Pompano Beach, FL
I have no first hand experience with the Stick modem..
But, I've used one t-mobile's ZTE hotspot(21Mbps--older tech) in a tethered configuration during a very long road trip(>3000 miles).
A couple observations.
One must get some separation distance between hotspot/modem stick and your laptop/pc. Laptop's/PC's can be very noisy(RF wise), especially in the bands T-mobile uses. The resulting interference can significantly impact performance.
Get some separation between your PC and RF modem, at least 4 or 5 ft.. 9 ft is better. Get a $10 USB extension cable and connect your stick USB cable to the USB extension cable, and it to the PC/Laptop.
Modern buildings have a lot of metal in them(outlets/appliances/framing/etc), relocate stick/hotspot higher up and/or closer to a window.
The more bars you get.. the better the performance..
Following some of these rules vastly improves performance, in major metro areas I would get up to 14/3Mbps in speed test results.
>One must get some separation distance between hotspot/modem stick and your laptop/pc. Laptop's/PC's can be very noisy(RF wise), especially in the bands T-mobile uses. The resulting interference can significantly impact performance.
That's not really true. My spectrum analyzer doesn't show any interference signal coming from laptops. There are FCC rules that limit that kind of interference to non-interfering levels.
I have a Zoom 3G USB dongle that works fine plugged into the laptop IF the laptop is in the right location. The problem is you can't get good reception most places where the laptop might be setting indoors so the USB device works better if you put it on a USB extension cable and sit it in a window. It also helps to use an outside roof mounted antenna for a strong signal.
Pompano Beach, FL
Yes there are limits imposed by the FCC.. But practical considerations means they can't test for these limits under all conditions.. And those conditions are going to vary from laptop to laptop, PC to PC, different CPU types, CPU clock frequency , memory types, battery state, power saver program, memory bus speed, rise and fall times on memory bus drivers, etc..
Heck the original author doesn't mention what type of PC's they have, it could even be something as simple as a missing chassis screw on a desktop PC.
P.S. I'm a expert in this arena..
Even the FCC part B radiation limits (300 uV/meter) can severely impact local cell receiver performance, since they are often listening for extremely weak radio signals.. -80 dBm, to -110 dBm. The cell tower they were using could be out of action, and they now depend on a less than optimal tower.
Even a shielded local transmission source(@ primary frequency) can easily swamp those distant signals. The only way to reduce interference is to add more shielding and/or more distance.
Thus my original suggestion still applies. (add more separation/distance, improve received signal strength).