2020-06-19

I managed to get a good start at the last-use analysis yesterday. It is pretty simplistic, but it seems to work alright. I want to expand it with aliasing support today. I haven't taken a hard look at Cosmins code yet, and I don't really know why it's so much more involved than mine. It would be nice to investigate a bit before our meeting at 15:15.

Robert says there's a meeting in Fritz' group that might be interesting for me today; Troels is doing a talk. It's at 11:00, so if I can get some good progress on the aliasing stuff before then, I'll try to go.

I didn't get that much done today, as most of the time was taken up by meetings. I did apply aliasing though.

According to Troels, bfast is not being tuned correctly with the new autotuner. Or it could be that moderate flattening is better than incremental?

I'm going on vacation! I'll be back with more updates on <2020-07-01 Wed>. Hopefully I'll be able to write a retrospective on this process before then.