Monetizing the Megawatt

If you get fifty Bitcoin miners and you get them sat down in a room and you show them these metrics, they will all look at you like you are an insane person, because what they understand is that there is no cost of producing a Bitcoin, but there are costs of producing a Bitcoin, and it's totally contingent on the individual who is actually producing the Bitcoin.

Hi everyone, and welcome back to Stephan Livera podcast. Today we're gonna be talking about Bitcoin mining and, energy and monetizing the megawatt. Joining me today is Robert Warren. Robert is the Head of research over at Bitcoin Park. I've known Robert kind of around, you know, the Bitcoin scene for a little while, been meaning to get him on, just haven't had the chance to until recently, until just now, and I know, you and the guys just recently hosted the Nashville Energy and Mining Summit twenty twenty-six. So, yeah, first off, welcome to the show, and maybe just give us a kind of a, a quick overview on what happened at the Energy and Mining Summit.

Yeah, huge pleasure to be here. So we, yeah, Bitcoin mining summit. This is now going on, almost the, the fifth anniversary of Bitcoin Park is coming up in June, so it's amazing because this has been kind of another sold-out show and it's only gotten, it's only gotten bigger and bigger every year, What we are talking about when we're talking about this energy and mining situation, you know, if you were looking back two, three years, which seems like, you know, ancient times in the world of Bitcoin mining 'cause things move so quickly, we were still figuring out a lot of the basic principles of just how to run these many machines, and how do you build the control systems, how do you manage the heat, how do you manage the connectivity? We were still having conversations around, you know, various pool payout methodologies, and in a very, very short period of time, that conversation- Has been almost totally upended, and it's been totally upended by this explosion in demand for particularly AI data center loads, these kind of AI, HPC loads, which has really changed the public market space. And then it has been radically changed from the explosion of folks that are, finding very, very creative business models that allow them to mine Bitcoin very profitably under very, very unique conditions. And that's everything from reusing heat To going in monetizing, you know, stranded generation assets in the middle of Africa. So the, the world has expanded really, really intensely, in, in the Bitcoin mining space. It has shifted in other ways. With that being said, it still feels like we're just at the beginning of the, the conversation. There's a lot of runway left to go in, in energy and mining.

Yeah. let's talk about, one aspect from your report where you touch on just energy as human flourishing, right? I think people are now starting to understand that. So can you just explain that from your perspective, why is energy human flourishing?

A hundred percent. So it's, it's really, it's the core assumption and what you're referring to, we, we produce, spotlight papers in conjunction with these summits that we facilitate. So the summits, they tend to, to sell out, you know, everybody wants to kind of be there in the room, but what's really important is that you're kind of sharing that message with people more broadly. So we produce these spotlight papers, this one was titled "Monetizing the Megawatt," and the core assumption that we make, and we think it's a function that you can make backed by hundreds of years of empirical data, which is that you don't have anywhere on Earth a high-performing, high-GDP, low-energy country. If you plot on, you know, an x-y chart and you have, you know, the GDP and the energy consumption of any, any country on Earth, you will always find one hundred percent of the time that places that are more consumptive, that are consuming more energy, are ness- Necessarily wealthier, they're more successful, and this is a bit of an inversion of the way that a lot of us, you know, I'm, I'm in my, my thirties, and a lot of us millennials, we grew up kinda during the era of, the peak energy efficiency, that we have to all switch to a hundred percent renewables, and coinciding with that was this idea that it was sort of a good to use less energy, that there was some sort of moral imperative to not use energy, and what the data says in actuality is that this is Actually the opposite of the way that we should be thinking about this. You know, renewables are great, we should be, we should be coming up with more and creative ways to produce as much energy as possible. We shouldn't be under any circumstances trying to get people to use less energy, because energy in and of itself is the basis for us being able to perform operations that were totally alien even fifty, sixty, seventy years ago. The amount of energy that we consume is directly related to our flourishing, whether it's the things, the, the im- implements that you have in your kitchen, your dishwasher, your refrigerator, whether it's the vehicle that is parked outside in your lawn, whether it's a Tesla or it's an internal combustion engine, gasoline or diesel, the ability to harness and use energy is directly correlated to the quality of your life as a human being in the twenty-first century. So we really, we wanna start with this initial inversion, which is get it out of your heads that we should be chasing less energy, and get it into your head that human flourishing only continues to expand. And if we continue to generate and consume more energy, it allows us to be far more creative as human beings and produce really, really, really substantial outcomes.

I'm totally with you, right? 'Cause, I mean, I think we're similar age, right? I'm thirty-eight, and, certainly I was frustrated by a lot of the woke politics and this kind of net-zero style of thinking, which has permeated a lot of the world, and now it seems it's, it's turning around, but for a while, even in Bitcoin land We were dealing with people who were trying to criticize Bitcoin mining for energy use, and then now all of a sudden with AI, it's funny that all of a sudden that's all out the window now. Now it's everyone's like, "Turn, turn it full around, blasters are on, you know, thrusters are full, we need to like drill, baby, drill, everything is like turned around." So, I'm curious what, what has that been

