From all I've read, directly buying Bitcoin seems to have a higher likelyhood of being profitable than the best GH/s renting offers I came across, this one included.

If the exchange rate goes up, my investment directly goes up, whatever the difficulty of the moment is. If it drops, I lose a part of my investment but if I were renting some GH/s with Minershare.net, the mining would likely stop altogether if it's not profitable anymore while I still could hold my Bitcoin "coins" as long as I wish for it to recover without worrying about the difficulty increase which would make mining again with my GH/s at Minershare.net later on even less profitable compared to holding my Bitcoin money.

There is enough information at http://www.vnbitcoin.org/bitcoincalculator.php to figure that out.

I'm considering they are honest but, indeed, that's a potential too...

In fact, I don't blame your business model so much, I find it smart from a certain point of view and I wish to succeed in the first place so that everybody "wins".

If the exchange rate with fiat currencies stays the same for the next months, the fact is that 30 GH/s at $120 with mining starting today is an interesting opportunity and you may even break even by mid-February but 30 Gh/s at $120 with actual mining starting in two months is a whole different story. Difficulty might be around 5,000,000 at that time and profitability, half what it is today.

For me, the biggest issue with your offer is that there isn't any fixed length of contract.
I could still manage halved profitability if I know I'll be mining for a year, for example, but when I don't know how long it will last?

I understand it's all about risk/reward and that it's everybody's responsibility to carefully read and understand for themselves what your offer is about but I find this kind of thread useful because some people may just jump on your offer not informed about all that it entails.

Thank you for your answers anyway.

My opinion:

From KnCMiner's website for the Neptune miner: "Shipment begins in Q1/Q2 of 2014".
That's not very precise and mining difficulty is increasing all the time (every 10 days or so).

https://www.kncminer.com/news/news-69

So, I would say, you are selling shares of the miners because you don't know yourself when exactly you'll receive them and how much you'll be profitable mining with them or even profitable at all given the mining difficulty increase we have seen in the last months. You're even smart enough to say that you'll stop mining when you costs aren't covered anymore so that you won't lose money. So, with this, you don't risk anything while customers take all the risks upon themselves involving the risk of not having any ROI at all.

Am I mistaken about it?

Looking at the Bitcoin mining difficulty still increasing fast, I'm not sure this device is worth buying from a profitability point of view though, except maybe if one can get it cheap and can overclock it as much as possible by soldering the appropriate resistors.

I bought a USB Bitmain AntMiner U1 a few days ago to mine Bitcoin and, today, I tested it both with cgminer as explained in the instructions from the company and with the latest version of bfgminer (3.10.0) at the stock frequency of 1.6 GH/s.

In the first case, I got up to 5% of HW errors but I don't see them in bfgminer.
The running frequency looks also more stable with bfgminer.

I uninstalled the driver before using bfgminer because cgminer requires the use of Zadig.
I use Windows (8.1) and if it doesn't work, you could also use "USBDeview" to uninstall the driver completely (http://www.nirsoft.net/utils/usb_devices_view.html). I used USBDeview myself but I suppose my problem was more the Comodo firewall which requires a restart of Windows before really allowing the miner through after adding it in the allowed applications.

I would have to test and compare the number of rejected shares in both cases but it looks like bfgminer works better with the AntMiner U1, maybe due to the fact that it can use the driver as is without modifying it with Zadig.