The Toll of the Halving on Large Miners

Adam Traidman (SBI Mining Chip Co), Jan Cepek (Slush Pool), Igor Runets (Bitriver), John Quigley, MinerUpdate

[00:00:00] John Quigley: Hello, everybody halving, welcome to this segment of the Bitcoin magazine, having live stream. My name is John Quigley. I I'm the head of research for crypto mining publication, minor update.com. And I'm very excited today because I'm here with three gentlemen who have a deep, deep knowledge of the crypto mining industry.

And today we're going to be talking about the impact the halving is going to be having on large mining farms. And we're also going to be talking about the impact, the having will have on hardware. But before we do that, we're going to get introductions from these three gentlemen. So without further ado, let's kick it off and let's start with yourself, Mr. Adam Traidman.

Adam Traidman: Great. Thank

you very much. good evening here from Tokyo. I am the CEO of SBI mining chip, which is SBI the large Japanese investment [00:01:00] banks foray into the design and manufacturing of advanced mining hardware. in addition to that, I also happen to be the CEO of SBI, and ripples joint venture called SBI ripple Asia, and also co-founded.

And the CEO of a company called BRD, which is responsible for the bread wallet app that many of you hopefully have heard about over the last many years, by virtue of sort of my past, I was originally an electrical engineer and computer chip designer based in Silicon Valley and, have turned to serial entrepreneur had been the founding CEO of several companies, raised a couple hundred million dollars and built great products and teams and exited those companies.

I get into crypto about five years ago. I was very skeptical at first and then realized that, since banking and lending are the largest opportunities in the world that, those investors who were pouring money into the industry weren't  just crazy as I initially thought. And perhaps today I am as much of an evangelist as everyone else, so happy to [00:02:00] be here and look forward to the discussion.

John Quigley: Okay. Sounds great. How about you Jan?

Jan Cepek: Yeah. Good evening everybody. My name is Jan Cepek and I'm a co-founder and co-CEO of the  company called Braiins. we are the company who's been running, operating and developing slush pool for the past seven years. So that's about the time when I got into Bitcoin and it was back in 2013.

Besides the slush pool project, we are also known for having an open source mining firm called Braiins OS, where we try to provide an alternative to the factory firm wares of the mining devices. so that's pretty much what we do.

John Quigley: Great,


Igor Runets: Hello everyone. My name is Igor Runets. I am, I am a founder and CEO of Bitriver

mining is the largest mining data center in Russia and the CIS region. I got in to crypto in [00:03:00] 2015 from traditional data centers and telecommunications space. And, now beyond the, just a Bitriver mining we also operates an energy company called Faraday energy company. So. Bitriver mining data centers and faraday energy company.

John Quigley: Okay. It looks like we have a very decentralized panel with them. Yeah. Professionals in Russia, Europe, and also Japan. For the first question I was wanting to put out there, it's an open question. So anybody can feel free to jump in and add their angles. all of you have experience with big mining farms and big mining operations over the past six to 12 months.

How have these operations been preparing for the having?

Adam Traidman: I would say that from our perspective, and we are not only a manufacturer of minors, [00:04:00] but we're also a large customer. Purchasing minors, that we developed internally, as well as those on the open market. And I would say that up to the havening, as you probably saw during the previous halvenings, you see an expansion of operations.

People want to take advantage of, the electricity rates that they've negotiated in order to mine as much as possible with the best hardware in class before the halvening happens. post havening. We expect that, you know, a lot of miners will be basically gutted from the market because electricity prices in the farms won't be able to support profitability, post havening, however, you know, those that were at scale prior.

who were able to secure great, electricity prices previously we'll, we'll continue and will continue well. So I expect that, you know, the first half of this year is a great couple of quarters for Bitcoin manufacturing and hardware companies like us.

