Closed MatthewGrim closed 6 years ago
Testing the output, they are in the wrong format ... lighting is in epoc seconds when it should be in YYYY::MM:DD::hh::ss and the range reports are not generating the right information either
The problem is not insurmountable ... the conversion from epoch seconds to seconds is just an offset. But I need to know the simulation start time in epoch seconds to make this work.
For the range, the data is the issue - 30% of the way through, we're generated 8GB. We're looking at a huge amount of data, and it'll be due to the range reports. I may need to restart because of this.
On the plus side, simulations match, so numerically we are doing what we think we are doing. I now need to solve the unit problem.
This is easy enough - the start epoch time is set explicitly in the simulations to be 31st of July at 10am. Applying this offset to epoch times will give the access time in seconds.
To do this, I will need to restart the simulations and re-implement the range function to handle the case when the ranges are also plotted in time.
I've implemented these fixes and the reports are being generated with much less data. This should hopefully speed thing up too.
Because the melds I ran comparing like for like hand made and programmatic sims matched up, I am going to close this issue.
I do not need to change the script for reading in the ranges as it should as based on Darian's implementation, it should be able to handle both cases.
This issue covers the validation of the generic scripts implemented in #25, by comparing results with hand generated simulations.