Hi Ray,
I don't want to unnecessarily belabor the apparent shortcomings (and the necessary, sometimes ingenious, workarounds) of HFC too much more, here. So I'll keep it, ahem, brief.
No one including myself wants to repeatedly read/write the same thing time and again, I'm sure. For first time lurkers who don't know what I'm referring to, review the past two or three months of discussions here.
These are the dominant points from your message which I walked away with:
(i) you're getting high quality delivery of MPEGs in the digital end of the spectrum (which, in reply to you uncertainty of same, is most likely in the ~500 to 7xx/8xx MHz region, right now);
(ii) you can still stand to use more capacity in the upstream if it were available, but it's not, although it's in the works.
When you say that "one is in fact in the works" referring to a bigger 'bone towards the head end, do you mean that they will be using some form of upstream spectrum allocation plan, such that individual (smaller-sized) clusters of users are assigned their own wavelengths through dwdm splits?
Or, do you mean that they are just now getting around to completing a fiber build for the distribution plant portion of the HFC from the h-e to the field nodes? Please explain.
I would think that you mean the former. If that's the case, do you know whose opticals they are using and what they are intended to do?
I think that --despite the high quality of video and audio MPEG delivery you receive-- your statement about the backhaul (capacity in the upstream direction) is operative, here. In many situations... perhaps yours in the "bonnies" of Bend is one of them... where pops are low (teledensities are not severe) the HFC model will suffice for some time to come, even for as long as the industry is in maintenance mode. Here I use the term "maintenance," as in methadone maintenance. Sorry for using such an analogy, but it makes the point.
If the digital delivery works well, it should. Channels are stacked in a pre-arranged or dynamically allocated manner in accordance with known parameters; the frequency plan is neatly laid out for same; and all is well. Given budgetary concerns in the down- and upstreams, however, it consumes more real estate than it's probably worth to you at some points in time, given where consumers' and home businesses' interests in general are gravitating to, when you are "not" viewing those channels.
Spectrum is limited, at best, due to the upper bounds on the black, and discretionary "head room" is virtually non-existent except for freedoms allowed within the dedicated 6 MHz downstream(s) and the more limited allocations in the upstream. At certain times you would prefer to have the lion's share of the overall pipe, within constraints which are dictated by end system designs (your pc and the targeted web site or other distant target), made available for two way data purposes.
Re: voice, would that be switched voice or packet? I've done some sizing of switched systems for HFCs, comparing those findings with what VoIP would dictate. The concerns here center around a percentage of subscribers that can come on line with voice before over-loading (again, the upstream in particular) becomes an issue. And of course, in some schemes the more upstream voice you implement the more you take away from an already limited capacity previously reserved for data.
The overloading factor is one of the reasons cited why vo/ip is so desirable. It takes up far less nailed up (or even dynamically assigned) bandwidth. Once again, however, what we see here is more balancing and tuning, and a great deal of trade-off engineering and workarounds required in order to compensate for spectrum shortages. In your situation there will likely be no problems implementing voice, but in higher density locales it's an issue if cluster sizes are still high. Which begs the question, do you know how many homes passed exist on your cluster?
Regards, Frank
btw, your attendance record here is really beginning to suck. As an anchor poster you have an obligation to the readers of this thread, so don't stray too far. And if you just happen see those other stragglers from that venerable firm of Lowe, Bemis, Howe, et al, please tap them on their collective shoulders and remind them that the holiday break is over and that the new millennium concerns were largely unfounded. It's now safe for them to come out, again. |
