They paid a few extra thousands just to get their gears hosted by Compass, while the power rate was tempting in Russia, I am sure the price difference would pay the power bill at least for the first year (it was anywhere from 2k to 5k) depending on the gear price at that time, but well, sadly newbies have to pay the price twice for not doing enough research.


I can't accuse them of anything without solid proof, however, I can speculate as well. Another point is the fact that they could have delayed responding to the Sanctions, I mean come on! it's not like the U.S police were pointing a gun at Compass's owners telling them to end the deal with the Russian host right now, EU sanctioned everyone in Russia (most of them don't even matter) and they shut their eyes on the Gas companies because well, you can't hurt your own people this way, Compass could have saved those newbies if they wanted, the law isn't black and white, there is always a way to reduce damage if there is a will, but it could be that they intentionally used this excuse to force the clients to sell their gears for cheap, for all we know, the gears could still be mining to Compass "new address" after they bought them 'back' for dust, but we don't know, do we?



That's also a good possibility, Compass is known for having long delays.

What's even worse, many people paid for future orders, not sure how Compass is going to handle that, imagine selling a gear you never saw hashing for 4k or less after paying 10k or more for it, that would be more painful than having had the miner run for a few months at least.

How did you perform the reset? The safest way would be using an Sdcard to flash different firmware, if that does not fix it -- you will need another control board.

I can't really read the kernel log of Braiin OS but it looks like there is a missing file according to this


Compare this line against the other miner, if it's not there, then chances are you need to flash a new firmware, also there is a dedicated thread for Braiins OS in the software section you could post the kernel log there they will probably read it better.

While the first part of this post is correct, the second one isn't accurate, the password in Cgminer is irrelevant to almost all mining pools, (x) or (blank) works just fine with nearly almost all pools, the password you use to login to the pool's portal has nothing to do with the password you set on your miner's configuration.

Someone who works with an official MicroBT distributor sent me a nice list that I wouldn't find otherwise.


This list can also be found on whatsminer's website > https://www.whatsminer.com/src/views/afterSalesPointInquiry.html

I think some of those guys might also fix Bitmain gears, we know at least (HMTech) which is listed here already does that, not sure about the rest.

Usually, your miners will be pointed to your own pool, the host doesn't own a pool nor is responsible for that part, you provide them your main pool and two back up pools and they will set the miners up for you, depending on how well built the facility is, this could be done via basic methods like an email or an online portal which you use to set up your pools and then they use some sort of a script that would copy-paste your workers from that database to your miners, some hosts could give you direct access to your miner using some of the technologies available. Different hosts will have different ways of dealing with that, but the common thing is that every owner will decide the pool they want for their miners.

That would have a higher probability if the miner was recently moved/shipped, but when the hash rate suddenly drops (like how OP reported) then 90% change the board is dead and there is nothing that can be done about it, at least not as far as switching cables and whatnot, but of course, it's always worth the try.

Another thing that can cause this issue is a bad PSU, sometimes the PSU just can't handle 3 hash boards, an easy to troubleshoot both the wiring and the PSU in the same step would be to disconnect 1 working boards wires (PSU + data ribbon cables) and attaching them to the bad board.

If that fails, you might get lucky with custom firmware, use high voltage and low frequency, if the board shows up, then slowly increased frequency and decrease voltage until you get to the point where the board goes missing again.

If all of the above fails, simply disconnect the cables from the bad hash board and leave it inside the miner.

First thing first, you should assume that the majority of those services are scams, so don't send your gears until you have done your research, however, the legal ones operate as follows.

They provide the infrastructure like internet, electricity, cooling, and all that's needed for mining gears to run, and then they will make you sign a contract to pay x amount per KwH your miner uses, an example:

Your miner is 1*S19 pro, it consumes 3kWh, and you pay them 10 cents per kWh, so you pay them 0.1*3*24 = 7.2$ daily, of course, they won't collect the fees daily, maybe monthly, maybe yearly, it all depends on their terms and conditions.

These folks supposedly pay less to the power company, so maybe 0.5 cents per kWh, so for every S19 they host, they make 3.6$ daily minus all the operation fees they have to handle, you as a client get your miner up and running by them, pointed to your pool, you make money, they make money, and everyone is happy until someone is not.

Compass on the other hand offers a sort of a modified version of the basic hosting, while the majority of other hosts will ask you to send your mining gears to them, with Compass you will have to buy the gear from them, by selling you an "overpriced" mining gear, they make some good money, and then again, more money for the hosting services.

There are many things that can go wrong with this business model, most hosts won't let you see your miners in person, they can just tell you "your miner is dead" and need fixing, send us $300 in fees, and if you decide to take your miner, they might ask you for another $300 for uninstallation and packing fees, it's doesn't matter how good the host is, they will have the upper hand in this mess.

With that said, there could be and probably are some honest and good hosts, and to get a good power rate, sometimes your place is not an option, so this hosting stuff might be your only option if you really want to mine, but then, it comes with an added risk.

A few days ago, one of the largest host providers 'Compass Mining' was hit by U.S sanctions against Russia (this is not a political discussion, I will report any off-topic).

Compas is a U.S based company that sells mining gear with hosting services included, thousands of people overpaid (around 13k) for a miner, when the war started, they sent out emails telling their clients not to worry and don't rush to move their miners and that they "Compass" can handle the situation if needs handling.

