Have you wondered how miners can improve profitability both short-term and long-term? Nick Hansen is the CEO of Luxor Technology, the largest North American mining pool for many altcoins and most recently, Bitcoin.
Nick discusses the economics, decentralization and the viability of pooled mining. Nick, at the helm of Luxor, also developed the platform used to facilitate the first legal Security Token Offering (STO) in the United States.
Listen on iTunes
Transcription
Scott Offord: Nick, how are you doing this week?
Nick Hansen: Great. How are you doing?
Scott Offord: Yeah. So for our audience, Nick is with Luxor, it’s a pool, and it’s fairly new. So Nick you were telling us that you just launched recently, this year?
Nick Hansen: Yeah. So we actually launched, the business was formed two years ago, and then we started with a SiaCoin mining pool. We’ve been mining Sia coin for about 18 months now, and are the largest Sia coin mining pool. And since then we started finding other Altcoins that we were really interested in, Liberty, or Decred, Monero, Z-Cash, high quality projects we think have a lot of technical reasons why they should… They have good fundamentals.
Nick Hansen: And we started looking for coins that were underserved by the pools that were in the ecosystem at that time. And we started building these pools, and that’s where we’re at now. We mine 11 different Altcoins, then we just built our Bit Coin pool in October which we launched in when we went to Chengdu for the conference there.
Scott Offord: That’s awesome.
Ethan Zurka: I’ve got a quick question for you, Nick. How did you learn that it was pronounced the Sia coin?
Nick Hansen: So I actually met the founder of Sia, David Vorick, he’s the CEO of Nebulous labs, and he told me that it was Sia. I always said Sia coin for the longest time.
Ethan Zurka: That’s exactly what I’ve said. And I’ve gotten in arguments with people about it, so I wanted to know how [crosstalk 00:01:40]
Nick Hansen: It is Sia, but almost everyone I’ve seen that hasn’t met David calls it Sia. Yep.
Ethan Zurka: Yeah. Yeah.
Nick Hansen: And I don’t… Yeah. Because there’s that singer, too, named Sia.
Scott Offord: Yeah. So Nick, you were saying that for the most part it’s Altcoins, and you’re just getting into SHA256, or?
Nick Hansen: Yeah. So we just got into SHA256 with Bit Coin in October. And that’s been going really well, we found two blocks which is pretty great. We’re PPS, which is a little bit interesting. I know we may get into that some point on the Pod, the difference between payment mechanisms. But we firmly believe in PSS, all of our pools are PPS. We experimented with different payment schemes early on, and just found that it only makes sense. I mean, I mine personally, our Lexor mines personally, as a business as well, and we’ll only mine at pools that pay PPS. And we can get into why that is.
Scott Offord: Yeah. Well, let’s talk about that now. Why is that?
Nick Hansen: Sure. Yeah, so with PPS you can actually just, if you can measure-
Ethan Zurka: I’ve got a quick question. What does PPS stand for just for listeners that are newbs out there that don’t know.
Nick Hansen: That’s a great question. So PPS stands for pay-per-share. I share is what your miner is actually giving to the pool to collect the reward. PPLNS means pay-per-last-end-shares, which is a way of essentially building a proportional pool. So everybody submits all of their shares to this pool. If you submitted 30% of the shares, you get 30% of the rewards when a block is found.
Nick Hansen: PPS, mathematically we can determine how much coin you should get for every share that you submit, because we know based on the randomness of a hash function how many hashes it will take to solve a block on average. So then we give you some portion of that block reward every time you submit a share. Now, the reason I always will mine PPS is because when you submit a share, you know exactly how much that share is worth. You can do the same exact calculation that the pool is doing, and you can determine, it’s the only way you can determine, if the pool is skimming.
Nick Hansen: Like with PPLNS, the pool could inject some shares into that pool, that fund of shares that I was talking about, and you would never know. So they would be injecting a few shares here and there, and then essentially skimming without you ever knowing, and it would be [crosstalk 00:04:04]
Scott Offord: So that’s possibly one way to be a little bit more transparent as a pool, is to do it the way that you guys are doing it.
