Re: [Sensor API] editor's edits

Thanks for the info Tran.  The spec looks like it says that the device, at normal/high usage at the max voltage, uses 3.4J in 24 hours?  That type of fine tuning of power usage in a high level orientation api  -- when we already have a way to unregister callbacks -- seems like an overkill.  What am I missing?

Doug


On Mar 11, 2012, at 9:44 AM, Tran, Dzung D wrote:

> Hello,
> 
> Here are some info after talking to the sensor team:
> 
> ------------------------
> The sensor will consume more power if sample rate is higher.
> 
> And some sensor like Accel and Light can set some threshold, that will use interrupt line to notify CPU that some sensor event happen. For example, we already use Accel threshold for Window orientation, that is for each 45 degree, the accel sensor can report a interrupt. For this case, the sensor is retrieved and saved.
> 
> Also, some sensors has low power mode, which you can put in low power state when you are not using.
> For example look at table 9 and table 5 
> http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATASHEET/CD00274221.pdf
> 
> ------------------------
> 
> Thanks
> Tran
> 
> -----Original Message-----
> From: Jonas Sicking [mailto:jonas@sicking.cc] 
> Sent: Saturday, March 10, 2012 12:48 AM
> To: Tran, Dzung D
> Cc: public-device-apis@w3.org
> Subject: Re: [Sensor API] editor's edits
> 
> On Wed, Mar 7, 2012 at 10:06 PM, Tran, Dzung D <dzung.d.tran@intel.com> wrote:
>> Hello All,
>> 
>> Here is the list of edits that I am planning:
>> - Remove Orientation, Gyro and Compass since these are already covered in Geolocation WG.
>> - Remove the "optional" keyword from findSensors(optional DOMString type). This will alleviate concerns about fingerprinting and privacy.
>> - Change SensorError to DOMError
>> - Many minor editing errors
>> 
>> As for WebIntents for discovery, I will leave that till after the F2F's discussion on how to incorporate this into DAP's APIs.
>> 
>> I would welcome any other suggestions.
> 
> Hi Tran,
> 
> I think there's a very interesting discussion going on right now with
> regards to how often to fire the event.
> 
> Doug's assertion is that the sensor is either on or off, and that when
> it's on, it doesn't make it use any more batteries to fire the events
> as quickly as we can get readings.
> 
> If that's the case that's the case then I think we can simplify the
> API considerably since we wouldn't need the watch function or multiple
> SensorConnection objects.
> 
> If that's not the case, then it would be great if someone could step
> forward with data challanging Doug's assertion.
> 
> But I'm not 100% convinced that the device orientation spec covers the
> full range of sensor data. For example you can only get compass
> 'alpha' information. You can't get full alpha/beta/gamma information
> from the compass.
> 
> / Jonas
>

Received on Sunday, 11 March 2012 17:55:31 UTC