Lets not sugar coat this and get to the point:
I've had horrible experience using MPD made LMR400 cables. First off, these cables were a PIM galore, riddled with high SWR and noise, and windy day? forget to hear anything but noise due to the cables swinging in the wind.. In addition, the crimping always failed too. Connectors are chrome plated cheap garbage, even though they claim silver, they are NOT silver. The crimping is not weatherproof either, they are not using anything to help weatherproofing at all, its just a regular POS shrink tubing... nada. So, when sitting outside the outer jacket will rust and the already crappy LMR400 PIM-galore performance will turn into utter trash with sky high SWRs.
For a 100 feet run, please, save yourself all the headaches and buy Heliax, 1/2" should be a good start; seriously, don't bother with LMR garbage. I can hear it... "here comes another heliax snob..." yes yes, I've heard that before, but let say I owned a ton of LMR400 too, I always thought it was the best thing since sliced bread...., I thought it was great... ha... ha... ha.... until I tried Heliax and chrome-less connectors... never ever ever in a million years will use LMR for anything longer than a 3-4 feet patch, and silver plated connectors or tri-metal ONLY.
In addition to that, I will reiterate what everyone here has already stated: use N connectors, preferably Commscope tri-metal ones for the long Heliax runs; the cheapo chrome plated ones suck at impedance as I've found out. If you want performance, be prepared to spend $$$, unless you live atop Mount Everest...
Yes, its hard to swap the UHF thingy, but I've recently done it, and it was well worth it, and once you have the setup running you'll love it: no more sealing connectors, ever, no more creeping SWR after a couple of thunderstorms... no more hearing the NOAA due to PIM garbage LMR400 swaying in the wind... etc..
EDIT: Oh, don't forget to ground PROPERLY the antenna/mast/radio to the same ground rod. Massive improvement when I did this in noise floor threshold reduction...