Nick Hansen: Yeah. Absolutely. So we think PPS, it’s obviously, it’s more risky for the pool. We’ve been attacked before. So for example, the way that a pool can be attacked is a nefarious miner would, they would be getting their PPS, they would be getting their rewards from the pool, but instead of actually submitting… That one time they actually find a block, instead of submitting the block, they discard it. It’s called a block discarding attack, or a block withholding attack. And typically it’s something that pools will do to other pools as a way to essentially defund them.
Nick Hansen: And it happened very, actually very recently with Slush, they were being block withheld. And they claimed that’s why their luck was so bad for the summer.
Ethan Zurka: Yeah. I wondered about that. I mean, their luck went phenomenally down. And so you can confirm that that was actually an attack that was done on them?
Nick Hansen: Well, I can’t confirm that. But I can say, I mean I-
Ethan Zurka: You can speculate?
Nick Hansen: You can speculate, right. That’s most likely what happened was there’s another pool, or some entity out there, that wants to defund them, and that’s how they can do it. So with PPS, now with PPLNS, that attack is now democratized across all miners, but with PPS the pool is funding that. So you’ll find a miner that has what we call a luck factor. The luck factor is we should know it should take approximately 100 shares to find a block, or whatever that number is. Sometimes it takes more, sometimes it takes less, but over time it should take 100.
Nick Hansen: If you’ve submitted 200 shares without finding a block, that means we’ve paid you two blocks worth of rewards without you submitting a block, and when you start getting into like 10X luck, like 1000 or so, then we’re really wondering, “What the heck?” Yeah, so that’s PPS versus PPLNS, some of the ways you can exploit them. And there’s a million payment methods. There’s Kano, I think, probably has the best PPLNS implementation, they have an entire week lookback period with a really long ramp on period.
Nick Hansen: I’m trying to think of some of the other ones off the top of my head [crosstalk 00:06:28]
Scott Offord: So that way you’re not getting as instantaneous payout, or?
Nick Hansen: Yeah. So with them, so PPLNS is it can be exploited by jumping onto the pool right after they find block, because then you’re more likely to have a large portion of the next block award. It’s called pool hopping, which you can read about it on Bit Coin talk and stuff. There’s a lot of people who talk about pool hopping, how to avoid it, is it actually bad? Some people would say yes, some people say no.
Nick Hansen: PPS, it’s not possible. Or you can pool hop all you want, you’re going to submit the shares that you submit and get paid for them. So, yeah.
Ethan Zurka: I’m curious, how often or how are you guys running the math on all of this? Do you guys have your own special software that’s doing it automatically for you? Or did you have to build that software and create it?
Nick Hansen: Yeah. We wrote it all. We wrote it all. The actual math is very simple. It’s just the number of shares, divided by the network difficulty, times the coin base reward. So 12.5 Bit Coin, that’s how much a share is worth. Then every time you submit a share, we just increment your balance by .001 [sutoshi 00:07:50], or whatever that number is today, and we just do that every time you submit a share. And that’s how it works.
Ethan Zurka: Can you tell us your thoughts about the halving that’s coming up, what impact do you think it’s going to have on the industry, at least from your perspective?
Nick Hansen: Yeah. Certainly. I don’t know if you know this, but everybody’s rewards are going to go down by half.
Ethan Zurka: Right.
Scott Offord: For Bit Coin [crosstalk 00:08:16]
Nick Hansen: Yeah. For Bit Coin, everybody’s rewards will go down by half. What does that mean? That means if you don’t have the best machines, or the best rates, you’re going to be mining at a loss. And I mean, we’re going to start seeing machines move from high energy cost areas, we’ve seen the exodus of S9s to most likely Kazakhstan, and other places with sub one set power.
Ethan Zurka: Venezuela.
Nick Hansen: Venezuela, and some other places in South America. Those S9s are all going to be gone, right? You won’t be mining with those at all. Versions of the S17, and the M20, and M20S, and now the M30, most likely I think those will be the only machines that will actually be able to mine profitably anymore. So what that means is if you have an S9, you should be selling it.
Ethan Zurka: Yeah.
Scott Offord: There’s a lot of them on the market right now.
Ethan Zurka: Yeah.
Nick Hansen: Yeah.
