Microwaves don’t ionize atomically. So there’s nothing happening there.
do you listen to the radio? Or have a house phone? Or a PlayStation?

You are the epitome of an idiot, I'll pass
![]()
You're hilariously dense! Imagine believing a cotton hoodie is really going to shield a person from non ionizing radiationI mean, you posted this on the last page, but *I'm* the idiot.![]()
So what is the collective solution to fight this? I dont want to hear problems anymore.....everyones deserves SOLUTIONS. Just harping on problems just makes everything worse.
You are severely underestimating what it takes to make autonomous vehicles work. I don't think you really grasp how much data is required and how fast that data needs to be processed and transferred for this whole thing to work.
You're here talking about Waze/Google Maps, comparing how much little data those basic location-based services need to the extreme data-intensive needs of Autonomous Vehicles. AVs rely on a shyt load of data. way more than any Waze/Google Maps would ever need. simply google "how much data do AVs use" AVs rely mostly on multiple sensors. Each one of those sensors on the car is generating a lot of data. You will get a lot of estimates but the consensus is they will generate an enormous amount compared to the small data that Ubers, Waze, Google Maps type apps need like you laid out. We're talking big numbers. Minimum 300-400 GB of data per day and easily much more from just driving a few hours a day.
4G is good enough for stuff like Uber, Waze, etc. they are just basic location-based services (i.e. share real-time status updates, location updates, traffic updates, etc.). But 4G cannot support the processes at the level/scale/speed that AVs will need. Those AVs cannot operate safely if they don't have access to a TON of data and access to that data fast. AVs will need to have damn near human-level reflexes so all that data needs to be processed and transferred extremely fast with very low latency. Lower than what 4G is able to provide. Think about the ability to react quickly to prevent an accident or one car sending data back and forth with multiple cars, traffic lights, reacting to road signs/weather updates, etc.). Think about you as a human driver. All the things you react to from your senses to allow you to drive a car competently. It's a lot more than just basic ass traffic or location GPS updates
Current wifi nor 4G can facilitate handling, processing, analyzing all that data as fast as AVs will need it. How long does it take for a message to get from the sensors in the car, to the computer in the car, then to process/analyze, make a decision, etc.
Decisions are local and sensors will be local - decisions will be local.
Cars are limited in speed. In town you are looking at sub 100 mp/h normally and outside of town nothing in excess of 250mp/h. That's slow.
"Current wifi nor 4G can facilitate handling, processing, analyzing all that data as fast as AVs will need it"?
Ignoring the fact that there is overlap the bulk of the analysis will be local. LOCAL (as it is now). There is no mention of a central server which operates as part of the car REAL-TIME decision making routines (not data). The cars just need to communicate their relative positions, travel vectors, state etc to each other. Data which in most cases they can source themselves via their own primary sensors. Just like we do ourselves. That additional information and any further information are not essential for the act of driving itself - whether automated or not.
The only reason that you need that sort of bandwidth is for video and other commercial services or to enhance safety beyond human senses (around corners, lights intention to change color etc) but again relative to the speeds at which car processing takes place light changing and objects around corners are in slow motion.
The commercial aspect as you open it up so that anything can be sent over the wire (for example adding/removing options to/from your car - engine tuning, suspension modes, running remote diagnostics, running live diagnostics) is a big driver of this. The hire rather than buy model of car "ownership" is a big driver of this. Control by making the car part of the network rather than a local user of the network is a driver. That is NOT the same thing as automated driving itself needing these. If you move control / processing from the car to the network then the network controls whether the car functions or not. If it is just information sharing, then you are talking about marginal gains.
Automated cars already work without remote processing. Now.
You do realise that real-time operating systems need to provide guarantees. No way are those guarantees possible with off-unit remote decisions.
"How long does it take for a message to get from the sensors in the car, to the computer in the car"
The speed of light - a bit or for practical purposes, locally?
![]()
![]()
Compared to processor / bit transmission speeds we are all moving in slow motion.
