Spoofing OS userAgent via CUrl; am I approaching this right?

So my school seems to be shaping traffic based on the OS of a device. For instance, I access a course video on my desktop (windows 8.1) and it works just fine (downloading the video at about 400k/sec) however when I attempt to do the exact same thing on my laptop (mac OSX), I get a choppy video downloading at about 30k/sec.

I assume that their is OS based traffic shaping going on (OSX and IOS seem to be too similar for whatever network shaper they are using). I want to prove my assumption....

I figure that if I can use curl to access the video at http://ift.tt/1AUTwrj from within my mac at a speed approaching 400k/sec then I will have effectively proven that my school is shaping traffic based on the OS. My problem is that when I run the following command, I see no different results...

curl -A "Mozilla/5.0 (Windows NT 5.1; rv:31.0) Gecko/20100101 Firefox/31.0" http://ift.tt/1AUTwrj

My question then becomes (1) am I using this command and/or method to prove my assumption correctly and then (2) is my assumption wrong? For instance, are they using MAC address filtering instead....

---

NOTE:

• I have already tried all the main browsers and confirmed this behaviour in both OS's (IE all 4 browsers in Windows DL the vid at ~400k/s while all 4 browsers in OSX DL at ~30k/s)


• I have tried using a VPN on both OS's, no effect.


• My IT department denies anything is wrong (even with screenshots) and they also suggest using firefox as the preferred browser (which is why my curl is using mozilla).