like from your perspective? You know, it's, it's funny because in the world of Bitcoin, very rarely do we find ourselves laying the groundwork for things that people eventually celebrate. And the initial conversation around Bitcoin mining, the, the goofballs got a really solid head start on us. They were publishing papers about how every transaction, you know, kills six baby whales and burns an Olympic-sized pool of water, and they were doing absolutely terrible science trying to convince people that we would eventually- consume all of the energy in the natural world, and what was spurned from that, that bad research, that very biased research were a group of individuals who Worked their absolute tails off to correct the narrative and to really through the, through the empirical data of Bitcoin mining itself, showing what it does when it's on grid, when it's off grid, whether it's reducing emissions in the oil field or whether it's stabilizing pricing by, living on a grid like ERCOT or living in, on any, ITO, ISO or RTO in the United States, its ability to stabilize pricing and essentially provide a buffer to, to grid pricing or, or, the actual production. function, the, the generator production, whether it's, a coal plant or a nuclear plant or a combined cycle, gas plant, the amount of work that has been produced rebutting those initial papers has been an order of magnitude larger than those papers, and so it has now become very, very hard, by looking at the actual world, what people are doing, and then by looking at the scientific world, the research that people are producing, to make these kind of silly arguments. So for like the first time in the- Bitcoin universe, people are actually using our data to go, "Oh, wait a second, this, this isn't actually something, this isn't something bad necessarily. " Now that being said, it is really important to distinguish that, you know, Bitcoin mining as a consumer of energy, is a very different kind of consumer than AI. You know, AI is really in the first inning in terms of, inning in terms of what they are building out. They have substantially deeper pockets than Bitcoin miners. They tend- To look for these, these much longer, much longer duration contracts, so they would come to a town and they would say, "We wanna sign a, a ten-year, a fifteen-year AI deal." Now, on the financing side of things, this is great for a lot of folks who are selling the infrastructure 'cause it's a lot of, i-i-it's a lot of sort of forward capital that they know is, is going to be coming to them, and these are folks with very deep pockets. In terms of the effect on the grid, that is something that we To really dial in. And the reason for that is because the Bitcoin network, you know what Bitcoin is willing to pay you for your hash rate. We have this great commodity called hash price. When it comes to the world of AI, HPC, these large data center loads, these folks are willing to pay substantially more than what we would be paying in the world of Bitcoin, and that starts to create some misincentives when it comes to consumers on the grid, because these, these large AI- Farms, these RGI clusters are willing to pay more than consumers. Now, that doesn't necessarily mean they're going to eat the world and, and, and turn off your grandma's, you know, heating in the middle of winter, but it means that there's a different way that these businesses operate that we have to be really cognizant of because the grid operator needs to make sure that they keep not just the lights on, but they keep the cost of the electricity that is going to the actual user, the people, you and me, in a reasonable range.

Yeah, fascinating. And, the other big shift, obviously to get your thoughts on, we're seeing a lot of these big, quote unquote, mega miners switch to becoming AI. Now, some of them are pivoting, you know, we saw this over the last few years, they would pivot, maybe some of them partway to AI, and some of them have just gone full AI and just like not even doing Bitcoin mining anymore. So, I believe IREN is in that camp, BitFarms is in that camp, and maybe some others out there. So, can you just

It's a great question. So the, the, the long and the short of it is that, everything is good for Bitcoin, this included, right? what it means in terms of the public markets, as in, what you have to remember is that the publicly traded Bitcoin miner isn't the same as you with your bitex or you heating your home or even a large private operator that has a hundred megawatts and is doing their own deals. A publicly traded company has to maximize the profit of its shareholders. And what has been happening behind the scenes is that you've seen really, really large investment from activist shareholders that have said, "Listen, you guys think you are Bitcoin mining companies, you aren't Bitcoin mining companies, you are infrastructure development companies. And the way that we know that is that you've been going around signing deals, scooping up gigawatts of available capacity, and you've done that successfully, congratulations. But now if you're going to actually steward shareholder capital and you're going to produce the highest ROI on it, you can't sit there and just ignore this new cap- customer base that has emerged. And so the transition is from a financial markets perspective totally rational and in fact the best choice for folks to make. What becomes interesting is that, okay, you've got this high dollar opportunity where you can make a ton of money signing these long-term AI deals. That means in the short term that folks are going to be, you know, ramping down a lot of their, their Bitcoin mining. So in the secondary market, this is great for folks that want to scoop up a bunch of, near recent generation machines on the secondary market because it's going to provide a lot of that, and that inevitably, provides a lot more opportunity for tinkering, a lot more opportunity for creativity. Same thing when everybody was upgrading to S-nineteens and all the S-nines were going off in pallets and a lot of people got started learning Tinkering in their homes. So from a, from a pleb perspective, from a pioneer perspective, the people that wanna learn, this is net positive because it means that as they're slowly emptying out or transitioning portions of their businesses into AI HPC, most of those machines, if they're over, you know, eighteen months, which is a common sort of depreciation schedule, most of those machines are gonna hit the secondary market and you, you, you and I now have the opportunity to, to find them, whether it's as a backup for our current farm or it's just a cheap unit that We have yet to find out what the stabilizing point is. A lot of companies say, "Listen, okay, we can do AI, HPC, it's gonna be really expensive, you have to build all this backup generation, maybe I have to go build a power plant to make sure I can do it successfully. " We can't also just get rid of all the Bitcoin mining, right? Because if we, if we as a company serve, have contracts where we have to, you know, provide services to the grid, which is very common in Texas. We're still liable for that, and we're still providing a valuable service to the grid, being a curtailable load, being someone that, you know, ERCOT can call on and say, "Hey, can you scrub a hundred megawatts? Can you scrub two hundred megawatts?" that's still something that is, is valuable that you can't necessarily get out of, immediately as far as what it does in the long term. I, I think it would be very interesting to see what these hybridized, these hybridized grid connected, you Future. As far as how long that takes, it's really only constrained by the ability of folks to build these data centers, 'cause these data centers are substantially more complicated than what we've been doing in the Bitcoin mining industry. You know, Bitcoin mining industry, well, we're happy to put a shed in the backyard and just keep the suckers running. That isn't gonna fly when you're trying to run an AI cluster.

