CPU Blocks S130 nRF51

[nordicdev]

Is it possible to have different characteristic of CPU blocks (I mean different number of CPU blocks or similar) with the same firmware running on two different nRF51 boards connected to app with connection interval of 20msec to 75msec?

[RK]

what are you asking about? What do you mean but CPU blocks?

[nordicdev]

I am talking about the CPU blocks that happen during Bluetooth activity (devzone.nordicsemi.com/.../)

[Stefan Birnir Sverrisson]

Hi

As you mention, the CPU is occasionally occupied (blocked) by softdevice interrupts during BLE transfer. If you have the same firmware on two identical nRF51 peripheral devices, there may still be different CPU blocking time and amount depending on what central devices the peripherals connect to. Android devices for example support different minimum connection interval and differ also in how many packets per connection interval they can transmit. So for instance, if you would have a phone connecting with long connection interval, it would need to send more data per connection interval than another phone supporting shorter connection intervals and transmitting the same amount of data. Sending larger packets or more packets per connection interval normally means longer post-processing time at the end of a connection event, therefore blocking the CPU from the application for a longer time..

There could be other causes for different blocking times than mentioned above, but this at least gives an example how different blocking times can occur for two nRF51 devices running the same firmware.

[nordicdev]

Hi Stefan thanks for the reply. Is it possible for the 2 different nrf51 devices to have different amount of cpu blocks if they are connected to two identical phones? Meaning, same phone manufacturer same model and same android version?

[Stefan Birnir Sverrisson]

I dont see how that could happen. If that is what you are seeing, I guess a sniffer trace would tell you what is the actual difference between the two transmissions.