-
It is not about driving. It is about services, improving capacity (to allow tech to expand) and video.
Remove all video, images and binary data streams from the Internet and we could probably go back to 1G without missing a beat.
Watch this video and boil it down to its essence.
Even the 5G stanners are saying that it's a bit faster than 4G and it only really counts for accident prevention (i.e. seeing around corners et al.).
You say a lot more than "a basic ass message with position info" but that's essentially what a car needs to know ... A message that says where you (other cars) are and which says where you are likely to be next (relative to my car). That can be inferred from a time sequence of position data or can be supplied directly by the supplying car which additional and pre-emptively indicates what is likely to happen next.
There is no list of additional types of data that have to be communicated by 2 cars (or even infrastructure) operating within the same frame that you can name.
funny enough, had a chance to do some cool things at work this year, so I better understand how the deep learning concept works. I was wrong. You were right on all counts.Decisions are local and sensors will be local - decisions will be local.
Cars are limited in speed. In town you are looking at sub 100 mp/h normally and outside of town nothing in excess of 250mp/h. That's slow.
"Current wifi nor 4G can facilitate handling, processing, analyzing all that data as fast as AVs will need it"?
Ignoring the fact that there is overlap the bulk of the analysis will be local. LOCAL (as it is now). There is no mention of a central server which operates as part of the car REAL-TIME decision making routines (not data). The cars just need to communicate their relative positions, travel vectors, state etc to each other. Data which in most cases they can source themselves via their own primary sensors. Just like we do ourselves. That additional information and any further information are not essential for the act of driving itself - whether automated or not.
Autonomous cars work today yes at a basic level. What Level 2? Level 3? Is it the highest level of autonomous like what most people are envisioning when they hear 'self-driving cars'? (e.g. my car picking me up from the airport or my car doing Uber to help pay itself off while I sleep) I don't think so and I don't think you will until you add 5G into the mix. 5G is what gives you that reliability guarantee or SLAs to support those real time OS. And again to your point I understand you would never fully offload the bulk of decision-making to the network. I think it will be a mix.Automated cars already work without remote processing. Now.
You do realise that real-time operating systems need to provide guarantees. No way are those guarantees possible with off-unit remote decisions.
"How long does it take for a message to get from the sensors in the car, to the computer in the car"
The speed of light - a bit or for practical purposes, locally?
maybe for downloading 4k videos it's just a bit faster, but for uses cases that require things like split second decision-making? nah. 5G is a hell of a lot more than "bit faster" 1ms of latency is a hell of a lot lower than 50msEven the 5G stanners are saying that it's a bit faster than 4G and it only really counts for accident prevention (i.e. seeing around corners et al.).
funny enough, had a chance to do some cool things at work this year, so I better understand how the deep learning concept works. I was wrong. You were right on all counts.none of the things 5G enables are critical for the car itself to functionally drive. All the necessary compute to make that happen is done onboard. What the 5G is primarily doing is enhancements, giving that extra level of safety, better predictability to avoid collisions, etc. through the V2X services like mentioned in the video. I still bet in the future some of the realtime decision-making will be offloaded to the edge(not some central server). Especially if the benefit you can gain with better predictability leads to less accidents on the road plus from a cost standpoint - your ability to cut down on the expensive onboard system if you can offload some of that processing.
Autonomous cars work today yes at a basic level. What Level 2? Level 3? Is it the highest level of autonomous like what most people are envisioning when they hear 'self-driving cars'? (e.g. my car picking me up from the airport or my car doing Uber to help pay itself off while I sleep) I don't think so and I don't think you will until you add 5G into the mix. 5G is what gives you that reliability guarantee or SLAs to support those real time OS. And again to your point I understand you would never fully offload the bulk of decision-making to the network. I think it will be a mix.
maybe for downloading 4k videos it's just a bit faster, but for uses cases that require things like split second decision-making? nah. 5G is a hell of a lot more than "bit faster" 1ms of latency is a hell of a lot lower than 50ms