> If this were 70.5GB instead of 705MB and the data exhibited identical performance, the final archive would be 4.1GB smaller<p>I don't think the savings for a file 10x as large can be extrapolated. This likely wouldn't work as the (default) window size for XZ would be too small; the similarities would be too sparse. To get the same benefits you'd need to use at least 10x as much memory by increasing the window size.