Scott Offord: So one thing that I was wondering, the block reward and all that technical stuff behind it isn’t necessarily my strong suit, but so yeah, the BTC, Bit Coin Halving is coming, but what impact would that have on Bit Coin cash or Bit Coin SV? What are their halvings, and what might happen there?
Nick Hansen: I don’t follow their halvings very closely. A lot of the hash rate, we may see a little bit of dissemination from BTC, obviously the hash rate is almost solely focused on BTC, except for whenever one of those other coins pumps a little bit, and then [crosstalk 00:09:59]
Scott Offord: It depends if there’s a war at any particular time.
Nick Hansen: Yeah. Yeah. And then people jump over to the other one for a little while, but most likely you’re not interested in mining at a loss. Now when the value of a Bit Coin share goes in half, after the halvening, then we may see some hash rate jump over to the other coins for a little while until their difficulty, as we all know, the difficult will increase, and we’ll find some equilibrium.
Nick Hansen: I imagine equilibrium will take a couple of months, though. So three to six months, there’s going to be a lot of churn in hash rate. And [crosstalk 00:10:33] we’re specifically targeting that.
Scott Offord: Do you think that there’s going to be some kind of a coordinated attack, or aggressive strategy, by some of these other SHA256 coins around the halvening?
Nick Hansen: I don’t think they have enough… I doubt it. It doesn’t seem like they have enough, they wouldn’t have enough hash rate to really do anything meaningful in any way. So-
Scott Offord: [crosstalk 00:10:56] But there’s proof of work, there’s proof of social, you hear people using those terms. But when Bit Coin SV first started, they were getting really excited and they though they were going to be winning the has rate war, but yeah [crosstalk 00:11:12]
Nick Hansen: [crosstalk 00:11:13]
Scott Offord: Yeah. How that might change, but…
Nick Hansen: If you’re a Bit Coiner, I don’t think you have anything to worry about.
Scott Offord: So other than SHA256, for you personally, what’s your favorite algorithm or coin?
Nick Hansen: Decred. [crosstalk 00:11:32] to be. I really like Decred, I know their DEVs pretty well, and the quality of their work is second to none. They have some of the highest quality code that I’ve seen. They’re lead, Dave, is phenomenal. He’s amazing. So probably one of the smartest people I’ve ever interacted with. So that’s definitely my favorite project from a technical standpoint.
Nick Hansen: Again, I really Sia, too. What they’re actually doing is cool. The ability to store files not technically on the block chain, but verify that they are being stored by a block chain, that’s really amazing as well. Those are the two projects that come to mind straight away.
Ethan Zurka: Can you tell me more specifically about the code that Decred, like just give me a comparison. Like what did they do that’s just really wowed you compared to another development?
Nick Hansen: They’ve been able to optimize this hash function that they have. Not specifically the hash function, but the way they apply it, and was able to speed up their sync time by like 90%. So you can now sync the Decred block chain in like less than an hour, which is incredible. So that’s amazing. That’s a pretty amazing feat. A lot of chains have this now, but it was really cool to see Dave able to pull that off.
Nick Hansen: And other people, sorry. Dave is the one I keep talking about, but there’s a lot of Decred developers. And I want to give them credit as well.
Scott Offord: Dave. Dave who?
Nick Hansen: Collins, I believe his last name is.
Scott Offord: Okay.
Nick Hansen: I know him as Dave SGH, which is screen name, but…
Scott Offord: So what’s next for Luxor?
Nick Hansen: Yeah. So our next big push is going to have an equi-hash profit switching pool. So we’re going to move, we’re going to do that as a stepping stone to SHA256. So we’ll do a SHA256 profit switching pool in Q2 of 2020. The equi-hash pool is currently running right now, but the profit switching pool is currently running now, and we’ll start rolling that out to public customers in January, most likely.
Scott Offord: So tell us a little bit about profit switching.
Nick Hansen: Yeah. That could be a Podcast on its own. Basically what we are going to do, is you just look at the network difficulty, you look at the trends in network difficulty, and then the trends in price, and see if you can find some regression that makes sense, and just find the most profitable coin at that point in time to mine.
