| Hi Joakim,
Thanks - that's a huge help. I hadn't read that document in a while, and I probably wouldn't have thought to go back there.
For what it's worth, the documentation seems to differ from the implementation. The connection interval unit of measure used in the documentation is 1ms, but the bluetooth protocol uses a unit of 1.25ms. As a result, the documented minimum connection interval of 16 (20ms), is implemented as 20 (50ms).
The flexibility to go down to 20ms as documented would be great. Is this a bug reporter thing? Has anyone else seen this?
Thanks, Peter Skinner
Ten One Design 201-965-0200 On Dec 12, 2011, at 1:09 AM, Joakim Linde wrote:
Hi Peter,
Sorry to hear that you are having issues. The iPhone requires that the requested connection parameters follow a few rules. These rules are documented in the Bluetooth Accessory Design Guidelines for Apple Products, available at:
On page 18 you'll find information about connection parameters. The rules are in place to enhance the user's experience with Bluetooth LE. For instance, the iPhone has to be able to handle Bluetooth Headset voice traffic and WiFi traffic at the same time as maintaining a Bluetooth LE connection.
The Bluetooth Accessory Design Guidelines for Apple Products is highly recommended reading to all accessory developers.
Let me know how it goes or if you have any questions or comments.
Thanks, Joakim
On Dec 11, 2011, at 7:35 PM, Peter Skinner wrote: Hello,
I have been working with CoreBluetooth on the iPhone 4s (iOS 5.1 beta), and I've spotted something strange.
The connection interval defaults to 0x54 (105ms). This is a great starting point, but could be longer to help the peripheral save energy, or shorter to increase data throughput.
After a connection is made, the 4.0 spec allows the slave device to send a L2CAP Connection Parameter Update Request to change the connection interval. The master device (iPhone) can then either accept or reject the new parameters. I've found the iPhone will always choose to reject the new parameters, regardless of their value, by sending response code 0x01. It's almost as if this portion of the protocol isn't fully implemented yet.
I could certainly be missing something here. Any insight would be greatly appreciated.
Thanks,
Peter Skinner
Ten One Design 201-965-0200
Do not post admin requests to the list. They will be ignored.
Bluetooth-dev mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden
|