Yeah. and then while we're on that What do you think it means for hash rate growth? Like, would you say hash rate growth might slow down a little bit because of this? Because now what was previously this big, these big public miners, they're, some of them are pivoting either partially or wholly to AI. What does that mean for Bitcoin mining hash rate growth? Or do you just think like it's still just gonna grow, but just like a bit slower than what it was before? Or where are you at there?

It's a good question. it will have an impact, right? You know, the-- when, when you have the, the CEO of Foundry saying that they can see the network slowing down on their end because there is a snowstorm in Texas, a snowstorm in the American southeast, you go, "Okay, those are pretty big loads. That's a lot of, you know, potential megawatts that are gonna be transitioning to AI." the question then is, you know, what? What, what does that mean? So what it means in one sense is that you are now actually kind of decentralizing the network even more. You know, there was a big, a big hullabaloo when all of these, publicly traded Bitcoin miners were ramping up, where people looked at them and said, "Hey, if you wanna get state capture of Bitcoin mining, Having a bunch of gigawatt scale sites is a pretty scary place to put all this hash rate. Now that being said, the Bitcoin network as a whole is so large at this point that this whole idea of an evil nation state is going to co-opt x y z amount of hash rate, is still very, very infeasible, still extremely infeasible, just because at the end of the day, the likelihood of any state, any government on earth being able to capture that Many megawatts and allocate them exclusively towards some, some evil purpose in a coordinated way, is a little bit, a little bit silly at this point with the size of things. What it could mean in practice is that more folks have the opportunity in the public markets all across the world to continue to grow their operations. You know, it's very reasonable to think that hash rate would hang out at a level playing, playing field or continue to drop a little bit over the next year or so. But that being said, the private, the private operators know that a lot of these folks are, are rolling off on a lot of hash rate, so they may just be transitioning this, previously corporate hash rate into their private hands and just happily expanding. So, the net out is that it's good for, it's good for decentralization, it is good for reducing the risk of state capture because it's no longer a machine that's living in a publicly traded, you know, audited, regulated space. And in terms of what happens after that? We'll just have to see, we'll just have to let the incentives of the network continue to operate.

Yeah. Now, another common discussion point that's coming up nowadays, especially in the, let's say, in the bear market or this mid-cycle dip, whatever, whatever we're calling it, is this concept of cost of production. And what we're seeing is sometimes like a TA or an on-chain analyst will sort of look at that and say, "Oh, look, we're in the bear market or we're in this dip," and typically the cost of production has been kind of like Let's say a bottom or a, a bottom zone for the Bitcoin price, and so they will typically put up a chart and they'll say, "This is the cost of production." Now, I know you've got issues with this. Can you explain what is the issue with, this kind of analysis of static cost of production?

I love it. I appreciate you, you, tapping right in the soft spot. So, you know, bit-- Bitcoin, you, you talk to a Bitcoin miner, you, you find, you get fifty Bitcoin miners together, and you And you throw a chart up in front of them and you say, "Hey guys, here's the cost of production. " And you use, you know, the most famous, models that are out there, as a guy Caprioli who has one, Glassnode has one that they, they've produced, and very smart folks who have gone through and put a lot of time into trying to make an estimate so that they can have something very beautiful and pretty display on a chart. But if you get fifty Bitcoin miners and you get them sat down in a room and you show them these metrics, they will all look at you like you are an insane person, because what they understand is that there is no cost of producing a Bitcoin, but there are costs of producing a Bitcoin, and it's totally contingent on the individual who is actually producing the Bitcoin. And so why are we, why are we splitting hairs here? Why does this matter? it matters because there are certain data points and statistics and things in the Bitcoin space that you can know. You can know, you can know price, you can know difficulty. You know, price, you go to an exchange and maybe there's an arbitrage opportunity between price, but you can very, very accurately get a sense of, you know, what is the price of Bitcoin right now at this very moment. And you can do the same when you look at the difficulty. Why? 'Cause difficulty is a single number that is in the protocol and it changes every two weeks and is shared amongst the nodes. And so you know with a hundred percent assurance what the difficulty is. You know with extremely high, high rate of accuracy what the price of Bitcoin is. And when you try to take something like the cost of producing a Bitcoin, you have to say, "Well, how do you know the cost of producing a Bitcoin? " And anybody who's trying to make a chart has to make a- A lot of assumptions. You have to make assumptions about the cost of people's input electricity, you have to make assumptions about the efficiency of people's operations, you have to make assumptions about, the, the way in which they're able to run their machines, kind of ignoring some outside factors that may be kind of inconvenient, and then you layer it all on top, at least as far as these metrics are concerned, with the notion that all of these metrics are backwards looking, and they are all backwards looking. Regulated by what the difficulty is and if people turn off or turn on. So it's not a very useful metric. It could be useful by argument to say, "I wanna know if there's some kind of two week tension happening in the Bitcoin miner universe, "and you might be able to say with x percent accuracy that, "Okay, I have a sense that there's more tension or there's less tension. If mining is expanding and the difficulty is going up, I would say, "Oh, it seems like there's less tension in the system. " If the difficulty is going down and you're, you're seeing the, the opposite happen, you might say in the macro sense, oh, it seems like there's, there's more tension, there's kind of a tightening because people are having to turn off and you're, you're hitting this marginal, this marginal miner. And you can say, you can say very rough, very high level, very macro-y type things about Bitcoin mining if you use those metrics. To those fifty miners sitting in a room, you can say nothing to them because each of them has their own cost of production, each of them has different machines. One guy may be getting all of his electricity for free because he has a deal, an off-taking deal, with the guy who, who owns the generation. The economics of that deal are entirely different from the guy who only operates on solar and only operates fifty percent of the time because the sun's down the other half of the time. That guy's economics are totally different from the guy who is heating his house. That guy's economics are totally different from the guy who has a little five megawatt site and he's buying into the, the futures, the electricity markets and the day ahead, in the day aheads and trying to basically hedge price or make a little extra off of that. So If you wanna look at things from a, a, a pretty single metric, a single chart, when you start to dig into them when it comes to Bitcoin mining, you realize there's actually very little of use that you can say. Anybody who's telling you that there's a single number that will give you the floor price of producing a Bitcoin doesn't understand the metric they're actually quoting, because you can't say that. Now, it might be helpful, you know, I'll, I'll, I'll give people credit to the, the folks who make this metric. It may be helpful

multi-year sort of expansion and tightening cycles. But there is no way on earth that you can say with any degree of confidence that the cost to produce a Bitcoin is eighty-two thousand dollars and we are below it and this means something terrible. It just, it doesn't exist in the world of Bitcoin mining.

