| ▲ | illamint 9 hours ago | |
I'm on a 2Gbps/2Gbps (but more like 2.5/2.5) DOCSIS4.0 plan in Seattle and as much as I hate to say it, it's been great. But like, why would they go all the way to XGS-PON when they can use their existing HFC access network, maybe upgrade that to DAA/RPHY, combined with existing subscriber outside plant, and deliver an incremental improvement? As much as I'd love 10Gbps+ service, 2Gbps is already bananas, and there's nothing stopping them from replacing the last mile with glass eventually, either. | ||
| ▲ | martinald 8 hours ago | parent [-] | |
Well firstly, DOCSIS is becoming an increasingly niche technology, and niche technology in telecoms means expensive. There are probably 10,000x more (XGS)G-PON CPEs being manufactured than DOCSIS4.0 ones. I assume similar scale for PON headend equipment. Secondly, PON basically has no powered equipment outside the 'exchange', which can serve 30k+ customers. DOCSIS requires loads of powered equipment, amplifiers, CMTS etc. This then costs money to power (and in high temps, cool) and more importantly breaks down more often. DOCSIS4.0 requires a lot of upgrades to the 'last mile' part so most telcos have decided to just go the whole way and replace with PON instead of doing all that work and getting none of the operational cost savings. The issue is the faster you push DOCSIS the more it shows up any problems in your last mile setup. | ||