One lingering question for 2012 is what will become of the Magma back-end platform? I predict that Synopsys will phase out the Magma Talus platform in favor of ICC. Why? It makes no sense for Synopsys to continue to field and support two different systems although it is likely that there will be some transfer of technology into ICC. Converting the existing Talus user base over to ICC is no small task and will likely take several years to complete as well as require incentives and utilities to move the existing base over to the Synopsys platform.
Timing verification is another story. Synopsys will capitalize on the acquisitions of Extreme and Magma to leverage the technologies in those products to develop and deliver the next generation PrimeTime platform. Once they complete this, they will have re-solidified their position as the industry golden standard in static timing verification.
It will be very interesting to see how the consumer-driven SoC market will evolve. SoCs used to be comprised of a processor, memory, various IP blocks, and the on-chip infrastructure needed to support them such as clock, power and communications channels. Now SoCs have multiple processors, large numbers of IP blocks, multiple on-chip communications channels and multiple memories. In essence, today’s SoCs are comprised of multiple SoCs as we used to define them.
The 2012 SoC will beget big challenges in design and even more so in verification. IP will become more important. And even though hardware performance and power will matter, system design and software will become the differentiating items.
SoC system design and verification will be especially active, because it is what the system does that really counts. (After all, the point of building an SoC is to deliver a winning end product.) To a great extent that will require a huge software and verification effort — under the schedule pressures that come from a hugely competitive consumer products market.