John Quigley: Okay. in terms of. Upgrading equipment prior to the having, there [00:05:00] was a lot of talk about, big mining farms having to upgrade to the latest gen, to increase their chances of surviving come past the halving, but have any of you experienced or believe that some mining farms that are lowest on the cost curve and are extremely efficient?

Maybe waiting to, after the halving on seeing some miners go bankrupt and trying to pick up their mining rates at a lower price.

Jan Cepek: That could be certainly the case. The question is if we are going to see the same kind of craziness that happened, for example, back in 2017, if there is going to be any significant.

A price action. I did some calculations before, before our panel. And actually I do publish those quite regularly on Twitter, on my own Brain’s, accounts, where I try to use the, efficiency of a minor compared to the current difficulty. And you plug in your [00:06:00] electricity price. And actually one of my colleagues turned it into a nice Tableau app where you can play with the parameters.

And it turns out at the current difficulty levels, the margins for, for miners who are, using their, as for example, S9s. at like 86 joules per terahash, which is done. Every setting. You could probably go a little bit lower than this, on the efficiency level. they are, having a margin of like 50%. So that means at this price level of Bitcoin, this margin would be wiped out on Monday.

But, so, so I am assuming that. A lot of farms are kind of like waiting. What's going to happen after this point in time. And at the same time they are getting ready for replacing the equipment, but they still see something like six or eight weeks of being operational. If their electricity costs. According to my calculations are around like [00:07:00] four or 3.5 cents, which, I'm assuming that the big miners should be in this range, including the operational costs.

Adam Traidman: Hey, Jan, it's Adam. I agree with you. I think that, you know, for, for the sort of large-scale miners who are operating with older equipment, like S nines, or perhaps the generation after like an s9 pro, somewhere around that electricity price, they're going to be taken off the market. But, so to, to answer the original question, I think actually people tend to upgrade after the halvening.

Not before. And the reason is they want to milk every dollar they can get from their existing hardware. Hopefully they've placed orders. So beforehand, because there's a lead time on getting the hardware rack. And right now, TSMC is way over capacity. and so, you know, even though with coronavirus and everything, and I, and I a reduction in say iPhone and iPad, you know, orders and Huawei having issues as well, still the orders for wafers that are being produced now replaced six months ago before Corona or three to six months ago.

And so you're going to see, you're going to continue to see [00:08:00] a, a lack of supply. And so I hope those folks got the, those orders in. Nevertheless, what I would say is this, while even after the halvening. When those folks who are, operating and who are, you know, barely flirting with profitability or breakeven, I should say, or just a little below, there's a human psychology aspect to it as well.

People don't like to be wrong. They want to think that price is going to go up. So sometimes they keep operating those miners. So for example, what you see is that the network hash rate will, will definitely go down after the halvening, but it'll, it'll go down over time. Right. And then as people get new miners to replace those old ones and they turn off.

Right then you'd expect it to come back up. And there's always this sort of nice, ratio between the, Bitcoin price and the total network hashing power. And we expect that to continue to, you know, to, to show the same type of correlation, because it sort of has to for, for the economics to work. But what we're going to see after the havening is that those people with the capital to buy the most power efficient miners and those people with the [00:09:00] lowest electricity prices, they're going to be the last ones standing.

And, and that's what it takes to win. in the operator space.

Igor Runets: We saw slightly a different dynamic in our data center. Almost all our large clients sold off all the generation machines by the end of last year, as it sold to retail miners in Russia. It's kind of interesting dynamics in Russia for retail consumers.

electricity price is very low in some regions like around one, a U S dollar cents. And like all our large clients, we have clients from many regions like Japan, China, the United States, Europe, and almost all of them, people to kind of sell a significant portion of that older generation machines by the end of last year.

And Whoa, what kind of installed on our local feed origination machines. so like, [00:10:00] Slightly different dynamic, yin our case,

Adam Traidman: you have

smart customers, Igor those that got in early and place their orders early. That makes the most sense.

