Hi @theakson, As a regular on DSLR I saw this speed test evolve and actually thought many things are a lot better now.
You will notice the DSLR test uniquely uses several locations simultaneously. This is quite a good idea in theory but in practice, it's been rocky. It's all written in HTML5 and for quite a while, the waveform never aggregated all these sites correctly in Linux Firefox.
I use a conky display showing what the waveform should look like but it never did.
System Monitor network plot can do the same thing. Compare and see what you get - does it match?
In the meantime, my OpenWrt router's SQM (Smart Queue Management) has gotten my buffer-bloat under control with a very nice waveform and the DSLR test also seemed to be correct. But from your report, maybe not. Or maybe buffer bloat affects this - I'm speculating.
Also, I attempt to run HTTPS Everywhere. For a long time, this test required it be turned off. But no more I'm glad to see.
Just my 2 cents.
EDIT: Had to try it - here's my results. I'm showing actual as a very "square" bandwidth but the DSLR test shows a much more slow rise on both waveforms that didn't actually exist.
BTW, you can post your results here, like this:
And yeah, I'm only 30/5 but on the LAN get well over 900mbps.