Hi Wang,
In addition to the difference between DDR2 and DDR3 regarding topologies and
loading conditions for ADD/CMD/Ctrl, ODT also has a difference.
For DDR2 4 DIMMs case, DQs use static ODT in a channel composed of 2 DIMMs. in
this case, whenever you enables ODT in a channel, a DQ in 2 DIMMs has a same
ODT value.
Even though it can help reflection noise to be improved, it can cause small
voltage swing for DC condition. There is a trade-off.
Unlike DDR2, DDR3 supports dynamic ODT to provide more voltage margins while
improving reflection noise. In this case, when you write data at a DQ Rx buffer
in a DIMM, the other DQ Rx buffer can have different ODT value to increase
voltage level. For user applications, some memory vendors provide ODT table
based on the combination of RANK/DIMM.
As Sanjeev said, SI simulation will be helpful to analyze the margin.
Best Regards,
Keeyoung
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On ;
Behalf Of Sanjeev Gupta
Sent: Friday, July 1, 2016 1:39 AM
To: xiaoguang.wang@xxxxxxxxxxxxxx
Cc: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: DDR2/DDR3 module issue help
Hi Wang
You can very well reproduce such scenario in simulation and root-cause issue
with proper models. DDR3 four-DIMM design working without any issue does not
imply that DDR2 design will also work. you need to properly investigate the
issue with your topology. DDR2 has different routing topology (star) compared
to DDR3 (daisy-chain). amount of loading is much higher in DDR2.
Thanks
Sanjeev Gupta
www.sigintegrity-solutions.com
On Wed, Jun 29, 2016 at 2:04 PM, Wang Xiaoguang <
xiaoguang.wang@xxxxxxxxxxxxxx> wrote:
Hi all expert,
I am doing the application test for ddr3 and ddr2 UDIMM module, but
some problem makes things confused:(two platform : INTEL, AMD, both
have 4 module slot ) For ddr2, it always hang up if 4 modules are
pluged in together ,however it can work if one of 4 modules is removed
from the MB (3 modules works), And what's strange is that such fail
can't be improved by change of frequency, VDD, Temperature...
But for DDR3, there is no such problems.
From my side, I think ,
maybe the odt resistance of ddr2 is not ok that makes signal weak at
DRAM interface when work for the full load(4 modues together), and for
ddr3, it can work because there is calibration for ODT, which can help
to adjust the signal quality.
But not sure about the rootcause,
It there some comments/explanation/advise from your experience?
Wait for reply.
Best Regards
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
