After getting complete data for a 10 gigamessage interbox test,
( proton-c, event interface, code here:
...I see that there is no gradual speed change at all over the duration of the
If you zoom in enough on the numbers, you can see an interesting little sine
but the magnitude of that wave is very small, (about half a percent) as you can
see from a picture of
the same data, but with an absolute scale:
It looks to me like proton-c is stable enough for any practical purpose.