Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Do you really need your iPhone all the time?
- Thread starter houssein31
- Start date
- Sort by reaction score
I forgot to mention this. To those who think an accelerometer alone is enough, you are wrong. I explained that you could get the distance you have moved but that's it. The accelerometer will tell the computer that you have moved 100Ft. If there's no compass, the computer will just add the distances and draw a straight line. That is pretty useless. You need a compass to give these distances, orientation.
And I doubt they would use GPS coordinations to calculate your movement because it's a lot of data for the cpu to process. Maybe the S1 chip uses the accelerometer to track movement and talks to the iPhone to verify the accuracy of the accelerometer. It could explain why they didn't bother putting a gps inside the watch. Also I didn't see the S1 chip to include a compass too, so maybe it's not just seeking gps location but also compass data from the iPhone.
Or else, the simpletons brigade will get you.
It's simple, pick an iPhone, get a GPS fix, put it in a faraday cage (no signal), walk around, and see how the path is recorded.
----------
yes, you know what a compass is? It's a magnetometer, and your phone has it.
----------
Of this... What?
Look, you don't need to understand this, you just need to grab me my burger.
You seem like an intelligent person yet you have not yet grasped the difference between theory and practical application. Your single example is disingenuous at best. I don't need a Faraday cage to test my 5S as it can't track me worth a damn without re acquiring GPS lock frequently. How in the world is a watch swinging in my wrist going to do it? What kind of sampling rate would it need to have and how quickly would errors accumulate as acceleration changes are not sufficiently resolved?
Sure, you can track an object's movement by measuring its acceleration. It's one thing to say that; it's quite another to create a consumer electronic device that can do it reliably. And I'm going to go ahead and say at any price since no valid examples have been provided and no company I know of is claiming it. I'd love to be proven wrong. As a fitness enthusiast, I'd spend good money on it.
All these posts without concrete examples or even a peer reviewed article or two seem to be descending into repetitive pseudo intellectual babble. Saying something over and over does not make it so. If the technology (applied) exists, great! I'd love to hear about it and how it might be incorporated into the Apple Watch.
I'm not wasting time on that, as those are things perfected in the industry, and the industry is not as open as the academy.
Idk how good these devices are on other companies, but let's see how this one fares.
Can't you see these devices operate at a frequency that's not comparable to any human or any animal, or even a traditional mechanical design?
...and you have done this?
Actually you don't need a Faraday cage. Do you even own an iPhone? This shows your basic knowledge of using an iPhone is lacking. You can simply turn GPS off on an iPhone.
Go to Settings: Privacy: Location Services: location Services: toggle offThis turns off the GPS but still uses the accelerometers/gyro (counts steps/milage).
Do this then go for a run or walk and track it. Let us know how well it works and post your tracking map.
Last edited:
Well said!...and you have done this?
Actually you don't need a Faraday cage. Do you even own an iPhone? This shows your basic knowledge of using an iPhone is lacking. You can simply turn GPS off on an iPhone.
This turns off the GPS but still uses the accelerometers/gyro (counts steps/milage).
Do this then go for a run or walk and track it. Let us know how well it works and post your tracking map.
...and you have done this?
Actually you don't need a Faraday cage. Do you even own an iPhone? This shows your basic knowledge of using an iPhone is lacking. You can simply turn GPS off on an iPhone.
This turns off the GPS but still uses the accelerometers/gyro (counts steps/milage).
Do this then go for a run or walk and track it. Let us know how well it works and post your tracking map.
You don't turn off GPS on your phone, you turn "Location" off on your phone.
Big difference.
No, it turns off the location API to app's. Including all data gathered by sensors.
Seriously. Don't try to talk about subjects you don't know.
After Tanegashima's post yesterday (now removed) I was going to leave this thread. But I don't think anyone confident on this subject matter needs to resort to insulting people.
So I thought lets go for a walk, with my iPhone, in a faraday cage..
iPhone 5S, Apple Maps App open, blue dot following me.
Put phone in faraday cage, loose cellular and blue dot stops following me. Take phone out and after few seconds blue dot starts following me again.
Could try and make a video showing this if anyone still has doubts. Although it is probably easier if you just try it yourself if you are interested.
So I thought lets go for a walk, with my iPhone, in a faraday cage..
iPhone 5S, Apple Maps App open, blue dot following me.
Put phone in faraday cage, loose cellular and blue dot stops following me. Take phone out and after few seconds blue dot starts following me again.
Could try and make a video showing this if anyone still has doubts. Although it is probably easier if you just try it yourself if you are interested.
It does NOT turn off the APIs to the accelerometers/gyro which is what you wrongly claim can map your run.
