The window of opportunity for 25G PON remains open

Omdia view

Summary

Operators are increasingly considering their next move beyond deploying 10G PON technologies. Both 25 and 50G PON solutions are now available in the market. This article examines the complex interplay between the technologies.

Both 25 and 50G PON solutions are commercially available

Nokia has been shipping 25G PON–capable OLT ports as part of its XGS PON line cards since the end of 2020. As of the end of 2024, over 1.7 million 25G PON–capable OLT ports have been deployed, but only a very small percentage (well under 0.5%) of those have 25G PON–capable optics at the OLT. 50G PON deployments have started with limited volumes in China.

Despite 50G PON’s commercial arrival opportunities remain for 25G PON

A simplistic assessment might be that because 50G PON is commercially available and can provide double the downstream capacity compared to 25G PON, 25G PON has missed the market opportunity. Some might argue that operators will now naturally deploy the higher capacity 50G PON technology and the consequence of this would be that Nokia, currently the only vendor shipping 25G PON equipment, is set to miss the market opportunity.

The reality is a lot more interesting and nuanced. In the first instance, part of the rationale for Nokia launching 25G PON was to differentiate its XGS-PON line cards. The vendor would argue with considerable justification that it is achieving this aim, as its XGS-PON OLT port market share has shown positive trends. In addition, more and more operators are asking for future technology roadmaps in RFIs, and the ability to offer 25G PON fits this requirement nicely.

Beyond this, though, the limited volumes of deployments with actual customers being served with 25G PON point to broader challenges in moving the market to multi-gigabit offers. Where such offers are available and realistically priced versus gigabit access, take-up is often modest and could be a low single-digit percentage of an operator’s total FTTP subscriber base. This shows that the business case for further network investments beyond XGS-PON is currently challenging.

In this respect, it could be argued that what is required is a technology that is as easy and as cost-effective to deploy as possible. 25G PON can fit this requirement since existing line cards in the field can support it only by adding new optics. As time goes on, it may then be the case that operators choose to deploy 25G PON for this ease and cost of deployment versus 50G PON, which has higher price points today, requires more complex components, and will need newly deployed line cards and chassis.

Initial 25G PON deployments could help spur growth in the 50G PON market, as real-world examples highlight

However, initial deployments of 25G PON are not necessarily bad for 50G PON; an operator in a market where a competitor has just deployed 25G PON would subsequently have a greater incentive to upgrade to 50G PON to establish a speed advantage, compared to an initial scenario where both operators had deployed only XGS-PON. If this scenario played out and one operator deployed 50G PON, then, again, that might not be a negative for Nokia since the initial operator that had deployed 25G PON would have a greater incentive than before to deploy 50 or 100G PON. Nokia would certainly be in a position to offer 50G PON, as this scenario could happen in the next few years.

There is some real-world evidence of these circumstances occuring. In Hong Kong, challenger operator HKBN is rolling out 25G PON, which will give it an initial speed advantage over competitors. However, in turn, incumbent HKT is likely to be among the first players to roll out 50G PON. HKBN believes it will have a window of 12–15 months where it can offer higher speeds than competitors. In this way, one can see how 25 and 50G PON deployments can feed off each other.

Vendors need to adapt their product offerings to spur growth in 25 and 50G PON rollouts

That said, all vendors are clear that making 50G PON as cost-effective to deploy as possible is hugely important. One aspect of that is the Chinese market, where greater volume deployments have the potential to deliver enhanced economies of scale and reduce costs. However, one caveat here is that while vertical integration of 50G PON chipsets and OLTs/ONTs by Chinese vendors can help reduce costs in some regions, the impact is somewhat constrained because it may not be possible to use Chinese vendors’ chipsets in all regions. Cost reduction for 25G PON is also important, and one aspect of this is improving supplier diversity. Nokia has its 25G PON chipsets, but other vendors will offer their own. Nokia also has its own 25G PON ONTs, but supplier diversity is improving for these devices. On the chipset side, for 25 and 50G PON, it is also worth noting that the move from FPGAs to ASICs can play a significant role in reducing ASPs.

A further element to consider is the most appropriate solutions if the overall demand for these technologies remains constrained by the current modest demand for the very high speeds that 25 and 50G PON solutions can deliver. In this context, 25 and 50G PON solutions that offer greater pay-as-you-grow flexibility for initial smaller volume deployments could be attractive. Pluggable OLTs could play an important role here.

Appendix

Author

Stephen Wilson, Senior Principal Analyst, Broadband Access Intelligence Service

[email protected]