« Open Source | Main | Formal Verification Usage with FPGAs »

Comments

Luc Braeckman

Dave,
During my Masters studies, we also took spice models, but our prof really liked Monte Carlo. No, not the nice city in the south of France, but the analysis of the extreme edges of the design you were working on. And hell, in the time that I was studying, the PC's didn't have the power of today's laptops or could be overclocked as the desktops. No we were working on a VAX-VMS that was connected with the university in Leuven (Belgium), and we had a link with that computer running at 64kbit. You can imagine that the graphics weren't great.
Anyway, the models that you stated should do it, but sometimes Monte Carlo can just show something more of the extreme edges the design is working on. Temperature, I/O voltage drift, Core voltage drop, ... it all has some influence. And with the core voltages dropping to close the Vbe (yes I know, today's FPGA's are CMOS based), one can imagine that the margin is closing. Anyone else has an other opinion?

Lattice User

Dave,

are there any plans
to offer faster functional
simulation models for the Lattice
DDR controller IP core ?

As far as I can see
today the netlists are used for simulation.


manish

yes i read it like i have query here you wrote
"The IBIS standard was developed by Intel many years ago in order to allow board-level signal integrity simulations to be performed without the need for a sophisticated SPICE simulation environment. Today, IBIS models are a common customer requirement. How important are IBIS models to you? How do you verify the accuracy of the IBIS models that vendors provide"


so if i have more than one two say various components to integrate to make components then i can inegrate all IBIS model and verify them at the simulator level to check SI before PCB design

The comments to this entry are closed.