I'll ask agin since you are the only one who believes this. Please provide any proof of this. Just one white paper or study. Just one map made by an accelerometer of a run and the device it was recorded. If you do this the debate will be over. Short of you offering proof of an unknown concept you ARE WRONG.
As proof it doesn't work. Here I started Nike+ and got a GPS lock. I then switched off GPS and walked to my car in a meandering fashion. At my car I switched GPS back on and got a fix, then stopped App. It clearly gathered accelerometer/gyro data as evidence by the distance and pace recorded. However there is NO path recorded on the map. The red dot is about where I parked. Oddly it gave me the starting point only on the web version and the ending point only on the iPhone App. Still proof it doesn't work.
Now I posted proof it can't be done. Now your job is to prove otherwise and not carry on with nonsense or personal attacks. Provide evidence or quite talking about what you don't understand.
Last edited:
It does not turn off those API's.
It turns off to Apple's location API's, that use the access/gyros to locate the device. Then App's use those API's, so Developers can use all Apple's work, so they don't need to analyze the sensor data.
----------
Read this book:
http://www.google.pt/books?hl=en&lr...KfFCFqZ0tbzYvQ7EMckI3FI&redir_esc=y#v=onepage
No, I'm not wrong. And you're going to see it when the apple watch, which completely lacks a GPS, arrives at the market.
Thanks for the great proof that it doesn't work. I guess you FAILED to read it.
Chapter 4 Internal Sensors, 4.1 Accelerometers
...However, most MEMS [accelerometers] offer relatively poor performance....
...Higher grade internal sensors discussed here are used in military ships, subs ICBM’ and space craft…and can costs in excess of $1,000,000 and offers a navigation shift drift of less than 1.8km per day....
4.4 Error Characteristics
...All types of accelerometers and gyro exhibit biases, scale factor and cross-coupling errors, random noise…..
Last edited:
It is an interesting book I agree, here is a part about needing a known starting condition.
Also interesting to see the accuracy and cost of systems used in aircraft.
Last edited:
I was being sarcastic of course, but it is a very interesting read. I bookmarked and hope to read it cover to cover (even if it doesn't literally have covers
).Strange that Tanegashima came up with something so factually accurate and posted as his/her source for misinformation.
I'll throw him/her a bone and offer a white paper that gives some info on accelerometer/gyro tracking with a smart phone. It's not accurate for mapping/tracking a run but at least it's in the right direction for what Tanegashima is saying.
http://www.cs.toronto.edu/~trein/csc2228/files/report.pdf
Too funny that I'm having to do the research for him/her.
You failed to read it!
The book is missing pages from Google preview and you didn't even noticed.
http://en.wikipedia.org/wiki/Kalman_filter (page 91 onwards, and page 117 onwards)
This answers all your quick and dirty and uninformed "research".
I did read it and I was already familiar with the Kalman filter and it use in GPS tracking. It's often called data smoothing. It has nothing to do with or provides any proof of your wacky theory of non GPS tracking/mapping.
As stated in my previous post, I offered you some info that has limited support of your wacky theory. I just can't find anymore supporting evidence to help you out. Sorry but I can't provide you with info of something that doesn't exists. It's up to you to find some hard evidence of it's practical implementation or demonstrated use in any device.
Whataver dude, you aren't an engineer, and on top of it, you hate Apple, so there's no reason for explaining it to you.
In an intellectual debate childish insults are no substitute for empirical evidence or facts.
from what I can gather from apples website they make it seem like the watch will have some form of local storage. And yes it will track your steps so you will be able to see distance just not location.
While it can (probably) estimate your distance in running it could be way off and it can't at all in biking.
In walking steps = distance as long as stride length is known or calibrated. However in running that is not the case. In running as speed increases your number of steps per mile decreases as your stride length increases. So if you run faster the aWatch will say you traveled less distance than you actually did.
For instance the average cadence of a runner is 180. So at 8mph you will have 1350 steps per mile. At 6mph you will have 1800spm. So if the aWatch's accelerometer's algorithms count 1800spm (6MPH) as a mile ran, then at 8mph you would need to actually run 1⅓ miles to register as 1 mile.
This is why accelerometers are poor and not generally used for measuring running distance.
Of course no distance can be estimated when biking.
Last edited:
Are you saying you can go for a run without your phone paired to the watch and then later, when paired, it will show you a map of where you have been running? No, it will not.
----------
Just wanted to make this post so I could have this quote for later...
The Fitbit Surge has a GPS built in, gets a week of battery life, is waterproof, and is less expensive. Apple better figure out this who battery ordeal, especially since we are not getting a GPS.