And even more, I've seen sometimes people post up a chart showing the, you know, the cost of production in these different countries, as though there's like just that one country. This is the cost for, for the USA, this is the cost in, whatever, Norway or- That's right. And it's just like, even that is just not granular enough, right? But let me steal that just for a second. Could it be that maybe in the early years of Bitcoin, when maybe, you know, let's say some of these other uses weren't really a thing and most people were just sort of paying a flat rate and you, if you just kind of assumed, let's like an average You know, commercial rate that somebody can get on a PPA, a purchasing, power agreement. Would the metric have had some validity in those earlier years, and it's, you know, it's just lost that over time because now, like, there's all these different uses and heat reuse and so on, or do you think like it never really made sense? I mean,

at the end of the day, we're talking about a decentralized network, right? You know, it, it would have probably even been less useful in the beginning days of Bitcoin for one simple fact, which is Operating under the pretense of one CPU, one vote, they didn't care what their electrical cost was. You know, if you, if you read, if you read about like, some of the early folks participating in Bitcoin mining, and you read, you know, Hal Finney's posts about, about experimenting with Bitcoin mining, there's a post where he goes, "Okay, I was mining some Bitcoin, it was fun, and then my room got hot, it was kind of annoying hearing the fan running all the time, and I turned it off." These guys weren't Thinking about business models, you know, they weren't thinking, and, and it would be many, many, many years until people were really in that mindset. there, you talk to folks that were Bitcoin mining in earlier days, and a lot of folks, I've heard stories of guys who were shocked to find themselves getting half a million dollar, a million dollar electrical bills because they thought they were just gonna set up, you know, walls of S-nines and didn't even know what their electrical rate was. So the, the maturation in the industry is- Something that has absolutely skyrocketed in the last, you know, three to five years. To think about, if these metrics would work in the early days, I think it's a, it's very much a different cohort of people who are even doing this kind of stuff.

Yeah, so it just doesn't really make sense, I guess. let's go forward now, so into your spotlight paper. You had a really interesting section around, Henry Ford's Crystal Palace, and you were talking a little bit about what he had to standardize to make that workable. So can you just explain a bit about that? I think it would be interesting for listeners. Sure.

So we, we draw this really, this really clean comparison where, you look at what was happening in the twentieth century through the Industrial Revolution and the The Industrial Revolution was this revolution, obviously, in, in productive capability, and we think about it in terms of, of energy consumption and production, you know, this, this core thesis, this idea that energy is human flourishing. And Henry Ford had this Highland Park campus. It was a hundred and thirty-acre campus that they built their moving assembly line. And the moving assembly line that Henry Ford built was this enormous, sort of industrial space, this warehouse space that, unlike a lot of, like, you know, Previous, previous industrial era warehouses had these long rows of windows, and I'm sure you've, you've seen things like this in, in the US, it wasn't that common at the time, but these long rows of windows would allow folks working inside the factory to actually get sunlight. And so this, this building was nicknamed the Crystal Palace. So what was it? It was the focal point, if you think of it in terms of nodes, it was the node for where production happened for the Model T. What it took to make that node operate continuously. And Henry Ford had a very specific goal that he wanted, which was this idea of continuous non-stop production from a raw material into a finished product. And he had this tool, the moving assembly line, that was going to allow him to do it, but he had to- Really control the inputs coming into that node, so he had to go and build energy producing assets. He had to build power plants to power them, he had to have coal, so he went and bought mines. To go and produce these, these vehicles at scale, he had to go, buy forests. So he's producing hundreds of thousands of acres or buying hundreds of thousands of acres of forests. He's having to manage all these supply chains and he's having to draw all these raw resources to Highland Park, Michigan so that he can feed this moving assembly line. This crystal palace. And his goal was continuous nonstop process. So he does things like he standardizes the wage, 'cause you have to have workers, obviously, right? And he's really trying to control everything to the point that he doesn't even have to warehouse, he's just going, going, going, going. Now, how does this tie into what we do today, right? his aspiration, this idea of, of continuous production, he couldn't get it in his era, the technology just wasn't there at the time, right? And if you hear about, you know, car manufacturers in the twentieth century, You talk about, you know, particularly, like the to-- Toyota productions and things like this, this sort of like, continuous production methodology, like people have continued to try to optimize for it, right? The just-in-time manufacturing, that you've heard of. What do we have in the world of now, loads like Bitcoin? We actually have a potential to achieve the idea of continuous nonstop production because our crystal palaces, instead of being giant warehouses with glass walls, they live in little chips. They live in silicon chips, and these silicon chips, totally unlike Henry Ford's giant campus, can be put anywhere on Earth. So you have drastically shrunk the, the supply chain requirements to power these crystal palaces, and you now have, by dropping a Bitcoin mining machine anywhere on Earth close to a source of energy, the opportunity to actually execute continuous production without waste. When I turn the Bitcoin mining machine on, it is hashing, I can be rewarded for that hashing because I connect to a pool that pays me, and when I turn the, the energy off, for whatever reason it might be. There's no warehousing, there's no like backup of hashes, like, "Oh crap, there's too many hashes in the machine, like we've gotta do something with 'em, they're gonna go stale. " It's an instantaneous process. And so when it turns off, I'm no longer producing hashes, and I'm no longer being paid for it because I'm producing a commodity. So we use that language of the crystal palace as a way to frame what's similar, which is that they want to have continuous production with discontinuous inputs, but to show why what's happening in the twenty-first Kinds of loads like Bitcoin and hopefully AI, you know, HPC loads can be a little more location invariant in the future. it's radically different 'cause we're actually achieving the goals of Henry Ford.

