Comments by "zenith parsec" (@zenithparsec) on "How Cell Service Actually Works" video.
-
6
-
3
-
This video is great if you don't know anything about it. At least you aren't going to feel any pain while listening.
I'm not a fan of videos that make people who are new to the topic feel like they've learned something, and people who know something about it feel like they've gotten dumber.
(Edit: especially if after watching the video, the new people can't explain what they learned.)
At 10:37 you literally say there must be some overlap between the cells, but show a diagram where there is no overlap at the intersections.
You get overlap by making the cellular radio signal having enough signal strength that it makes it a larger radius than the radius of the 'hexagons' of the cell. (and those cells are conceptual, and are not actually equally spaced hexagons in anything but the most trivial implementations.. (my excuse for not being clear is this is a comment written as I watch the video. what is yours?)
Having just talked about the microwave transmission, you probably should have explicitly said the overlapping signal needed to be the cellular signal, just to clarify you weren't saying range of the line of sight for the towers needed to overlap. While that seems like a silly thing, remember that this is teaching people who know nothing about it.
Oversimplifying, saying one thing and showing another, and you even drew the frequency spectrum in the reverse of the usual direction.... not attaching numbers to things, except for when you use them and don't say where they come from (e.g. QAM stuff), and the microwave transmission towers you just described are much higher frequency than the "ultra high frequency", but you didn't do a callback to add links to add memory. (Also, you just used stock footage and no call back to the radio signals while talking about the remote towers.)
When you reference a thing you said earlier, people go "Oh yeah, that. Ok.".
(Edit 11:13 to 12:06 You're conflating frequencies with channels. The center frequency has 0 bandwidth. You can't send a signal. It's not that there you are limited to a certain number of physical wavelengths for cellular signals, it's that there are only a fixed number of channels of a certain bandwidth in the range they were allocated. Higher transmission speed needs more bandwidth, all other things being equal. you could have introduced the word 'bandwidth' when you got to FM in the first part. And then you could have talked about how the number of available channels was the width of the available slot divided by the required bandwidth per channel. (You show in this section you're not afraid of doing math with "divide by 7" part.)
I'm not able to continue. It hurts too much. Apparently the good stuff happens next. I'll pretend he did it well.
At least he explained the changes between LTE and 5G, and how the virtualization on commodity hardware of non-radio portions of the infrastructure allows it to scale high speed connections more easily that previous generations did. But what really (i'm guessing remember) stood out was the warning how this boon had the potential to increase the amount and severity of disruptions in disparate locations if, for example, a cloud provider hosting some components had network problems..
Yay.
I need an ibuprofen.
1
-
1
-
1