Nick Hansen: So we’ll have four coins underlying that pool. We’ll have Komodo, Zcash, Pirate, and Horizon, Zen Cash. We’ll be analyzing all four of those coins in real time and deciding which coin is the most profitable to mine at any point in time. And then switching the hash rate over to each of those chains to maximize profits. It’s a really common feature people have asked for. We get requests for profit switching a lot, especially from our equi-hash miners. They’re not behoofen to any chain at any point in time, they just want to like get their Op-Ex covered and then typically what they’ll do is they’ll mine their Op-Ex, and then switch to the coin they’re actually interested in, say Z-Cash or Zen.
Ethan Zurka: Are you going to be using some sort of artificial intelligence or something to determine this? Or is it just number crunching?
Nick Hansen: So it’s just number crunching now. But, like you said, that’s something we will definitely start doing is we’ll start looking at getting smarter about which coins we’re going to mine. At first it’ll be a rather naïve approach, just looking at difficult and price with a very small regression. But then we will start seeing if there’s a way we can aggregate more data and decide which coin is more profitable to switch to earlier. So what you would really like to do is know which coin is going to be the most profitable, not the coin that is the most profitable.
Ethan Zurka: That’s why I asked about AI, is I’ve actually heard in the community some people that are working on that technology right now as we’re speaking, of predicting where you need to be. It’s not what fits the bill right here right now, it’s where you need to be in next 24, 48 hours to be the most profitable.
Nick Hansen: Yeah. So the one thing that we have a bit of a benefit, being at the pool level, is that we’re very nimble to switch. A lot of people that are doing what you’re talking about are mining, they’re switching the miners themselves. Which is, that’s quite the task. I don’t know if you’ve ever been to a scaled mining farm, but switching a scaled mining farm between pools is quite a task.
Nick Hansen: But being at the pool level, they connect to the pool, and then the pool just decides which work to send out. And there’s no disconnects, the miners don’t even know that they’re mining different chains at any point in time. So we just send out different jobs.
Ethan Zurka: So that’s very interesting that you say that. Would it be possible for other pools to maybe side-mine? Like let’s say I’m connected to a pool, and I’m using just my S9 miners, but would it be possible for that pool to send a different coin’s, even though it’s the same algorithm, but it’s mining for different coin’s work, and then my mine would be working on behalf of that. Is that a…
Nick Hansen: So are you asking if they could be mining Bit Coin and Bit Coin Cash at the same time?
Ethan Zurka: Not at the same time. But let’s just say every now and then, periodically, they send a wave of Bit Coin Cash to be mined by my miners, or just the pool in general.
Nick Hansen: Absolutely. And I’m very confident in saying that Bit Coin pools do just that.
Ethan Zurka: Oh, wow.
Nick Hansen: They’ll be switching coins underneath the hood and you won’t even know. So they’ll give you the Bit Coin PPS rate, whatever that is at any point in time, but they’re most likely switching other coins underneath that.
Ethan Zurka: Wow. Okay. So I was always curious about that. So my miners may have not just only done Bit Coin in their entire history, they may have done Bit Coin Cash, and some other-
Nick Hansen: PPC, Terra coin, yeah.
Ethan Zurka: Wow. Okay. That’s-
Nick Hansen: Yeah. So at the pool level, we can do that. You won’t know. We can switch out whatever job we want to send, your miner won’t know. As long as it’s submitting shares, your hash rate and everything will… You’ll be getting paid, you’ll have hash rate, and you won’t ever know.
Scott Offord: Do you think there would be a way to opt out of a certain one? Like let’s say if you had something like morally against one coin, or something like that, one project. Could you click a check box and opt out?
Nick Hansen: Not technically. You could… We, as Luxor, we will, if we ever decide to start opening up to profit switching, we may consider that. If say… Maybe in equi-hash it’s not as important, because people are less dogmatic, but in Bit Coin, most likely people would uncheck that Bit Coin SV box.
Ethan Zurka: Yeah. So yeah, I was really curious how the hash rate was being kept up. And now it kind of makes sense to me that, like you were saying, that under the hood things can happen that are beyond our control, that the pool can control.
Nick Hansen: Absolutely. Yeah. And so Stratum V2 would change all of that. So when Slush, that’s one of the things that Slush… I don’t know if he actually called that out specifically, but you would not be able to do that. Because say… So what Stratum V2 allows you to do, essentially, is pack the transactions the way you want to as the miner. So I’m mining my SHA256 miners, and I’m sending the shares back to the pool. Well, along with that is which transactions I want to go in the block.
