Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

The most data intensive thing you can stream to a device is video. For two reasons:

1. The whole point of video is that you need all of the data, in order

2. Video is the only thing, really, that requires that much data per second of sensory input.

Mobile devices aren't really getting more pixels. And hardware isn't improving enough the we have spare cycles for decoding/decrypting/rendering 16K video or whatever, and even if it did, the batteries can't handle that. It's obvious to me that we don't need more bandwidth. My phone would be instantly dead if I was downloading and processing data at a gigabit per second.

Other than downloading very large files (why?), I don't think we'll invent any new use cases that can make use of more bits or second.



And carriers limit video bandwidth. I forgot the details, but on certain tmobile plans you got capped to 1080p or a lower resolution, only on the higher plans can you get whatever your phone will support. Let’s the user pay less for less fancy video, and the risk of stirring up the net neutrality pot.


> There is nothing new to be discovered in physics now. All that remains is more and more precise measurement.

Lord Kelvin, 1897


It's not even remotely the same thing. It's a simple fact that there's nothing that requires gigabytes of data on a mobile device that can't be more easily done server side, unless it's real-time. And if it's real-time, the maximum amount of data is the maximum fidelity of data that the device can output (pixels times refresh rate plus lossless audio, either the device specs or its compute ability to drive the hardware). Right now, even if you max out your device's physical ability to output data into the real world, you're not coming close to maxing out your connection. The network isn't the bottleneck, and it's probably not even in the top three.

Investing in more bandwidth is a massive waste on the scale of billions of dollars when you consider that time and capital could be better spent improving existing coverage, lowering latency, and making the service more reliable. Creating a next gen standard that requires new hardware and new devices to support a use case that doesn't exist and can't materialize with today's devices is just silly.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: