By: rwessel (robertwessel.delete@this.yahoo.com), September 26, 2007 2:26 pm
Room: Moderated Discussions
Jonathan Kang (johnbk@gmail.com) on 9/26/07 wrote:
---------------------------
>This does not, however, get around the fundamental issue of the link having to
>be active at all times. Remember that whatever power figures given for parallel
>links such as CSI are only the active power figures. When comparing it to a serialized
>link, even a low-powered one like the one Rambus has, one has to take into account average power.
OTOH, you can (at least in theory) turn off all but one of the parallel serial links, and still have quick (low latency) access, and the full bandwidth is "only" a few hundred clocks away. Obviously you could do the whole power vs. load tradeoff on a dynamic basis.
---------------------------
>This does not, however, get around the fundamental issue of the link having to
>be active at all times. Remember that whatever power figures given for parallel
>links such as CSI are only the active power figures. When comparing it to a serialized
>link, even a low-powered one like the one Rambus has, one has to take into account average power.
OTOH, you can (at least in theory) turn off all but one of the parallel serial links, and still have quick (low latency) access, and the full bandwidth is "only" a few hundred clocks away. Obviously you could do the whole power vs. load tradeoff on a dynamic basis.