Nick Hansen: Well if you’re mining Bit Coin, the transactions that are in Bit Coin SV are not in the blocks, or not in the memory pool, so you wouldn’t be able to submit those blocks. You wouldn’t be able to submit those shares because those transactions are not in the memory pool.
Ethan Zurka: What are your thoughts on the morality of like, for myself, when I join a pool I think I’m mining a certain coin. And of course, to a degree, I am mining that coin because I’m getting paid back in that coin. But in reality, I’m actually supporting other coins. Some, like Scott mentioned, that I might not be in support of. So what are your thoughts on the morality of that?
Nick Hansen: Yeah. I wouldn’t say that it’s… That’s a good question. I’m not actually sure. I wouldn’t… Okay. I wouldn’t say that it’s, I’m not sure if it’s moral or immoral. Yeah, I wouldn’t do it. Let me put it that way. My company will not do that. So I guess in that sense, then yeah, I would say that it is immoral. I would, but that doesn’t that everybody would say that.
Nick Hansen: A lot of people are rather utilitarian, right? They just want to get the most money for their coin, or for their [crosstalk 00:21:07]. And then by proxy, to their coin. So yeah, that’s a good question. That’s a good question. I don’t know. I don’t have a very solid answer for you, which makes me a terrible Podcast guest.
Scott Offord: No, so on a different note, that dog, is that a Corgi?
Nick Hansen: That is a Corgi.
Scott Offord: Walking around, it’s kind of like [crosstalk 00:21:35]
Nick Hansen: He sees, yeah, he’s [crosstalk 00:21:37]
Scott Offord: [crosstalk 00:21:37] has his…
Nick Hansen: Yep.
Scott Offord: [crosstalk 00:21:38] right?
Nick Hansen: Yep. That’s Rogue. She’s a little Corgi, yep. I have a cat, too.
Scott Offord: Nice.
Nick Hansen: That’s Hermes.
Scott Offord: Hermes.
Nick Hansen: I’ll lock them up next time.
Scott Offord: No, No.
Ethan Zurka: No.
Scott Offord: Yeah, it’s a treat.
Scott Offord: So to change topics a little bit, it sounds like you know a little bit about decentralization, and economics. What other topics do you want to hit on today?
Nick Hansen: Yeah. So let’s talk… There’s two. But one I’ll give some tips to your miners on how they can be as profitable as possible for as long as possible, we’ll go on that one in a second. But we can talk about decentralization first. So right now you can see a lot of the hash rate is really centralizing, right? Not just to a pool, but to a region. Mostly the Sichuan province in China. Some people think there’s more than 60% of all Bit Coin hash rate is within an eight hour radius of Chengdu, China.
Nick Hansen: Now, is that a problem? I mean, it’s not great. But is there any way around it? I think the only way we’re going to get around it is with an American manufacturer, a mining manufacturer. So and I have heard there are a couple potentially coming, we’ll see. [Obilist 00:23:07], they tried, and failed rather miserably. But hopefully they can try again. I hope they can. Because I think they ended up failing, but in their failure they succeeded. They actually shipped a miner, an American miner, which is something we should kind of applaud.
Scott Offord: But they themselves haven’t really shown any interest in doing anything for SHA256, I don’t think.
Nick Hansen: Yeah, I don’t believe so. I have to imagine that they would, though. If I were running that business, that’s what they should have done in the first place, I believe.
Ethan Zurka: If you could advise an American manufacturer on what should they be designing for? So basically you’re fighting two different things. You’re fighting manufacturing costs, so go to a very, very small process, seven nanometer, five nanometer, even three nanometer, the foundries to produce those chips, it becomes phenomenally expensive. And then we have to balance that with power efficiency. Would you say it would be better for an American company to produce miners that are balanced more towards efficiency through a small chip process in the dye? Or don’t worry about the efficiency, just get the cost low so people can buy a cagillion of them and get them up and mining.
Nick Hansen: Yeah. I think long term, I think you have to focus on efficiency. The watts per jewel, I think that’s the bottom line number. And anybody that’s making a miner needs to be thinking about that.
