Hi Brad,
>>My ire is directed not at Mr. Coluccio, but at whatever fool wrote the original article.<<
Happy to hear that someone's ire is not directed at me. [g]
>>Maybe they were referring to the distribution amplifiers needed for a coaxial cable system. <<
Precisely what I assumed the author meant. Outside plant powering requirements.
OSP Powering is something that will grow geometrically in some respects, due to continuing increases of penetrations and the numbers of services offered; and, due greater emphasis on lifeline and other factors surrounding reliability, once alternative services are introduced to HFC architectures, such as voice and data.
But the offsets will lie in VLSI replacements of components, the increased use of fiber (which requires fewer points of regen/amplification per unit length), and increasing other dependencies on optical technologies, on a per unit of bandwidth measurement basis.
It's also, as I infer you already know, something that is among the thorniest of problems facing terrestrial service providers to solve, where price-performance engineering is concerned.
There are too many variable factors in each of these models to consider, for an author to make a blanket assertion one way or another, I agree. Some of these factors, including that of the "always on" web tone attribute for voice and Internet component, enter this situation, as well. How does one compare one model to the other, in fact, when simplex satellites do not offer the return path that HFCs do, and are relegated to entertainment purposes, almost exclusively?
In order to establish the ground rules (assumptions) in creating a pair of models for comparison, in order to test these theories, we'd need to specify which form of cable TV delivery architecture was being used [straight coax, HFC, FTTC, etc,], the modalities of each user [of which there are bound to be many and varied of, depending on availability of offerings by any individual service provider], units per home, etc.
For example, the powering of straight coax in large un-optimized topologies where they have yet to use fiber, would require far more powering, IMO, than that of HFC, with increasing distances from the head end. Overall DBS model power consumption, in contrast, remains constant.
Staying focused solely on the terrestrial model for a moment, the serving area distribution profile would come into play, as well, where uptake rate and economies of scale would favor high density demographics, and would be very poor for low density areas.
I almost went off into a home grown attempt, here, to develop a model for comparing these attributes, but I'm sure these issues have been covered already, ad nauseam, by the engineering pros with the right modeling tools. Although, I'm not sure if they took the most recent changes in behavioral factors into account yet, such as the changing individual services, which have come to spell out the nature and being of the convergence landscape.
Perhaps Denver Techie or WTC or someone else can point us to some work (urls would be helpful here, folks) that already demonstrate these comparisons.
Thanks for highlighting this point, as it was something that also struck me and lingered for a while in my mind as being too vague.
Regards, Frank Coluccio |
