the port-forwarded isnt a fast one though, limited to one core at about 700khash/s.

I suspect there's some difficulty receiving blocks if all the nodes you're connected to are 0.3.9 or lower.  We need enough of us so that at least one node you connect to will be 0.3.10.  The problem will start to go away when we make up more than 1/8th of the network.

It'll help if you port forward so you can get lots of connections.

I'm working on upgrading my CentOS machines (have to custom compile for it), once I get them updated, they should be able to connect to as many clients as they want/can then.

Here's what I have:

All of these have the updated blocks, all of them had the block chain wiped, reloaded old one, updated all versions to 0.3.10; I hope that is the right number so far?

IP Address:
12.53.130.10 - (Linux CentOS 5.5)
99.59.167.147 - (Linux Mandriva 2010.0)
64.186.154.76 - (Windows Server 2003 x64 Enterprise)
64.186.155.104 - (Windows Server 2008 x64 Call Center Edition)

As you can see, I like variety in OS I guess.  

I've opened them all up to as many connections as they can take (with the undocumented connection option now), so I can kill two birds with one stone. Help restore the network and test part of the netcode