John Quigley: I definitely want to come back Igor to some of the attractive elements of mining and Russia, but I want to touch a bit more on the s9.

It was some research recently by an analyst coin metrics, [unintelligible], an estimated that. The current network hash rate is represented by 23% is represented by Antminer S9s. How is this going to look in the months after having will the Antminer S nine become obsolete, or we still see it represent a significant share.

Adam Traidman: Th they, they have to become obsolete because even if you still have them and your electricity price is low enough such that you can operate them barely profitably. You're doing that as an opportunity cost of making a significant more amount of revenue. If you up [00:11:00] grade those machines. So the trick is this, can you, or do you, can you find, or do you have the capital to replace them?

And the industry has been built on this notion that you have to pay ahead of time for the minors upfront. And that really precludes. People going in and basically replacing things en mass. Right. And that's why you've seen some mining companies, as you said before, either they fail, or they get bought up for pennies on the dollar after halvenings.

and then companies like ours since we're a bank, you know, we offer financing for miners very much. Like you might get financing when you buy a car. And you know, that financing with installment payments, affords companies, the ability to use the asset to actually pay back that the cost of the asset. And while the interest rates aren't, you know, aren't as low as what you'd get off for a car loan, for example, you know, they still make a lot of sense in the mining space.

So I think they're going to be gone pretty soon. They have to be those s9s.

John Quigley: Okay.


ahead, Igor.

[00:12:00] Igor Runets: Yeah, I would agree with that. It depends on availability of a capital for minors. Obviously, a, some miners who bought machines and [unintelligible to keep running it their s9s for as long as possible.

but like, all the miners like a larger professional, my more professional minors who were able to replace the machines as [unintelligible]. We'll do a right after halving. I agree with this.

John Quigley: Okay, Jan. you also agreed that we'll see an upgrade from, from all miners who can stay in the business?

Jan Cepek: Well, I'm maybe I didn't express myself, precisely, but this is what we are hearing from the big farms.

They're essentially getting ready for a replacement of the hardware. So they usually place their orders already. and they're just waiting for the shipments. And once these miners are installed, I mean, Then what could happen to you would either [00:13:00] have a rise in difficulty that would obsolete the old hardware even faster, or if it stays constant, that means that the farms that did have too high, you know, electricity costs and operational costs, that could not operate s9s, would just go out of business.

Adam Traidman: Hey actually, Thomas, do you mind if I, I entered a question into Igor and Jan, if you don't mind, I'm really curious because I think they're right. That a lot of people have placed orders. I'm curious what everybody thinks about the recent instability at Bitmain. And their ability to fulfill the orders that people have placed over the last several months.

I assume everyone saw that there was a fight at the Bureau in Beijing the other day, which is caught on Twitter. And, the business license was, for a short period of time stolen by, amazingly the chief legal officer for Bitmain. And then he was thrown in jail by the police. So I'm very curious what Jan and Igor think about that.

Their capability of delivering on their contracts.

[00:14:00] Igor Runets: It's a good, it's a great question. And we saw like a lot of hesitation from when you have our large clients to place orders, who is at Bitmain, even, even before this, we definitely see a kind of a Bitmain's share, a significantly decrease in recent year.

Jan Cepek: I cannot speak to like specific things because we, we're not like directly connected to the mining hardware, but from what we are hearing for the miners, they try to diversify any potential risks. So they usually that's at least what they say. They try to diversify their orders between Whatsminer and a Bitmain and see what comes first, pretty much.

What we also heard is that it is not quite sure. Like if the, for example, the s nines are going to be [00:15:00] delivered on time and, in enough quantity required by the customers. So the mass production could be delayed because of all the coronavirus things, some stuff like that. So maybe, maybe Adam could do have

input on this as well.

Adam Traidman: Jan.  you said S nine. Do you mean S 19?

Jan Cepek: Sorry. I meant s19s.