Scott Offord: Yeah. Jewels per terra hash.
Ethan Zurka: Yeah. Jewels per terra hash is very important. And that being-
Nick Hansen: [crosstalk 00:24:57] M30 just set the mark, right? At 33?
Scott Offord: Yes. Supposedly.
Nick Hansen: Yeah. We’ll see.
Scott Offord: 30. Well, they said sub 40 Jewels per Terra hash, but what that’s actually going to be, who knows.
Nick Hansen: Right.
Scott Offord: [crosstalk 00:25:12] If that’s actually going to come out [crosstalk 00:25:15]
Ethan Zurka: Scott and I both have seen many miners that you look at the specs, and then when they run real time in the farm, and even the farms have told us this, that they can be all the over the place. You know?
Nick Hansen: Yeah. Other advice, yeah, I think they have to go for SHA256. That’s the only way to do it. And it needs to be, specifically if America wants to really be involved, like really in the conversation, we have to figure out how to either manufacture them in the United States, or through a US entity, such that you don’t have to pay the tariffs. Because the tariffs right now are really the killer, right? It makes it essentially impossible for anybody to mine at scale. People will buy 10 or 15 miners and just pay the 25% out of principle, but if you have somebody buying 1000 plus S17s, that becomes rather untenable with a 25% tax on that, or even more.
Scott Offord: Yeah. Yeah. And right now, Bit Main actually does have a manufacturer, a facility in Malaysia, which is starting to help people in the United States to avoid that tariff. But other than that, all the other manufacturers right now are based in China, so…
Nick Hansen: Right. Let’s see, so decentralization. Yeah, we’re headed in the wrong direction currently. But that doesn’t mean we can’t turn around. And we’ll figure it out.
Ethan Zurka: Nick, what are your thoughts on maybe these companies, because we all know that China is kind of an unstable country, and sometimes it’s for block chain, and then other times it’s against it, and there are claims that Bit Coin has been outlawed there. What are your thoughts on those Chinese companies migrating to other countries like the United States, and Canada, and South America, if you will.
Nick Hansen: Yeah. You mean the companies, or the minors themselves?
Ethan Zurka: Well yeah, the entire mining facilities just uprooting. I’ve kind of been hearing a buzz that that’s kind of a possibility, that that’s being thought about.
Nick Hansen: So that’s really hard to do. It takes a lot of time. So you know that they’ve moved minors from the Sichuan province to different provinces during the [crosstalk 00:27:42]
Scott Offord: The half season.
Nick Hansen: Yeah. They actually do that currently, but within the country. Now would you move them out to another country and leave them there? I don’t know. That’s a pretty big ask, I think. I would imagine what’s going to happen, what will actually be happening, is that net new minors, or the new minors that are being built, that they wouldn’t end up in China.
Ethan Zurka: They would just shift them to the new location?
Nick Hansen: Yeah. I don’t think they’re going to take… This is all in a hypothetical world, because that’s not going to happen. Right now, all the miners are just going to end up in the Sichuan province in China, like 80 plus percent of them will probably just end up right there. Because it makes the most sense, right? You just put them on a truck, ship them over there, and plug them in. And they pay cheaper probably then we pay anywhere over here.
Nick Hansen: In an ideal world, you wouldn’t take those mines and move them somewhere. You would just create new mines with the machines coming out of the pipeline.
Ethan Zurka: That makes sense.
Nick Hansen: Yeah. Yeah, and the other topic was so how can miners, how can they maximize their profits. So I don’t know, this goes from the miner that just got their first GP rig set up, and is now mining Ethereum or something, to a miner that has thousands of machines in a data center mining Bit Coin with optimized firmwares and everything. The one thing that I notice people don’t watch as closely as I think they should, and they really need to keep an eye on it, is the pool efficiency. The efficiency of their miner’s connection to the pool.
Nick Hansen: So like when you submit a stale or an invalid, they’ll show up as hash rate, but you don’t get paid for them. And a lot of people don’t monitor that rate. So if you’re mining, say, let’s say you’re mining Monero, and you have a 97% efficiency, and you’re mining at a pool that takes two percent fee. Well, since you’re losing three percent to your poor efficiency, and now you have a two percent fee, you’re actually mining at five percent fee, effectively.
