|
; U& u+ ~" e& [4 lPersonally i have different opinion on this. Most of the designers thought reduce skew rate is the only major concern on the DDR2,3,4 rounting. This isn't 100% correct. This is because skew rate will be affected when SSO happen. And, if you didn't route at same layer(same reference plane), the coupling noise will be different. Consequently, you will have significant skew rate even you route in same length for those DQ0-7. Thus, it is important to bear in mind that same length doesn't 100% lead to 0 or mininum skew rate. Just my 2 cents from SI and PI point of view. |
|