Adam Traidman: Ah, yeah, I, yeah, I, I have a lot of visibility into that. Actually. I think the, the ODMs in China. we're not terribly delayed. I mean, in our case, for example, you know, a couple of weeks’ time, nothing significant TSMC, basically made up for all issues and there are not significant delays there at all.

So w we haven't really seen any, any issues out of, wafer or chip manufacturing, that said, I think that, because of the slowdown, just in general, consumer purchasing of, electronics, you're going to see kind of a, you still see over capacity in TSMC now, but in Q three Q four, you're going to see that come down a little bit, [00:16:00] because right now the orders have been reduced because of Corona in Q2.

Right. And so that manifests itself in Q three and Q four as the chip manufacturing level and the ODM level. And then it will come back with a vengeance. Right at the end of the year. So we're going to continue to see a difficult lack of supply. I think right now, with all the old hardware in the mining space, slated for replacement, I guess Igor's customers are smart.

They place their orders late last year. Not everybody's as smart as them. A lot of folks are doing that now, or post halvening. A lot of folks are hesitant because, if you saw Canaan's public, publicly reported, results, to the NASDAQ, after they went public, just a few weeks ago, they are projecting some pretty scary numbers going into the next quarter. Bitmain’s, you know, obviously having a lot of legal trouble and, hope for all the best with Whatsminer guys.

But I think we all know that there's some legal issues there as well. So there there's a lot of instability in the industry right now. And that makes it really difficult for large mining companies to just send tens of millions of dollars to these folks, you [00:17:00] know, and, and, and hope for the best. So it's going to be really interesting to see what happens.

it's I I'm thankful for it. It resulted in us selling out of our first-generation systems way earlier than we expected, due to the instability with the other players. but I think that if you look back at the history of like computers and microprocessors, where Intel became of the monopoly after competing with like AMD and Cyrix and all that in the eighties and nineties, I actually think we're going to see the same thing in the mining space.

I think we're going to see the same sort of trend. I think there's going to be one player in the major chip area, and then there'll be a lot of system companies or a few system companies just like you can buy computers from Dell and many other companies today. So I think it's just going to get too costly.

And then your cashflow situation and the margins are too low. For too many companies to try to all make their own chips. So we'll see if that comes through or not.

John Quigley: When do you think we'll arrive at that kind of a system where we have one dominant player?

Adam Traidman: Yeah, in order, I mean, think about how long it was before Intel [00:18:00] finally killed AMD and Cyrix.

It took like a fricking decade. But I think in this case, it's got to go quicker. You know why? Because these other companies are just not that financially strong and, and look at the like crazy levels of governance, instability, and police action. Right? These companies, I mean, it rivals like a movie and somebody's got to make a movie about this by the way they totally should.

So I would say it's going to be a lot faster. I would say definitely within two or three years.

John Quigley: For sure. It's like a, like a Hollywood movie. MicroBT seems to be putting, a lot of pressure on the Bitmain market share. I see that they're getting support of some companies. Do you think by the 2024, having, we might have a new leader in the market?

Adam Traidman: Well, I hope that leader is me. I have a lot of respect for dr.

Young, at MicroBT as well. And I think they're doing a great job and, you know, we work with them as well. So, you know, we've actually worked with the guys we worked with Jihan, we worked with a Nangeng at Canaan, too. So [00:19:00] we're really at SBI. We're very much into competition. Right? Like we love to create competing situations because we think that competition.

Breeds great products and grows industries. So, you know, it's a very, very small world that we all live in. but I, I very much hope that we can bring some legitimacy and, you know, sort of stability to the industry because I'll be damned. We haven't really seen it thus far have we?


John Quigley: Igor. I wanted to bring it back to you because I know bit river have operations both in China and in Russia.

And can you touch on some of the differences mining in these two regions?