Yeah, it's really fascinating idea. Take us to this idea of monetizing the megawatt and some of the different examples you have. So, talk, talk through this concept of monetizing the megawatt, you know, into, the Bitcoin mining

relevance. Sure. So when- When we think about monetizing the megawatt, you-- most people think about Bitcoin mining and they say, "Okay, well, you have to buy--" They start from the, the, the machine, right? You gotta buy the most efficient machine, and then you gotta go hunting for electricity, and you have to go plug in and, and maybe negotiate and strike a deal. Figure out what you're going to, set your operating cost on, and then you just kind of sit there and pray and hope that, hope that the difficulty doesn't get too hard and that your machines don't become obsolete. And, you know, in our view, this is a very, this is a very thin understanding or perspective on the way that Bitcoin miners are actually operating in the world. And it's not 'cause people are dumb or bad, it's because they just don't know enough Bitcoin miners to understand how much folks are thinking about how to operate their, their operations So the way that we think about it and the way that we, we share it in the paper is sort of as this balanced equation, right? And on one side of the equation, you have the revenue that you generate from hashing. And yeah, that's one way to generate revenue. Okay, well, what are other ways that people are generating revenue? Well, they're also generating revenue through what we call a heat credit, right? This could be the idea of, "I'm going to use the excess heat, to, to, to, push into a glycolated line that's going to melt snow off my driveway or it's going to heat my home." how is that a credit? Well, it's a credit because I'm not just having a sunk cost of propane or natural gas or electricity, Reducing a bill somewhere else by using this heat, it's a utility to me. And then we had this idea as well of a load credit as something that you can generate revenue by. So there's really, an an example of a load credit would be something like, I'm buying in the day ahead markets when I, I want an assurance of a price. So today, it's, you know, X, I, I'm gonna buy, you know, a couple of hours of, of megawatts for, four PM, five PM tomorrow, and I Gonna get a good price on them, but if tomorrow comes and that price is way, way higher, I can actually sell that, that load back into the market and I can make money on the spread. and that's kind of the simplest example of it. There's all kinds of stuff, very fancy financing, very fancy programming that's happening in ERCOT where you're getting, certain kinds of rebates and certain kinds of, bidding into the markets and certain kinds of programming. this is a really expansive opportunity for folks who are providing a lot of Price and grid stability to, to the grids that they operate on. however, these are all ways to generate revenue. There's not just one way you generate revenue in a Bitcoin mining operation. You generate revenue from the hash rate, of course, the heat credit, the load credit. And if you think about it in that way, and then you back out what it takes to run these operations, you know, the primary cost being the electrical cost, the operational cost, and then the OpEx, whatever it costs to, you know, keep your filters up, keep your systems running, replace needs to be replaced, you wind up with an understanding of what the profitability these businesses can actually be. And when you look at the shapes of them, they're very different from each other. They're very different 'cause they're trying to optimize for different things. Yeah, interesting.

So, just playing it back a little bit. Let's say the older understanding of Bitcoin mining, or let, let's say the more naive interpretation, is just like, get the mining rigs, find the cheapest rate for power, and, you know, plug it in and sort of hope for the best with network difficulty. And of course, there are more professionalized forms of that, but that was kind of a model, let's say. And then what we're talking about now is more like, well, hang on, let's expand out our possibilities and consider different forms of revenue. Some of that will obviously be Bitcoin mining. Mining, you know, hash rate, we get paid out Bitcoin and obviously sell that in the market or whatever, whatever we need to do to pay the expenses. But there's also this heat credit, there's also this concept of, flexible load or this idea of curtailment, and so these are other ways in which we can- get revenue as a Bitcoin miner, and so, yeah, I guess that, that's probably, how I'm reading that. do you wanna just talk us through some of the notable examples?

