Hi, Jason, Why only a factor of 2 faster instead of a factor of 10? What is the limiting factor other than ethernet speed?
10 Mbps throughput over a 100 Mbps link (10% efficiency) seems not so great but understandable/acceptable. 20 Mbps throughput over a 1,000 Mbps link (2% efficiency) seems kind of bad. Thanks, Dave On Apr 28, 2013, at 11:21 PM, Jason Manley wrote: > ROACH2's PHY is different. It's now Marvel (as opposed to a National Semi on > ROACH1). It works reliably at 1Gbps. > > I bank on 10Mb/s on ROACH1, which is achievable using the normal katcp and > tcpborphserver2 without any trickery. You can get better if you write your > own PPC code. > > ROACH2 should be about a factor of 2 faster, but I haven't benchmarked it. > > Jason > > On 29 Apr 2013, at 00:08, John Ford wrote: > >>> hi john, >>> >>> i think all roach1's PPC ethernet ports need to be used at 100Mb/sec port >>> speed, >>> independent of when the boards were made, and not at 1Gbit/sec. >>> >> >> Thanks, Dan. I thought that a PHY chip had been changed, but I couldn't >> remember. Now I seem to recall that the Phy is just Not Good. >> >>> it's tricky to write PPC software that gets more than 20Mbit/sec >>> out of the port, but i'm not sure about this 20Mbit/sec number. >> >> OK. I wonder if anyone else has experience with this? I guess we'll test >> it if not... >> >> John >> >>> >>> best wishes, >>> >>> dan >>> >>> >>> On Sun, Apr 28, 2013 at 1:57 PM, John Ford <[email protected]> wrote: >>> >>>> Hi all. >>>> >>>> Can someone tell me if the ethernet port on the newer ROACH boards is >>>> reliable at 1 gb/s? >>>> >>>> Also, what is the maximum data rate expected through that interface? >>>> >>>> Thanks! >>>> >>>> John >>>> >>>> >>>> >>>> >>> >> >> >> > >
