5G – Barrel scraping

Ericsson is scraping the bottom of the barrel for a use case.

• Ericsson has had to use cloud-based virtual reality to demonstrate the benefits of 5G over 4G which clearly indicates that the use cases that are regularly touted to justify investing in 5G, such as digital and autonomous vehicles, make no sense at all.
• Ericsson’s video (see here) uses a cloud VR system where the headset merely sends and receives video signals and the processing of the experience is done in an onsite control room.
• One unit is connected with 4G and the other with 5G both running under optimal conditions.
• When the user moves, the accelerometers send the changes to the control room which processes them and sends the new picture over the wireless signal back to the headset.
• In the video, professional football players then attempt to play football with a real ball using the headsets and it becomes very obvious that the player using the 4G headset is suffering from the 19ms difference in latency between 5G and 4G.
• Ericsson states that the difference between the two is 19ms but RFM’s own measurements have been able to get 4G latency close to 10ms, leading to its opinion that the real difference between 4G and 5G is 9ms.
• It is worth noting that using 19ms rather than 9ms in RFM’s calculations does not in any way change its conclusions with regards to the uses cases that work for 5G and those that do not.
• RFM research has already identified cloud-based VR as a potential use case for 5G (see here).
• This is because, in VR, the virtual environment has to react extremely quickly to the movements of the user in order to avoid nausea and, in this case, to be able to track an object in the real world.
• This is similar to a demonstration done by Nokia in 2016 (see here) that highlighted the importance of low latency in industrial robots that need to be adjusted in real time.
• Industrial IoT is another use case for 5G that RFM has highlighted as viable (see here).
• However, what is missing is the use case for autonomous and digital vehicles and here the argument for 5G badly breaks down.
• This is because RFM has calculated (see here) that all autonomous driving systems are far more latency tolerant than the 5G proponents would like, meaning that 4G could equally do the job as well as 5G.
• This reinforces RFM’s opinion that the drivers for 5G roll-outs remain predominantly the greed and fear of mobile operators who are all desperate not to get left behind.
• This is likely to result in an initial burst of spending in order to claim coverage and then a pause while everyone works out what to use the capacity for and the handset issues are solved.
• Hence, I don’t expect real spending on 5G much before 2022 or 2023.
• This means that there will be a good time to take profits on infrastructure vendors that run up in anticipation of a huge spending splurge that ends up getting meaningfully delayed.