Yeah, happily, and I'm, I'm, I'm happy to actually start with one that, was probably the largest misconception. So under this naive view of, you have to be big, you have to go and buy the most expensive machine, you have to be the most efficient and the most aggressive, it led to this misconception in the, in the marketplace where people assumed that if that was true, well, all of the megaminers are just gonna eat the world because they can buy the most machines at the best rate and they can buy the newest machines and they can go- They'll get the cheapest electricity. The assumption was that, oh, the mega-miners are going to consume the Bitcoin universe, and look at that, they have some unfair advantage in their economy of scale. And the, the short answer is that that's simply not true, because the unit economics of a Bitcoin mine are contingent on the unit economics of the Bitcoin mine, not the scale of the Bitcoin mine. When you look at examples and, and take the mega mining example. They are at the point now where they have access to all of the tools of high finance, and when you layer on-- and this is maybe a, a rough example to start with, because it, it very quickly becomes the most complicated-- when you layer on the treasuries that these, these Bitcoin mining companies are holding, you know, tens of thousands of Bitcoin, they are now able to play very, very advanced financial games, not just in the treasury, but in the production of the asset. That which feeds into the treasury. So they try to create this kind of, this kind of beneficial loop where the, the asset that they are producing, they are able to not just, produce it in reference to the unit economics of production, which is, you know, the electricity cost, the opex, yada yada, but then they think about the full life cycle of that as, wait a second, once we actually have the Bitcoin now in a treasury format, we start to re-monetize that in the financial system. So they're creating a capital side of it, yeah Of like transmuting it this like incredible, and complex, mechanism that they're, they're operating under. That being said, what is more competitive with this kind of mega miner way of operating? Well, one of the most beautiful examples of this that we feature in the paper is the example of gridless compute, and loads of people have heard about gridless compute, and it has this kind of gut, you know, this-- it's very exciting 'cause these guys are out, out in the bush and they're cruising all over Over Africa and they're finding these big generation assets that are underutilized and they, they go out and they drop a Bitcoin mine on it and all of a sudden these generation assets are, are producing way more, way more electricity and they're making money. but what people fail to grasp because they're kind of thinking under this old model of bigger is better, is that. That model, the gridless model, is one of the most robust, ways of mining Bitcoin because what do they do? They do something very, very interesting. When they find one of these underutilized resources, and say it's a five megawatt hydro facility that is only producing, you know, one megawatt. They go to the individual who owns and operates this facility, and they don't say, "Let's sign a deal where I'm gonna pay you two cents per kilowatt or five cents per kilowatt. " They sign-- they, they, they show up and they find the person running the facility and they say. Would you like free money? And they say, "Um, yes, I think I would. " And they go, "Great, here's what we can do. " The increased marginal cost in producing on hydro. Whether you're running at one megawatt or four or whatever your capacity is, it's tiny, it's nothing. You know, you buy, you buy more grease for the bearings, but this is fifty, hundred year old technology. This isn't stuff that, I have to go buy more coal to run my power plant. The, the increased marginal cost on producing one versus four megawatts is nothing, nothing. Gridless comes and they say, "Hey, we'll take the rest of that, whatever your capacity is, whatever you can give us, and we're not gonna compete with any people in town who want to buy the energy, and they can't because the people in town who are buying the energy are typically buying it at, unfortunately high rates, you know, thirty, forty cents a kilowatt hour. What we will do is we'll take that excess energy for free, and we will give you a profit share. Now what does that do for the model? What it does for the model is that you have something that is just pure waste. Purely wasted opportunity cost, which is the ability to produce additional electricity, and then you have Gridless, who now is monetizing all that excess with an input cost of zero. It doesn't matter what pub co you're looking at, they are never going to get their input costs to zero.

Yeah, it's not gonna be an example of where cost of production, if you're looking at the, the TA and chart, chart, chart analyst guys, it's totally wrong for Gridless, right? Because what is their cost of production? I mean, it's ridiculously low. Now, to be clear, I've done an interview with, Philip Walton, the CTO of Gridless, I think that was from, October last year in Lugano at Plenary The thing is, like, look, it's hard to do that, right? Not everybody has the skill set to do what he and the team at Gridless are doing. Exactly. Right? So that's kind of, yes, it's very low cost, but not everybody can do it.

A hundred percent. And this, this I think is where the subtlety becomes, at least, at least from our perspective, a lot more interesting, because the people who are willing to go through the work to do this very difficult stuff, 'cause they're, they're, they're operating in the remote Roads to get to this asset, to get to this community, how they set things up in a robust enough way that it can actually operate remotely and be monitored remotely, and then how do you train people on site who can actually manage the operations of it and keep it moving? So that model is incredible, but no model comes without its own externalities. You know, there's, there's other examples where, the one I love to give is that, you know. Me, at my house, here where I am, I'm always going to be able to produce Bitcoin at a lower cost than every large publicly traded Bitcoin mining company in existence. Why is that? It's because when we bought the house, the house had solar panels on it. And the solar panels after the first year here, we saw covered the full cost of electricity, so the, the bill just kept coming back zero. And then of course, at the end of the year, they give you like six dollars for whatever you produced, right? Well, what did I do? I set up a Bitcoin miner in my basement, and in the summer I just vent it outside that, heat exhaust, and in the winter I turn it around 180 degrees and I pump all that heat into my HVAC system, and so it does a little offset on what my natural gas costs are, but the cost to run that Bitcoin miner is zero, it doesn't cost me anything. Why? Because instead of a hydro facility in the bush. I have a solar, a solar array on my roof, and that solar array, it's already paid off, and it doesn't, it doesn't take more to, to produce more electricity from it, it's just the sun shining. and it's a very kind of passive, very kind of, kind of dumb way where as long as I don't have, you know, 20 machines plugged into my, into my, into my system, I don't get an electrical bill. The cost is still zero and I'm still producing Bitcoin and getting the utility of the heating. So, yes, these are new modes of mining Bitcoin. Yes, they have difficulties associated with them. There's been this explosion in open source mining tools and folks that are trying to solve for these problems. Are there beautiful clean consumer solutions here today? Kind of. We have these baseboard heaters, we have these bit chimneys and all these other tools that people have. are we fully integrated into, you know, commercial HVAC systems? Absolutely not, we're not there. But the case is proven out and people understand that this is a, a highly efficient way to, to actually run your, your Bitcoin mine.

