Greg, Perhaps a CASPER maintainer can correct me if I'm wrong, but I believe I am using the black box, because the files used are .ngo files. There is also a "black box definition" bbd file. The pcore is called ddr2_controller_v2_00_a if that helps any. Thanks, Glenn
On Mon, Sep 8, 2008 at 7:29 AM, Greg Gibeling <[email protected]> wrote: > Am I missing something here? Are you synthesizing the memory > controller from source? Or are you using the black box? > > -Greg > > > -----Original Message----- > > From: [email protected] [mailto:bee2-users- > > [email protected]] On Behalf Of G Jones > > Sent: Monday, September 08, 2008 1:22 AM > > To: BEE2 Users > > Subject: [bee2-users] Difficulty meeting timing with DRAM in ISE 10.1 > > > > Hello, I am cross posting this to the bee2-users list in case anyone > knows > > more about the DRAM controller for the BEE2. I'm not sure if the version > > used by CASPER is outdated, but I appreciate any insight. > > Thank you, > > Glenn > > > > > > ---------- Forwarded message ---------- > > From: G Jones <[email protected]> > > Date: Fri, Sep 5, 2008 at 11:53 PM > > Subject: Difficulty meeting timing with DRAM in 10.1 > > To: [email protected] > > > > > > > > Hello, > > I am migrating my designs from ISE 7.1 to 10.1 and have run into trouble > > meeting timing for designs using DRAM which previously met timing in 7.1. > > Specifically, all of the failing constraints have the following form. > > * NET "ddr2_controller_1/ddr2_contr | MAXDELAY| -0.209ns| > > 0.266ns| 1| 209 > > oller_1/data_path0/data_read_controller0/ | | | > > | | > > dqs_delay1_col1/delay1" MAXDELAY | | | > > | | > > = 0.057 ns | | | > > | | > > > > 36 such constraints are failing in exactly the same way, but some of the > > 0.057 ns constraints succeed. > > I am trying to clock the DRAM at 200 MHz and the system off sys_clk2x = > > 200 MHz, so nothing too extreme. > > I wonder if the timing models became more conservative between 7.1 and > > 10.1? One noticable difference is in 7.1 the voltage was initialized to > > 1.5 V but in 10.1 it's at 1.4 V. > > > > I appreciate any pointers on this problem. > > Thanks, > > Glenn > > > > > >