Nick Hansen: So you can squeeze out three more percent if you can find a pool that has either better latency, processes shares faster, gets block updates out quicker, there’s a lot of things that pools can do to reduce that inefficiency. And that’s something I think a lot of miners should really take a hard look at, is see if their miners are running efficiently.
Ethan Zurka: [crosstalk 00:30:23] a lot of sense, yeah.
Scott Offord: Ethan, I think that’s something that we might be able to help people to start looking at, as well, with our website cryptomining.tools. Right now we have an uptime check box where you can say, where you can assume that your miners are always going to be up 100 percent of the time, but realistically that’s not the case.
Scott Offord: But I think another field that we can also add is something like efficiency, or stale shares, because [crosstalk 00:30:52] knocking off two to three percent right there. That’s big.
Ethan Zurka: Yeah. That’s significant. I’ll definitely bring it up with the developers. I’m sure they’ll be happy to do more math.
Scott Offord: And Nick, how do you help miners to see that and to pay more closer attention it?
Nick Hansen: Yeah. So we display all of that data on our site. So we collect those shares, we still collect them as statistics. We obviously can’t pay invalid shares, or stales, but we collect them and display them. So when you submit… Really you should have zero invalids. Unless your miner came on and submitted a weird share right when it started, you should have zero invalids. Start with that.
Nick Hansen: And then you should have very low stales. Like .1% or lower stale rate. So those are some bench marks you should be looking for. We have our miners that are very close to our pools, and we have like .05% stale rate. So that’s a good bench mark. Now if you’re getting like one to two percent stales, you got to find a better pool. You got to find a pool that’s closer to you.
Scott Offord: And it has more to do with that physical geography that you’re looking [crosstalk 00:32:06]
Ethan Zurka: [crosstalk 00:32:06] network lag, the latency.
Nick Hansen: So it can be that, or it could be the pool is processing shares slowly, and is taking a long time to get a new jobs out to miners. We time all that stuff internally to make sure that we’re sending out jobs with very low latency, meaning when I say a new job, say the block height has just changed, we make sure that we get that information from the network, and get that information out to the miners in less than a couple milliseconds. That’s what we’re trying to do. So-
Scott Offord: Okay. So some of it has to do with the pool itself, some of it has to do with the mining facility, and maybe how robust our network is, or [crosstalk 00:32:47]
Nick Hansen: There’s a million reasons why you could have a high stale rate. But just identifying that you have a high stale rate, that’s the first step. Because then you can start investigating. We’ve had miners that had a misconfigured data center, where it’s introducing latency within the data centers. So they were measuring latency from the data center to the pool, and we’re looking at a really good latency there, but it was actually latency coming from within the data center.
Nick Hansen: Oh, another good number to add is you your ping to the pool should be less 100 milliseconds. You should be looking at less than 100 milliseconds latency between your rigs and the pool. And that needs to be your rigs, not your network.
Scott Offord: And is that something that you can help show back to them? Or is that something that the miner has to look up themselves?
Nick Hansen: The miner would have to look that up themselves. I’m trying to figure out a way to do it. There’s no way in the protocol for us to actually reach out to the miner and get them to respond so we can measure that time. It’s something I’m trying to do, but right not there’s no way for me to do it. So you would just have to measure that from your miners. So make sure you have a machine that’s on the same network as your miner, and then run a ping test or something to the pool that you’re using, and make sure that that ping rate is less than 100 milliseconds. And if it’s not, there’s some ways to mitigate a higher latency. We can get into that, as well. You can compress the shares with a proxy or something, but that [crosstalk 00:34:22] technical thing you may not want [crosstalk 00:34:24]
Scott Offord: Yeah. Well we’re coming up to the end of the show, here. So why don’t you just tell us one more time where we could find you, what algorithms people should use, and [crosstalk 00:34:38]
Nick Hansen: Luxor technology, you can find at Luxor.tech. You can find us on Discord, hit me up on Twitter, NickHanson600. We’re pretty much around, we have a great Discord channel, we chat with people all the time. Help them set up, help them de bug, anything? No? Okay. And that’s it. Yeah. Thanks for having me on guys, appreciate it.