I have been running tomato for quite some time on my buffalo WHR-HP-G54, version 1.09 had over a year worth of uptime when I finally upgraded and restarted it. The upgrade was due to the fact that it would sometimes forget to renew the DHCP lease automatically after I upgraded to Verizon FiOS. Version 1.23 has been up and running for about a month, the problem looks to be fixed Anyway, there does not actually appear to be any problem with the router, it works fine. My current service is Verizon FiOS 20/20, the router is connected directly to the ONT with a cat 5 cable, the coaxial connection is not used, nor is the original ActionTec router supplied by Verizon. Back when I was with Comcast, where the connection was only 8/2, the bandwidth graphs were always fairly smooth. With the FiOS connection, they are usually smooth, but when I start moving a lot of data the graphs sometimes gets these huge spikes were choose way up, goes way down, then goes back to the line it was on before. It typically doesn't do this if the combined throughput is under 10 Mb/sec, but almost always does when it gets close to 20, especially if there is a lot of traffic in both directions. Is this an indication the router is close to some kind of internal limit? As I said, there does not appear to be any problem with the connection, it has not had a moment of downtime since I upgraded to 1.23 a month ago. I am just curious about if this means anything. Thank you Ed Added: the spikes do not show up on a bandwidth graph in eMule, which is usually one of the application using most of the bandwidth. It also does not show up on the bandwidth graph in Windows XP, so I don't believe it actually is showing a huge swing in throughput, I think it is just a display thing.