EDA DesignLine Blog
Tell us What You Think
We want to know what you thought about this Discussion. Let us know by adding a comment.
Hogan says retooling underway for custom 2.0
James Hogan
1/11/2013 11:53 AM EST
Custom 2.0 SPICE verification – Precision a requirement
For Custom 2.0, designers no longer find it acceptable to gain increases simulation speed and capacity only at the expense of sacrifices in precision.
Custom 2.0 applications include custom digital, memory, clock and I/O, mixed signal, analog, radio frequencies (RF), microwave, and microelectromechanical systems (MEMS). The analysis complexities are increasing at both the circuit level and the system level. From custom digital to microwave, transient analysis is required. From high-speed I/O’s through RF, various types of noise analyses are needed. From analog through microwave, we need harmonic analysis.
The crux of the Custom 2.0 simulation challenge is that the cost of failure is so great with high volumes at 28nm and below, that simulation coverage becomes even more critical. Even with the increasing use and size of simulation farms, simulation speeds must improve by a factor of 5 to 10, and simulation capacity must improve 100-fold when compared with traditional simulators. And all this must occur without sacrificing accuracy.
The market continues to be fragmented with best-in-class simulators winning over having a one-tool-fits-all simulator for all these different areas, i.e. standard cell, memory, I/O, mixed-signal, analog, RF/microwave, MEMS, and EM. Further, as with Custom 1.0, engineers can be reluctant to change from a simulator that works “well-enough”.
By acquiring Magma in early 2012, Synopsys gained a preliminary monopoly in the custom digital, memory, and standard cell library markets. However, the loss of a second-source SPICE vendor (Magma) in the custom digital, memory, and standard cell library markets immediately opened opportunities for Cadence and Berkeley DA (BDA) to build market share as replacements. Both Cadence and BDA have already added new offerings and gained initial penetration into these markets which are new to both of them. Synopsys resources have now shifted from fighting Magma to integrating recent acquisitions including Magma, SpringSoft and Ciranova, which have tools that target Cadence Virtuoso in the analog/mixed-signal markets.
National Instruments and Ansys both have the cash-on-hand to enter the circuit simulation market, and actually tie in a powerful transistor-level analysis engine to rapidly and accurately analyze system performance and power. An acquisition such as BDA would grant them immediate standing, as BDA has evolved to a going concern with a base of over 100 customers.
For Custom 2.0, designers no longer find it acceptable to gain increases simulation speed and capacity only at the expense of sacrifices in precision.
Custom 2.0 applications include custom digital, memory, clock and I/O, mixed signal, analog, radio frequencies (RF), microwave, and microelectromechanical systems (MEMS). The analysis complexities are increasing at both the circuit level and the system level. From custom digital to microwave, transient analysis is required. From high-speed I/O’s through RF, various types of noise analyses are needed. From analog through microwave, we need harmonic analysis.
The crux of the Custom 2.0 simulation challenge is that the cost of failure is so great with high volumes at 28nm and below, that simulation coverage becomes even more critical. Even with the increasing use and size of simulation farms, simulation speeds must improve by a factor of 5 to 10, and simulation capacity must improve 100-fold when compared with traditional simulators. And all this must occur without sacrificing accuracy.
The market continues to be fragmented with best-in-class simulators winning over having a one-tool-fits-all simulator for all these different areas, i.e. standard cell, memory, I/O, mixed-signal, analog, RF/microwave, MEMS, and EM. Further, as with Custom 1.0, engineers can be reluctant to change from a simulator that works “well-enough”.
By acquiring Magma in early 2012, Synopsys gained a preliminary monopoly in the custom digital, memory, and standard cell library markets. However, the loss of a second-source SPICE vendor (Magma) in the custom digital, memory, and standard cell library markets immediately opened opportunities for Cadence and Berkeley DA (BDA) to build market share as replacements. Both Cadence and BDA have already added new offerings and gained initial penetration into these markets which are new to both of them. Synopsys resources have now shifted from fighting Magma to integrating recent acquisitions including Magma, SpringSoft and Ciranova, which have tools that target Cadence Virtuoso in the analog/mixed-signal markets.
National Instruments and Ansys both have the cash-on-hand to enter the circuit simulation market, and actually tie in a powerful transistor-level analysis engine to rapidly and accurately analyze system performance and power. An acquisition such as BDA would grant them immediate standing, as BDA has evolved to a going concern with a base of over 100 customers.
|
|
|
|
|
Navigate to related information