Hi Tenchusatsu; Sure it's late at night, but really. The entrance and exit pins for a Rambus channel signal are most certainly not the same signal pin. Consequently, these are two distinct pins, and I counted them correctly. (Okay, I may have missed one or two.)
If we were counting power pins, I could agree with you, all the Vcc pins are the same. But the input and output signal pins are quite distinct.
On the motherboard, the entrance and exit signal pins are on two different traces. On the RIMM module, the two pins have two different names. It matters which way you hook them up, they are not reversible (one at a time). The two pins are not the same pin. They carry the same signal, but delayed in time.
The way you are counting them, we should count the address lines for SDRAM as 1/2 of a signal pin, cause they only go into the DIMM, and don't have to come back out. I mean, really, give me a break. It's really a matter of semantics, but your way of counting "signal pins" is silly. It might make sense if I were counting "signals", but I wasn't.
A more reasonable suggestion would have been to count the total number of motherboard "pin pairs", as that is usually a better indication of routing difficulty than the total number of "signal pins." But this would be unrealistic since most of the Rambus signals have extraordinarily difficult routing requirements.
-- Carl |
