| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
|
Colleagues, A couple more comment
w.r.t EEE. 1. In 1GBaseT1, the wake time is longer as there are specific slot for wake up to allow efficient
powerdown and in 10GBT, there less of a restriction and thus more difficult to power up the analog in time. 2. In 1GBaseT1, the OAM channel is still fully operational in LPI mode, thus, if we only had low (very low) rate
maintenance data need to send out, we do not need to wake up the link at all. Michael. From: George
Zimmerman [mailto:george@xxxxxxxxxxxxxxxxxxxx] I almost forgot that I promised to look up the EEE timing parameters from 1000BASE-T1. They are 3.6 us for T_s (time that sleep signal is transmitted before transmitter is turned off) and 10.8us for T_w (time to wake – the longest time the system has to wait before transmitting). Generally these will scale with baud. For example, 10GBASE-T is just a little shorter than this (it is 800 Mbaud, just a little faster), but 40GBASE-T is a lot shorter, 0.56us max for T_s, and 1.12us for T_w. George Zimmerman, Ph.D. President & Principal CME Consulting, Inc. Experts in Advanced PHYsical Communications 310-920-3860 |