Hi Laurent,
I've been looking at it a little lately.
One thing that occurred to me is that the 2 strategies - RLE vs
uncompressed - might be easier to follow and manage if they were broken
out into separate classes:
MarlinCache
+--- MarlinRLECache
+--- MarlinUncompressedCache
...jim
On 10/9/15 3:20 PM, Laurent Bourgès wrote:
Jim,
FYI I worked on improving Marlin patch #4 to send you asap a new webrev:
it will be simpler with only 2 variants (uncompressed and RLE+blockFlags).
I also fixed a bug in the array cache related to crossings (ptrLen !=
ptrEnd) in Renderer.endRendering () and wrote a test that makes all
cached arrays to resize above their initial capacities.
It is a bit faster on my ellipse tests: 285ms and 420ms instead of 300
and 450.
I will evaluate using xcoord instead of runLen in the tuples and see if it
saves some add operations.
It seems promising so I will try implementing it but probably after
sending the new webrev.
Did you look at the tile flags approach ?
Any other advice on the heuristics I experimented ?
Will you have some time next week to review that patch ?
PS: What is the plan to perform the concrete Marlin integration into jdk9 ?
Laurent
