The peer list will make all the difference. On private trackers your likely to get a lot of high bandwidth peers to feed you download, but peers taking from you are scarce. Also, because both clients are operating on the same IP address, they cannot connect to the same peer, so whoever gets the fast download peers first wins the download race, and whoever gets the most peers asking for your upload will win that race.
The only way to test them so that all is equal is to run the test on a lan where you control all peers.
To get the best performance on either client, you need to have your max upload speed capped at about 80% of your tested upload speed, so there is enough bandwidth for prompt replies to peers wanting to trade with you. If your running two clients at the same time, that makes it much more complicated.
If you want to run a more accurate test, try doing them one at a time, but still remember that the results will be subjective.
I can tell you from my own testing, as well as testing by our development team and independent experts (see the robb topolsky paper), both bitcomet and utorrent will perform equally well in a private tracker environment.
Some users may find that one client or the others default settings work better for their connection, but with all conditions being equal, they both work the same way.
BitComet does have some strong advantages on public tracker, but only if resources outside the bittorrent swarm are present. This is where the emule and LTseed are handy.
Regarding LTseed, without registering for CometID, you can connect to upto 40 LTseed peers, and if you register for CometID, the more you upload and run torrents, the more connections you can receive. This works much like bittorrent protocol, rewarding those who upload the most, but with CometID, your total upload for all torrents increases your rank, where with bittorrent protocol you could be the best uploader in the world for ten years, but when you start a new torrent, your no better off then any other peer.
I have seen cases where users have only been about to get a max of about 100kB/s from bittorrent peers, and shortly after enabling LTseed the speed jumped upto over 1000kB/s. This isn’t always the case, but it’s not an uncommon experience. Naturally you’d have to be on a public torrent with other bitcomet peers to see this benefit.
The really beautiful thing about LTseed protocol is it doesn’t just benefit the bitcomet user, it can benefit the entire swarm, because all data downloaded from them is then uploaded to the bittorrent swarm. This can breath new life into dead torrents, and LTseed will only upload when there are no bittorrent peers available to upload to.