As I work more and more with PfSense and research more into firewalling with Linux, I can't help but to notice that there are so many options available for QoS. I've learned so far that a significant portion of my connection 15%-30% [depending on my queue's] is wasted and/or not available to my machine when during that moment, I'm the only one using the connection. I have six machines on my network, four of those are F@H machines that rarely use the internet for anything. My main desktop and my roommate's laptop are the two top bw consuming machines. She rarely get's on her laptop [netbook], and when she doesn't it's simply browsing.I need to find a solution that will maintain priority on certain ports but not block all of the availability of the connection for the other hosts. It seems at the moment that pfsense isn't sharing the available bandwidth between queues. You first have to setup what % of the total bw you would like to designate to each queue, [a port or group of ports]; then if there is any available surplus/unused bandwidth it gets shared out to other queues, until it is needed. Ideally this would never waste anybandwidth if it was running efficiently only serving a few machines at a time, however, I'm seeing serious losses in download and upload speed, as much as 40KB/s-100KB/s, it's a bit unnerving.Overall I need to either find a better policy setup for the queues, or find a better more efficient solution altogether Linux. I love pf though, maybe the end result should be testing out a dedicated OpenBSD with similar rules. This kind of takes away all of the nice efficiently accessable graphs, and many other things, but I'm more concerned with my connections efficiency. Sorry about the babbling on, just wanted you to get a feel for the entire setup.