Overclock.net › Forums › Industry News › Technology and Science News › [BGR]Driverless cars could be the big thing that vaults Google over Apple
New Posts  All Forums:Forum Nav:

[BGR]Driverless cars could be the big thing that vaults Google over Apple - Page 20

post #191 of 196
Quote:
Originally Posted by 47 Knucklehead View Post

Required? By whom? Oh yeah, the Federal Government. So, once again, we are going back to the LEGALITY of being FORCED to comply with this standard, to eventually replace 260 million cars, at what cost? I'm sure these cars aren't going to be free, and this technology certainly won't be free either. Let's just say that this "Super Technology" only adds $10,000 to the cost of a car (which IMO is ridiculously low), are people going to stand by and pay that much more for this? Who is going to pay the 2.6 TRILLION dollars? Lower insurance premiums? Maybe will help, but unless they give me my insurance for free, I'd still save money in insurance premiums by driving the old way ($1500 over 6.67 years JUST covers the added cost of the technology).
Why would this work differently than any other regulation imposed on cars right now? Chances are this is going to become a standard feature at some point, once it's cheap to implement and more proven. At which point it will be regulated into all new cars, and at some point we'd probably make it illegal to drive on public roads with a non-automated car. That last part is a long ways off though.

As for cost, we don't even have a basis for that yet, you're just making up numbers and saying "LOOK HOW BIG THIS NUMBER I MADE UP IS!". I can point an actual number that car accidents cost somewhere a bit under 260 billion dollars a year in the US (I believe that number includes injuries, auto damage, ect). So if you want to compare you made up 2.6 Trillion number, you can amortize the savings to the economy over 10 years. Though that would be assuming it eliminates all accidents, more likely it would only reduce them by some currently unknown factor.
Micro Dance
(8 items)
 
  
CPUMotherboardGraphicsRAM
i7-4790k z97i-plus EVGA GTX980 SC 16GB Team Xtreme 2400 
Hard DriveCoolingOSCase
512GB Samsung SSD 830 NH-L12 Windows 10 EVGA Hadron Air 
  hide details  
Reply
Micro Dance
(8 items)
 
  
CPUMotherboardGraphicsRAM
i7-4790k z97i-plus EVGA GTX980 SC 16GB Team Xtreme 2400 
Hard DriveCoolingOSCase
512GB Samsung SSD 830 NH-L12 Windows 10 EVGA Hadron Air 
  hide details  
Reply
post #192 of 196
Quote:
Originally Posted by Xeio View Post

Why would this work differently than any other regulation imposed on cars right now? Chances are this is going to become a standard feature at some point, once it's cheap to implement and more proven. At which point it will be regulated into all new cars, and at some point we'd probably make it illegal to drive on public roads with a non-automated car.

Because, as I just got done saying, every single regulation in the auto industry that has happened in the past (be they seat belts, airbags, backup camera, etc) have not required that all EXISTING CARS be retrofitted.

By passing a law that says that all cars on the road MUST be automated, this completely flies in the face of over 100 years of legal precedence.


Quote:
Originally Posted by Xeio View Post

As for cost, we don't even have a basis for that yet, you're just making up numbers and saying "LOOK HOW BIG THIS NUMBER I MADE UP IS!". I can point an actual number that car accidents cost somewhere a bit under 260 billion dollars a year in the US (I believe that number includes injuries, auto damage, ect). So if you want to compare you made up 2.6 Trillion number, you can amortize the savings to the economy over 10 years. Though that would be assuming it eliminates all accidents, more likely it would only reduce them by some currently unknown factor.

You are right, my $10,000 number was made up. It is what I think might be good start, but it is made up. It very well could be $50,000 or $100,000.

As far was your $260 billion number, yup, that can be documents. I won't challenge it, but when you factor that there are 260 million cars in the US, that $260 billion number means that per year, it costs each car $1000. So, we can then divide the ADDED cost of the computer controlled car by the average useful life of that car, and if it's more than $1000 ... then you haven't saved any money ... ASSUMING that with 100% computer controlled cars, there are ZERO accidents. And given that only 1000 miles out of 140,000 miles driven in them, there HAD TO BE human interruption ... because the computer controlled car failed in one way or another.

