However, put a bigger device on a car or on a boat and that can be useful.
I guess you can use it to top up then. e.g. Bring 1 litre instead of the 2 & rely on this to cover the shortfall.
>It's not worth it.
Not yet. Cutting edge near prototype stuff rarely is "worth it".
Where my curiosity lies is in energy consumption. There's a power button on it obviously, so how much power does it use? I'm not an expert in thermodynamics/rates of condensation but I know my dehumidifier is a freakin power hog and it's doing exactly the same thing as this. My guess is that power consumption is the biggest hurdle here.
 http://www.water-gen.com/product-water-generator.html and
im curious if this device on your bike would cause you to burn more water than what it produces
Guessing the diameter of a drop at 1mm, and assuming it to be cubic, that is 60 cubic mm per hour. Rounding up, we get 1 cc per hour. A U.S. teaspoon is just under 5 cc. Alternatively, get a small cup of water every 24 hours.
There can be about 30 cc of water in a cubic meter of air, so you would have to pump 16 cubic meter of air through this thing and extract all the water to get at 500cc per hour. With a 10 by 10 cm opening, that is a 1600m long column. A bicycle can easily do ten times that speed, so that sounds doable, but that is at maximum efficiency and the article doesn't say he has done that, so I fear that 500cc is wishful thinking.
"Under the right climatic conditions, Fontus can easily produce half a liter of water in an hour's time."
We already know that it can produce half a liter per hour, but those same climatic conditions are when a cyclist needs significantly more than half a liter per hour.
I still doubt that number. Looking at http://atmosphericwatersolutions.com/wp-content/uploads/2012..., I see a machine that weighs 38 kg, consumes 373W of power, and produces 'up to 17 liters of water a day'. This thing would be an order of magnitude smaller, and produce almost the same quantity of water. Possible? Maybe. Likely? No. Because of that, I would like to see experimental confirmation of that half a liter per hour before accepting it as a given.
Also, see http://en.m.wikipedia.org/wiki/Atmospheric_water_generator which claims the U.S. army and FEMA use a system that needs a gallon of fuel for 'up to' five gallons of water. If something like this were known to be more efficient, I think it is likely they would use it.
I'm guessing this is limited by how much cooling you can get out of it.
The cooling system has to be cooling the air that passes through it down to the dew point, if the ambient temperature and the dew point are too far apart, it's going to consume too much power to be practical.
Linking a small solar cell (I would say ~1W from the shown surface) to a Peltier element (known power hog, a really small one takes at least 4W) is not a solution for efficiency.
Wikipedia, and a post by 'someone' tells us there is ~30cc of water per cubic meter of air at 100% humidity and 30c. (This could be a reasonable set of envoronmental conditions in Washington DC for example).
The mass of water in 100% humidity air is
28c 27.22g http://www.wolframalpha.com/input/?i=mass+of+water+in+1+cubi...
30c 30.34g http://www.wolframalpha.com/input/?i=mass+of+water+in+1+cubi...
So we can extract ~3g, or 3cc of water from saturated air, with the use of ~2.6kj (assuming 100% efficiency, which is unlikely)
Some of the highest power density batteries I've handled are 5 cell, 5500mAh li-poly batteries for model aircraft use, with a total energy capacity of around 180kj,
Or ~ 200g of water.
I think I'd rather carry a water bottle.