Igor Runets: So first of all, I will correct us this a little bit. So we mined both in Russia and in China currently all our colocation data center facilities allocated in Russia, not in China, but we have a lot of experience both in Russia and China.

so, so some pros and cons of [00:20:00] each of those countries. So Russia. Why we decided to go with Russia for our data center business. It's a lot of cheap, sustainable excessive capacity here. So there a lot of huge hydropower plants were built in Siberia, during like 60s or 70s. And now we have access to a lot of our sustainable green hydro power capacity [unintelligible] and has, this is definitely a plus. And actually kind of unique feature of that hydropower [unintelligible] that it's stable.

We don't have huge fluctuations between the summer season, a winter season. So, the water stream in Siberia is very stable and, because there are no huge fluctuations. So sustainability or stability inexpensive power is kind of [00:21:00] good to feature of Siberia. Another feature is a cold climate. We save a lot of money on the cooling or the devices and.

After that, very supportive and, predictable, regulatory situation with federal government with local government, I believe as also has three main pros of mining in Russian. Mining in China is kind of slightly different. people are kind of more focused on a short, shorter term, not long as a long term stability as that's why we see kind of people acting opportunistically, moving devices from, for example, Sichuan to other places and to moving back to Sichuan and for summer, it's kind of a very unique characteristic where a unique feature of China, but a very big advantage of money and in China, it's kind of proximity to all the manufacturers, proximity to supply [00:22:00] chain.

sometimes people I've been able to build mines and, deliver devices to mines much faster in the China is a significantly faster in China. So as there are some pros and cons, I believe, for us as I am. Who focused on kind of a long stability, and actually is a little bit more attractive for, for data center play, because we have access to very predictable, low cost, sustainable electricity, good, predictable, regulation.

And, our clients are ready to wait for an extra week or so. Okay. Before the device has been delivered.

John Quigley: Okay. Seems like, yeah, it seems like Russia is, has a lot of attractive clients as a mining region and it will be interesting to see how the hash rate share changes over the next four years. We've got to wrap up soon.


Igor Runets: We're

working on this to change the distribution of hash power.

[00:23:00] John Quigley: Yeah, let's see. Let's see how it shapes up. before we wrap up. Another, big thing in mining, apart from getting low electric rates and good capital deals, is obviously the firmware you use to run the device and how you can boost your hash rate capacity, increase the efficiency of your hardware.

And yeah. Brains have been doing some extraordinary work in this area. Do you believe this is going to be a huge, angle for minors to stay competitive going forward? Using firmware on the latest generations of hardware.

Jan Cepek: Oh, well, what we're seeing is that there is a lot of, firmware projects that, have come to the market that try to optimize the settings of individual mining chips on the inside of the machine.

And this is also the path that we also took with brains OS where we still do have the open source community edition, and then we have a brains OS plus a that we [00:24:00] have as a commercial product that. Does all these tuning algorithms currently supporting the  S9s we're bringing the S17 support, but actually there's another interesting aspect of, or that we sort of like, we didn't anticipate that this could happen, but there were a few people that kind of pointed that.

we have quite a few firmwares available for, for different machines in the market, but they all built on top of CG minor. As soon as they, like extend the functionality. they seem to be like breaking the GPL licensing because. they don't give the users the IP behind, you know, the tuning, even though it's part of CG minor code base, which is GPL.

So technically they, any customer that takes the firmware should, should receive a copy of the sources and so on. And this goes down also to the manufacturers who build their products. On top of open source [00:25:00] software components, not just speaking of CG minor, but we have, we have, you know, Linux kernel, the Linux distribution and all these things, and they keep all that stuff closed.

so there may be some, some shift in this area, and we try to provide solutions for this because that's, that's what Braiins OS is about.

John Quigley: Yeah,

I know. There's been a lot of. stuff going on in the firmware industry recently. Yeah. Unfortunately we don't have time for that. It'll have to be a discussion for another day. so we've got to wrap up.

Thank you, gentlemen. You're all doing an amazing job in the mining industry and happy half