Yeah. And, so this heat, while we're on the heat thing, I guess we're still, we're seeing, as you said, like this, this retail use heat These miners that are also heat, home space heaters and this kind of thing, then, yeah. So do you wanna also talk to us a bit about, this concept of flexible load? So I know you have an example there from, Megawatt in Indiana. Yeah,

Megawatt are fantastic guys. They, they are, They're fascinating because it's sort of like, I, I spent some time, when I, when I was working in the Bitcoin mining industry specifically, getting to experience a lot of the way pubcos think about managing load, and they have these brilliant, brilliant guys and gals who are, essentially executing very, very complex trading regimes on the asset that they have, this energy, and it's hundreds of megawatts at a time, and you can make a lot of money. If you are particularly on ERCOT, capitalizing on these certain spikes in price or, or inefficiencies that are happening in the market, what's fascinating to me is that Megawatt, a privately held company whose strategy is to find these smaller sites, you know, on the order of like two to ten megawatts, they have figured out the same sorts of techniques that a lot of these pubcos are doing, and they're doing it on a smaller scale. Now, why is this interesting? This is interesting because- What they're able to do is they're able to not, you know, steward hundreds of megawatts and create this huge influence on the grid, but they are, they are in effect Bidding into day-ahead markets that gives them some price stability on their production, 'cause they know their break-evens, right? They know what machines are running and the efficiencies and the locations. It also gives them the opportunity to generate a little bit of extra revenue when you wind up with price disparities when that day actually shows up. So I buy for tomorrow, and I, I buy because the price looks particularly low, but maybe I think that the price might, might go up. I can make an arbitrage if, if tomorrow comes and all of a sudden something happens and the price is substantially high Higher, I can make an arbitrage trade on selling that back to the grid, which isn't my, my sort of main source of revenue driving. My main source of revenue driving is still coming from mining Bitcoin, but what I'm doing is I'm actually able to smooth out my business operations. And from the perspective of the grid operator, 'cause you have to remember, if you're, if you're grid connected, it's not just you sucking up megawatts, it is you monitoring a pricing mechanism, trying to figure out some sort of, Most people aren't doing PPAs or trying to do kind of live price management, or bidding into the markets. it's you in conjunction with the grid operator, the grid operator who is managing the functioning of this grid, because their mandate is to the consumer to make sure that the, the frequency of that grid is consistent and that you're not, you're not browning anything out, blacking anything out, or blowing anything up. so you're now kind of playing this game with this, this, this grid entity, and you're providing a bit On the financial side of things, and the unique thing about Bitcoin mining that makes it different from sitting in an office and just trading energy futures, radically different from trading energy futures, is that the worst case scenario is that, if you are properly hedged and a trade goes the wrong direction on you, you can take delivery. You know, you remember during COVID when everybody was buying, you know, tankers of oil because the, the, the price went, you know, negative, and then they figured out that these were all physically deliverable options, and they were like, "Oh crap, I need to figure out what to do with, with these, these train cars of oil that I just purchased." Bitcoin, Bitcoin miners that are properly hedged and properly financially managed don't have this issue because if they are participating in the financial side of grid pricing, they always have the fallback. Which is that they can just take physical delivery of the electricity and use it for mining. Yeah. It's amazing, they, they, they are providing their own hedge. Interesting, yeah.

yeah, and then, yeah, so, some other interesting examples I saw, obviously the upstream data is an interesting one, you know, I've done podcasts with Stephen, but for listeners who aren't familiar, give, give people an overview there and what's happening with the upstream

data. Yeah. Stephen's amazing, and his, I've, I've had the pleasure of like seeing his model just evolve and evolve over time. And the one thing that, you know, he, he has consistently said is that I, I run an oilfield services That he's always said, but I think it goes over people's heads, exactly why he's saying that and what that means, because they think, "Oh, you know, it's Steve, he's the, he's the black box guy, he's the guy who builds these, these upstream data, remote load centers. He builds these hash generators, which is a big engine on one end that then has a Bitcoin mine on the other." And Yeah, that's-- those are all things that he builds, but fundamentally he's an oil field services company. And what does an oil field services company need to do? If you look out into, you know, the, the, the Great White North, if you're in the, the US and to Canada, and you look at Alberta and you look at Saskatchewan and you look at all the oil exploration that's happening out there, you have this really robust industry where people are going out and, and exploring for oil. And when they explore for oil and they decide that they're gonna- develop a particular field, you have to create these islands, you have to create these hubs, if you're going to put a pipeline in these collection hubs. And you have this thing where it's sort of like building a town. you don't know how many people are gonna move to the town. You don't necessarily know, if there's gonna be twenty thousand or fifty thousand or a hundred thousand In the same way, you don't know really know you can guess, but you don't know explicitly how much gas is going to be produced. And because of that, when you're building your energy generation, like you have to put big generations at these islanded sites, you have to always estimate on the high end. You have to always estimate on the high end. It's why your electricity bill has like a peak demand charge, because the grid is never allowed to be too small for what it is needed for. It always has to be sized to the maximum load. You're never allowed to, not have enough capacity. So what does he do? Steve says, "Okay, listen. What are people doing now? When they're doing the exploration and they're building these islanded sites, they have to put these giant generators, and it's because they just don't know what the oil development is going to look like. And those generators, they can be rented, they can be purchased, but they're expensive. Remember, you're, you're throwing a very large engine on this site and you have to run it. And then what happens is in the early days of actually developing the site, there's just not going to be a lot of load on that engine. So if it's a, if it's a megawatt engine and you're running it at ten, fifteen percent capacity, it's just not a very happy engine. It wants to be operating at a higher heat threshold, at a higher load, at higher RPM

Tools that are, are backed by Bitcoin. As he goes and he says, "Listen, okay, you can rent this engine. It's gonna be a huge pain in the backside to run. It's going to really get throttled when you're ramping up and have all these wear issues. you could do that, or you could come to me and you can actually pay less for your engine rental. And the only trade-off is that we're gonna figure out some arrangement where either I'm gonna be mining the Bitcoin or we're gonna do some kind of split on the Bitcoin. But when it comes to you, what you care about, which is developing an oil and gas site, and you might not even care about the Bitcoin, all you care about is a generator, and all you know is that you have to oversize it. Steve can drop one of these hash gens on your development site, and he can say, "Okay, maybe at the beginning, I'll have to run more of the Bitcoin mining 'cause there's not enough load on the engine. And maybe as you ramp up production, we'll have to dial it back, but you don't even have to care about that. I will manage that whole problem because you're only concerned about producing the oil. You don't even have to be concerned about the Bitcoin. All that you know is that you're gonna save a lot of money on the operation and rental of this generation over the long term. So comparatively, it's an easy sell. It's an easy sell, it's a good deal.