But hey, at least you are now thinking along the way that I am, by factoring in other costs, and beginning to see this isn't the "wonder item" that some are touting it to be .... like the you don't hear any more about it because it has been shown to not be all that great as first hyped ... Chevy Volt. thumb:
Edited by 47 Knucklehead - 2/1/13 at 11:13am
post #193 of 196
Quote:
Originally Posted by 47 Knucklehead View Post

Because, as I just got done saying, every single regulation in the auto industry that has happened in the past (be they seat belts, airbags, backup camera, etc) have not required that all EXISTING CARS be retrofitted.
Maybe, maybe not. It doesn't really matter as that's far enough out it's purely speculation anyway. Closer is the likely requirement on new cars, but even that is probably 20 years away depending on how quickly this technology develops/how effective it is/how much it costs.
Quote:
Originally Posted by 47 Knucklehead View Post

But hey, at least you are now thinking along the way that I am, by factoring in other costs, and beginning to see this isn't the "wonder item" that some are touting it to be .... like the you don't hear any more about it because it has been shown to not be all that great as first hyped ... Chevy Volt. thumb:
I still think it's a wonder item. Just because it doesn't solve every problem or isn't ready for prime time yet doesn't mean it isn't a awesome.

Also, your statistics are out of date. The 1k/140k numbers (which you take to mean that the driver was basically in-control) are back from 2010, it's up to over 300k as of last year, and there are multiple demonstrations of completely un-overridden driving on streets (including the blind "driver").
Micro Dance
(8 items)
 
  
CPUMotherboardGraphicsRAM
i7-4790k z97i-plus EVGA GTX980 SC 16GB Team Xtreme 2400 
Hard DriveCoolingOSCase
512GB Samsung SSD 830 NH-L12 Windows 10 EVGA Hadron Air 
  hide details  
Reply
Micro Dance
(8 items)
 
  
CPUMotherboardGraphicsRAM
i7-4790k z97i-plus EVGA GTX980 SC 16GB Team Xtreme 2400 
Hard DriveCoolingOSCase
512GB Samsung SSD 830 NH-L12 Windows 10 EVGA Hadron Air 
  hide details  
Reply
post #194 of 196
Quote:
Originally Posted by Xeio View Post

Also, your statistics are out of date. The 1k/140k numbers (which you take to mean that the driver was basically in-control) are back from 2010, it's up to over 300k as of last year, and there are multiple demonstrations of completely un-overridden driving on streets (including the blind "driver").

Agreed. My stats are out of date, they are up to 300,000 now. But somehow I doubt the next 160,000 miles improved all that much from the first 140,000 mile number, 0.7% without "human intervention", but who knows, maybe it got 50 times better since then, and is up to 35% now. I'd love to see the actual numbers.
post #195 of 196
Quote:
Originally Posted by 47 Knucklehead View Post

Agreed. My stats are out of date, they are up to 300,000 now. But somehow I doubt the next 160,000 miles improved all that much from the first 140,000 mile number, 0.7% without "human intervention", but who knows, maybe it got 50 times better since then, and is up to 35% now. I'd love to see the actual numbers.

Do you even code bro? smile.gif

If the AI was pretty good 2-3 years ago....it only gets better and cheaper. Besides, they probably are running simulations on servers first before loading the software into the cars.
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
post #196 of 196
Quote:
Originally Posted by DuckieHo View Post

Do you even code bro? smile.gif

If the AI was pretty good 2-3 years ago....it only gets better and cheaper. Besides, they probably are running simulations on servers first before loading the software into the cars.

Yes. The article I cited was 2 years ago. Hence why I said 50 TIMES as good. I think a 50 times improvement in 2 years is very generous. I only wish that graphic drivers and other software was 50 times as fast, stable, and accurate in only 2 years time. wink.gif

But as I said, I would like to see updated info.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Technology and Science News
Overclock.net › Forums › Industry News › Technology and Science News › [BGR]Driverless cars could be the big thing that vaults Google over Apple