Someone must have already figured this out. I'd like to know at what electricity price Gridcoin breaks even, assuming roughly average returns. I know this is a function of GRC's price, say, in dollars, and the number of users. I was idly speculating about the possibility of covering electricity costs in places with cheap electricity like Iceland. Theoretically, distributed computing projects could earn useful foreign exchange.
Of course, this could just be another one of my bad ideas.
The problem here is this question depends on a lot of factors. What do you define as "roughly average returns"? Obviously a device like a Raspberry Pi is going to use a lot less energy than a full server or multi-GPU setup. What is your definition of"average" here?
Good question. I was thinking in terms of average RAC per box, loosely speaking. That is, something like mean, median, etc. I was thinking in terms of personal computers that could be found at a home or school.
The point is to determine the circumstances in which running Gridcoin is profitable for common types of computers that one would already have on hand. This may very well be a pointless exercise, but a negative result is still a result.
You just planted 0.05 tree(s)!
Thanks to @rufusfirefly
We have planted already 3599.56 trees
out of 1,000,000
Let's save and restore Abongphen Highland Forest
in Cameroonian village Kedjom-Keku!
Plant trees with @treeplanter and get paid for it!
My Steem Power = 39372.51
Thanks a lot!
@martin.mikes coordinator of @kedjom-keku
You might find this post from @ilikechocolate interesting.
It looks at the idea of normalizing BOINC credit based on estimated power usage per operation (FLOP), there are some calculations for common processors and graphics adapters.
You could probably use the figures together with statistics form a State of the Network post to compare cost per 1K Recent Average Credit with the Gridcoin reward value per 1K Recent Average Credit.
You would have to do the comparison for each project separately and re-calculate as new statistics become available.
Thanks. I do like the idea of tying GRC rewards to energy expended. It's clear the implementation would be difficult, if not impossible.
Here's mine. My 2 rigs consume 650 watts while crunching.
Rig 1: Ryzen 7 1700 3.75ghz 16gb 3200mhz ram rx 480
Rig 2: I7 920 @3.8x ghz 12gb ram rx 480
I make around 32 GRC/day and my electricity cost is between between $5.91CAD and $0.091 CAD so let's just assume $0.07 USD per kilowatt-hour.
At 15.6 kwh just for crunching, it costs $1.092/day in electricity or in other words 49.64 GRC a day to break even on the high end. On the low end, it would require me 32GRC/day at current rate. In effect I'm always on the low end because my consumption rarely goes over 36 kwh/day. All my lights are LEDs and I dont use my large oven anymore. Bonus: Crunching is my heating in the winter.
PS: I could probably decrease it a bit by undervolting.
Thanks. The heating is an obvious benefit. About how much heat (in USD/day) does your rig generate?
Obviously, the returns and costs are low. Assuming USD 0.02/GRC and assuming away heating benefits, the breakeven electricity price is about $0.04/kWh. That's the number I'm looking for.
Now assume a school running Gridcoin for a net profit of $0.10/day (very optimistic). 1000 boxes would generate $100/day or $52,000/year, a not insignificant sum. Of course, electricity would have to be very cheap to begin with, an extremely unrealistic assumption. I looked at Iceland's electricity prices and read that prices are rising as demand grows to levels above some US states.
So, my idea is pointless. Run BOINC for the science and get GRC as a bonus. That's what we've been doing all along. Maybe in the future, GRC will rise enough to make BOINC profitable, but don't make any decisions based on that.
You might as well consider it infinite for me. I dont have to heat at all thanks to crunching. My appartment is about 30 by 24. ( I would have to pull my measuring tape to verify)
It's not really infinite. The return to the heat can be quantified as the cost of other heating avoided. I assume you live in a cold climate. How do control the temperature besides opening the windows when you feel too warm?
I assume @ragnarokdel gave flat dimensions in feet, so it is around 70 sqm. His rigs use ~6000 kWh per year, what gives ~ 80 kWh per square meter per year. (If my calcs are ok).
For many (in non extreme weather areas) flats / houses you can assume heating demand is at 100 to 200 kWh per sqm.
But gas heating is often ~3 times cheaper than electric heating.
In extreme case when source of heating is grid electricity, roughly you get a net profit equal to mined GRC switching from standards electric radiators / heaters to CPUs / GPUs.
3 generations old pc might use the same energy as a new one and produce 5x less GRCs, another variable to account for (just loose numbers).
We have extreme climates, our houses are really well isolated and the snow also acts as an isolant. Since I live in a basement appartment, most of my surface is either underground or covered with snow in the winter.
Of course it's not actually infinite. but I dont use heating except maybe one or two days a year at those super peak lows (-35°c let's say). I wish it was separate on the bill so I could give you a number but it isn't.
In Québec most building even with water heating use electricity, older houses might be different as for my apartment, it's in a house basement but it's definitely on electricity as well.
As for summer, I dont mine in heatwaves and yes I crack a window open. Being in a basement keeps it relatively cool.
Thank you for spelling this out!
Congratulations,
you just received a 16.27% upvote from @steemhq - Community Bot!
Wanna join and receive free upvotes yourself?
Vote for
steemhq.witnesson Steemit or directly on SteemConnect and join the Community Witness.This service was brought to you by SteemHQ.com