Of course these solutions (except the mentioned isolator, see DaveBSC post) will do NOTHING against common mode garbage loops.
Not even when battery powered hubs are used and the +5V is interrupted.
Why ?
Because the common mode garbage travels equally well over the ground line as well as the +5V line and datalines.
The ground line (and datalines in a lesser degree) will still conduct the RFI garbage ground loop regardless of how clean it is powered UNLESS there is true isolation.

One should realise that isolators WILL break a physical electrical ground-loop but only attenuate RFI garbage marginally and would suggest to use clamp ferrites WITH these isolators (that should be battery fed or have a lin power supply and NOT cheap lightweight SMPS).
Don't do a single pass though a ferrite but get the clamping types (the ones you can open and close) with holes just big enough to fit 3 windings of the USB cable through it and do as much windings through it as possible.
a single pass winding is about 3 to 20x  LESS effective as when 3 to 5 windings are used.
More than 7 windings won't give better attenuation.
Make sure the clamp closes well and there is no widened air-gap between the two halves of the clamp.

Above is advice given based on actual performed RFI measurements (in the past on USB outputs) and will actually do something against common mode RFI which is the culprit in 99% of cases where troubles are found (mostly strange noises heard in the audio path) in conjunction with such isolators.

Be warned that it is often those cheap lightweight smaller SMPS that have bad or insufficient filtering (you know the ones that come free with every cheap device) that cause the MOST RFI garbage !

Ferrites only 'help' if you are experiencing weird noises in the audio that is caused by RFI common mode garbage.
It is not easy to determine if noises are caused by audiofrequent groundloops or RFI without testing or measuring.
Fertrites can only attenuate RFI a certain amount of dBs which could, if the attenuation is enough to become too low to cause problems (below a certain treshold it has no influence anymore, that treshold differs).
Groundloops where garbage is in the audible range (or just above it and 'mirrors' into audio by interference) will NOT be addressed by ferrites at all... NO influence in the audible range but only above 100kHz.
These loops can only be broken by isolators as suggested by DaveBSC.

For a total package you need a real isolator between the PC (or source with high amounts of garbage that causes audible problems) and the DAC.
Such an isolator would still conduct RFI common mode garbage which potentially COULD cause problems as well under certain circumstances.
To lower that the ferrites will help.

So Isolator + ferrite between PC and isolator (preferably battery or lin supply fed) + (optional) a ferrite between isolator and DAC.

Ferrites alone on USB will likely do very little to nothing IF one is plagued with audible noises in the audio frequency groundloop.

The audio frequency groundloop could also be 'broken' by powering the DAC via an isolation transformer.
These transformers do nothing against RFI ground loops.

When you are not experiencing any weird noises I see no reason to apply all these costly measures myself.

So groundloops exist in 2 forms, DC to audible frequent loops and RFI loops.
If only 1 bothers you address that one, to be safe address both.

Alas, I cannot comment on subjective improvements when these things are applied because this is not my area of expertise. 
Note: the mentioned isolators do not go beyond 24/96 as mentioned by DaveBSC.
An option to get rid of a LF loop, if that is plaguing you (but is far from ideal) is to feed the DAC+ amp from an isolation transformer instead.
There will still be RFI garbage and audiogarbage through the USB but the impedance (and thus current) of that audible frequency groundloop is higher so induced currents are smaller.
It is these induced currents that are audible.

example of such a clamp:
http://nl.farnell.com/tdk/zcat2235-1030a-bk/ferrite-cable-clamp-filter-100/dp/1844071
Note: this is NOT the one you will need and have to find one with the proper inner diameter depending on the used USB cable and number of windings (between 3 and 7)

example of multiple windings through a ferrite:

yeah. I kind of feel a bit bad for my short replies, heh.
hmm. well as I was saying I just wanted a good USB hub setup so that I could use my ODAC and external HDD and charging my iPad... while compromising sound as little as possible. I haven't actually tried it with the laptop yet, so I might have good sound though if I add a hub that may change.
but yeah as long as I am happy I should perhaps not look into changes... will report back when I have the laptop.

so summary is, a good usb hub psu is expensive, smps powered one might even sound worse than just using laptops usb? and there are no noticably better choices around $50?

thanks a lot for the information, hope it will not be needed though!

yeah I figured. I haven't bought a hub yet but safe to assume they all come with smps right. battery is completely dead so it runs from the outlet.
I am willing to look into cheap DIY psu if possible, but I think since nothing was mentioned earlier that is a no-go
so isolator and battery power for cheapest solution in terms of sound... yeah. batteries hmm.
thanks again!