Story board for this paper:
(1) NuSTAR observed LMC X-1 three times. Once on its own (40 ks), once with XMM (60 ks), and once again on its own (40 ks).
(2) NuSTAR sees lots of variability in the source, with the diskbb-related component apparently varying a lot during an observation, while the >15 keV emission varies from epoch-to-epoch.
(3) During the XMM / NuSTAR observation we saw various "flares" here meaning flux increasing by factors of 2x in both XMM and NuSTAR. For one of these we see some evident of spectral evolution through the flare.
(4) In all three epochs, the epoch-integrated spectrum shows that the source is in a different state compared to what Jack Steiner saw when they measured the spin using Suzaku. We see a softer power-law component in all three states (more like 3) compared to the 2.5 that Jack saw.
Note: In general we haven't had great success in modeling the Fe-line with a reflection model because the statistics are too poor. This being said, there seem to be a lot more models out there (if not public, at least accessible if you know the right people...)
(5) From these data, we can't confirm the spin measured by Suzaku. The RXTE analysis explicitly assumes that the source always has a power-law index of 2.5 in order to get the spin measurements. Given that we've looked at it three times and have seen a softer non-thermal component I need to make a probabilistic statement against the RXTE analysis (Gao et al, I'll link the paper in here).
(6) Let's compare the source explicitly to Cyg X-1. These are both wind fed systems, so why is their behavior so different? What's special about LMC X-1 (or Cyg X-1, for that matter) that allows one source to undergo state transitions and the other not to. Something to do with the angular momentum of the stellar wind for the LMC X-1 companion truncating the disk?
(7) How can we interpret the spectral evolution of the flare? It looks like the harder emission goes first, then the softer emission responds. This implies that there is some energy transfer in the disk, but the timescales are pretty small. Can we use this to measure the viscous timescale of the disk? How else can we interpret this?