An architectual dilemma

[JaapvanEkris]

Hi all,

Just a question about an architectual decission I'm trying to make for the next version of the RowingEngine, based on changes in the drag calculation. In essence, the drag calculation gets closer to the MovingFlankDetector. And this poses the question, would it be wise to push the drag calculation into the movingFlankDetector, and change its function entirely to a representation of the flywheel (and thus also make it responsible for calculating angular velocity and torque)? Than the RowingEngine would transform to interpreting those FlyWheel properties into ErgoMeter properties (like strokes, but also linear distance, etc.)?

Alternative is to do all these calculations in RowingEngine, and make that an even more complex object, and let the flankDetector be just that: a way to determine wether the flywheel is powered or not.

Any suggestions to resolve this dilemma are welcome :)

Jaap

[laberning]

Hi Jaap,

cool that you plan to work on this!

I think the cleanest option would be to move that new calculations into a new module, i.e. named Flywheel. So RowingEngine would use Flywheel and that uses MovingFlankDetector to analyse it's state.

We also should consider testability of the Flywheel since we already have seen some quite different flywheels on different rowing machines. So would be really helpful to design it in a way that allows us to automatically test it against raw recordings from flywheels of different rowing machines.

Feel free to PM me to discuss this further.