What I was referring to would be the ability to control the modes ourselves. You indicated in the next quote that the machine essentially self tunes itself based on the room temperature. I am a person who runs a bank of 741's in the winter to heat my home, I like being able to overclock them in the coldest months and in fall/spring play with the settings to adjust the heating to my needs. So my suggestion is to save money and put out an inefficient miner, making it an efficient heater. The product pitch should be "The heater that pays", as that's what's being produced.

My reasoning here is that it should be more cost effective to use older/obsolete chips when compared to current gen gear. This can help get the cost of the product down to the point where it makes sense to someone to pick a few of these up. I also wouldn't launch it with multiple colors available for your first run, it has to be cheaper to buy all 1 color as opposed to being able to offer 3. If the product does catch a market share and gain adoption for heating future models could have a color option, but for now I'd say the obstacle is getting the cost down for people.

Crowdfunding is a good idea to get the capital needed for a launch. I don't follow it much but it's generally seen as more of a donation to inventors to help them get a prototype together or finance the first run of a product. Usually there are some "gifts" involved, or the opportunity to buy the product at discounted rate, or just the opportunity to buy first.

Best of luck to you with the project. Like I said before I like the project, I was excited when I saw Canaan putting a similar product together. They ultimately decided it wasn't viable outside China. Hopefully you can find the right group to buy these.