of compilers and floating point errors

This week has marked a departure from the last few weeks. I fully flowcharted the entire digital stethoscope system and decided on the feature set I want to include in it. Yesterday and the day before were spent plugging away at code, getting one of the filtering routines and a sample buffer to work.

For those who are so inclined, the buffer will implement a ping-pong routine, which alternatively feeds from two buffers through the filter and into another set of two buffers. This way, data flows through the system with very little interruption.

Essentially, one can think of this as a set of two buckets being used to put out a fire. One bucket gets filled while the second is being tossed onto the fire, then that empty bucket gets refilled while the other, now full, bucket gets tossed.

Today's work has consisted of lots and lots of debugging. There is a strange floating-point error in my computer system that makes sin(pi) =/= 0. I am trying to resolve the error in order to successfully test my moving average filter routine with a large dataset produced in the computer. If that checks out alright, I'll begin moving on to other features within the code that need to be accomplished.

Hopefully, tomorrow I will be able to get my hands on the actual evaluation module I will be using in the final project, which will allow me to do further debugging and development.

I apologize for the dryness and length of my post, but a lot has happened since my last post. To make up for this, I have included a picture of where I'm working from, which is not terribly impressive, but it certainly less dry than the rest of this post.