Fantastic. I, I think we've got time for one more example. can you just talk, maybe pick one example that you think might be interesting or, yeah, maybe the, what have you got? Cholla in, in, yeah, I'd love to talk about Cholla. Also got, oh, Futurebit. Yeah, maybe talk, talk about the Cholla one.

Cholla, Cholla is great because they, Cholla, you know, Cholla Petro, Gideon Powell is the entrepreneur that manages that. It's within his family and Brad Cuddy is the lead on the Troya Innovation Lab, and this is one of the ones that I think is, is a fantastic way to end, because I wanna, I wanna root us back in the bigger picture, which is that everything that we are doing, it is important, it's incredibly important to keep the conversation as connected and as open as possible. And when you talk to Brad Cuddy and you ask him about the Troya Innovation Lab, he will say in no uncertain terms, "We- Pay more for our labor, we pay more for our opex. We are strategically located in West Texas where we can get access to ultra low cost of power, but this Bitcoin mine doesn't exist to maximize the revenues that we can generate in Bitcoin. He says this Bitcoin mine exists as a testing base, as a petri dish, as a place where you can bring in the politicians, you can bring in the entrepreneurs, you can bring in the technology, you can- You can put in an upstream data hash hut, you can put in a gigahut, you can try new types of firmware, new types of operating systems, new types of management systems. That's the purpose of the Choiya Innovation Lab, and this is something that doesn't really exist anywhere else in our industry, and that's why it's so important what they're doing, is that they, they have controlled for the, for one factor, which is ultra low cost of electricity. You look out to West Texas, West Texas is where, you don't just have kind of the, the, Oil and gas world of, production, but you also have tons of solar and tons of wind, which very often pushes electrical costs to zero or just, just under zero. if I can control for that ultra low cost of energy, I can use this facility as a conversation hub, as a community innovation hub, where we can instead of being sandboxed on our own sites and producing information and, and not really being in a position to share it or further the conversation on the- This industry, I can actually produce information that people know is good information. We're an innovation lab, you know? We, we are where you go and where you send your equipment if you want it to be tested, and if you wanna have conversations with the people who are actually on the grid operation side, people that are actually writing the policy. So it, it, I think is, one of the most beautiful examples of what we should be doing and continuing to do, which is creating spaces that we can produce the high quality information, not to say anybody's doing it. Anything wrong? You know, you're, you're not doing anything wrong if you're just, you know, privately mining your Bitcoin or publicly mining your Bitcoin and minding your business. But in terms of educating people on the power of these types of tools and operating across all these kind of typically siloed parties, there's, there's no more exciting place, I think, than the Troy Innovation Lab because of what they're doing.

Fantastic. Yeah, I think there's just so many cool examples. So I definitely recommend listeners check out this report, make sure you share this podcast so people can listen to in the show notes, we'll obviously we'll link the, spotlight paper, and, I guess Robert, just before we let you go, anything else coming up at Bitcoin Park that people should know about?

There are some things coming up, so we're, we're super excited because, About a month from now, just under a month from now, we have what's called the Bitcoin Takeover in Austin, Texas. So it's gonna be two days, proof of payments on the first day, proof of work on the second. it's gonna be the 12th and the 13th, Thursday, Friday, and that is at Bitcoin Park, Austin. So we have some brilliant folks coming in from the payment side of things that are going to be talking about everything from how do I scale payments infrastructure from, you know, one to billions of people, Scape look like? What are the hurdles? What are the conversations that are happening across the large companies in the space, all the way to how do I build UX to make it accessible to people? So this is, this is our, our event, it's sort of the, the, the re-indieification of South by Southwest, which has decided that they wanna go mega. You know, we wanna keep things small, we wanna celebrate all the indie-sized building that's happening in the Bitcoin space and this sort of payments and entrepreneurial side. So that'll be the twelfth and th Next month of March, and then after that, we're going into, going into a ramp up for the Texas Energy and Mining Summit, which is going to be a couple months thereafter, and the Texas Energy and Mining Summit, and if, if you can't tell by how excited I am, by, Detroit Innovation Lab, this is gonna be in May, so May 20th, 21st. The, the theme that we're working on for that is this idea of, you know, Nashville was monetizing the megawatt, Texas is Laboratory Texas. What we are focused firmly on is celebrating the heck out of Texas is the place in the world where you have West Coast tech bros, West Coast brains in the programmatic space, combining with East Coast financial brilliance. And this isn't just folks coming from the East Coast, but also, you know, Dallas is standing up their own, their own stock market, and they are innovating across every single level of the stack on what it means To be a large load and operational load on a grid in the state of Texas. It, it is an example for the rest of the world, and it is happening nowhere else on Earth. Nowhere else on Earth do you have this level of intellectual horsepower actually producing innovative solutions live in living color in the state of Texas on ERCOT. So that's what we're focused on for that event on May twentieth, twenty-first.

Fantastic. Well, yeah, and I'll just say from my side, I, you know, I've been to some of these events myself. I will actually be at the, the Austin one, and, yeah, listeners, check it out. And of course, if you enjoyed this, make sure you share this episode. Robert, thank you for joining me. Stefan, it's been a huge pleasure. Can't wait to see you then.