To: George Dawson who wrote (12095 ) 2/11/1999 8:19:00 AM From: Frank A. Coluccio Respond to of 42804
George, whereas OC12 was once thought to be the high end of infrastructure hardware (going back about 8 to 10 years), it has now been relegated to a "mux stage" in the greater scheme of things, from a carrier perspective. As an access technology speed, it is still highly viable, especially from the standpoint of CPE gear, but in the carrier domain, it usually resides in a box that is extensible to higher speeds, through the swap of a board, or the flip of a chip. Using multiple discrete devices at OC12 to reach the OC48 threshold is not an elegant way to go. Nevertheless, there are sill many types of situations where OC12 is more than adequate to meet user demands. The costs you cited are not derived in quite the same manner as you (or your reference) has depicted. They don't calculate in a linear fashion, in other words. For the sake of discussion, if you pay $10,000 for a T3, the cost of 3 T3s, or an OC3, is not $30,000, rather it may only be $22,000. The cost of four OC3s, or an OC12, in turn, will not be $120,000, rather it may only cost $75,000. Economies are achieved by compacting many interfaces into a single port, over a single fiber path, as in an OC48 versus multiple (16) OC3s, or 48 individual T3s. Still, these are not chicken feed, although we are seeing steady declines in the cost per OC-n route mile as time progresses, and as more entrants begin to compete on price. Thus far, pricing has not been cost based (rather, it's point of reference has been the voice circuit, an arcane manifestation of monopolies gone by, but still with us, nonetheless). But increasingly, pricing will become more aligned with actual costs, to the point where more affordable provisions will be in place. HTH, Frank C.