featured paper

From Chips to Ships, Solve Them All With HFSS

Sponsored by Ansys

There are virtually no limits to the design challenges that can be solved with Ansys HFSS and the new HFSS Mesh Fusion technology! Check out this blog to know what the latest innovation in HFSS 2021 can do for you.

Click here to read the blog post

Featured Chalk Talk

Transforming 400V Power for SELV Systems

Sponsored by Mouser Electronics and Vicor

Converting from distribution-friendly voltages like 400V down to locally-useful voltages can be a tough engineering challenge. In SELV systems, many teams turn to BCM converter modules because of their efficiency, form factor, and ease of design-in. In this episode of Chalk Talk, Amelia Dalton chats with Ian Masza of Vicor about transforming 400V into power for SELV systems.

Click here for more information about Products by Vicor