A few days ago, they sent out emails telling their clients that the new sanctions require them to stop dealing with the Russian host and the clients have two options.

1-Pass ownership for their miners to Compass mining so that they can liquidate the gears in Russia on their behalf.
2-Contact the Russians directly and deal with your own problem (not recommended by them because of a high risk of losing the gears).

Today, someone reported that they will get 4.5k for the gear he bought for 13k (probably an S19 or M30) although that isn't likely because s19s brand new go for 7k range, these are used miners that MUST sell fast, and where? in a heavily sanctioned country, so only some Russian whale may buy them, so my guess is, not even 4.5k is possible.

The moral of the story: having someone else host your gear is very risky even if the company is based in your country and you think they can be highly trusted, it's not about Russia, it could be anywhere else, you can't control the outcome, you may never know when you might be forced to pass ownership of your miners and hope the host can sell your mining gears for whatever they can, and you won't be able to reject the offer regardless of how much your gear is actually worth.

The was a huge one, I believe this has to do with the previous drop of - 1.26% (some hashrate went offline for a while for some reason), If I am correct, we should not see 5+ anymore, it's just way too much hashrate IMO, 3-4 should be somehow the limit as long as we don't get a negative.

You can keep guessing or you can save yourself the trouble and post the kernel log, 99% of the time it will tell you exactly where the issue is, a fan loop like the one you described could be caused by many things, you have to get the clues from the kernel log.

If you decide to post the kernel log, please use the code tag.

Or, one last bear trap before heading back to at least the ATH



The way I see it is that fear is high, many people are waiting for a crash, longs are not increasing by much, more shorts coming in trying to make a killing over the coming trend crash, to me this is a perfect setup for a new rally to the upside, even if this trend line gets taken, it's more likely to be a bear trap and the last shakeout before heading back to 60k at least, this whole thing feels like Q1 2019.

It sure does, however, there is a downside to it which is the cost per hash,  an S9 will most likely last a lot longer running underclocked at 9.5th instead of the default 13 to 14.5th, but then assuming the miner costs $400 shipped, it becomes 28.5$ per th vs 42$ per th, you do save on power but the most often the lost hash rate surpasses all that.


Buying different batches might be a good idea, but not in the way you mentioned, batches come out in quantities of thousands, usually monthly, so buying from different people does not mean you will get a different batch, they could all have bought that one bad batch you were trying to avoid.

Usually, the first a few batches are the worst, when a few months have passed and the manufacturer starts getting a ton of warranty repair tickets,  they "might" fix what seems to be the issue, but then missing the first batch might also mean paying a lot more for the same miner, it's pretty hard to time this, luck is a major factor.


It's both, for an example antmine S9 are known to have a dead middle board due to a lack of airflow around the first a few chips facing the front side caused by the fan motor itself, so the longer it runs the likely to fail, so this gives you a linear curve more or less.

With something like 17 series where very often the issue is just one bad solder ball shorting the shit out of the board, it could be safer to buy a used miner which has been repaired, if your 17 series miner makes it through the first 6 months, it's more likely to survive another 6 months, but then again, different batches will have different results, the ones that later came from Malaysia were a lot better, it seems like the robots/people that did the soldering were doing it in a better way.

I also remember buying an early batch of S9k where I got miners without heatsinks attached, the heatsinks were sitting in the case doing nothing, it seems like they forgot to heat the thermal , on the other hand, I recall Phill got a different batch that arrived in good shape and worked for a while (not sure about the current status).

So ya, it's just a mess, very hard to predict and it's part of the risk you have to take when you are looking at a business than can get you a 1-2 year ROI, I highly doubt that even drug dealers have such a short projected ROI, , this is why when you invest in mining, you can win big just in the same way you can lose big, it's a very risky business, especially for those who are just starting, I always tell the majority of people NOT to invest in mining, at least not before knowing the full story, because things look pretty damn good on paper, you go to an online calculator and think to yourself that you will get your money back in 6 months, reinvest and double your hashrate in 3 months, get that many PH worths of hashrate and start making 10k a month while doing nothing, it's exactly how things look like in an excel sheet like the one you made, but that couldn't be further from the truth.

نعم يجب ان يكون التحويل صادر من ذلك العنوان, الارسال اليه لن يبين لك نوع العنوان.



ليس كل المحركات تضع بيانات ال input script كاملة, هده البيانات موجودة في البلوك ويمكن لاي محرك بحث او اي شخص يمتلك full node اني يراها, ولكن اظهار البيانات للمستخدم يتطلب موارد وكل مازادت البيانات كلما زادث الموارد, لذلك البعض لا يضهر ال script اما لتقليل استخدام الموارد او لكونهم يعتقدون انها ليست مهمة اصلا.

Per year, keep in mind that the 20% I used accounts for many other things which will be hard to predict, the global average failure rate IMO is about 10% and double that for fixing your gears, for example, one of the members here offers one working 17 series board in exchange for 2 dead boards, so if you get say a total of 20% failure it means the final outcome would be 10%, but then there is a downtime, shipping cost and a few other things involved.

But keep in mind, not everyone will have that 10-20% failure rate, some people reported a 3-5% yearly failure rate, others reported up to 90%, if you get a bad batch that makes up 50% of your purchase, you will be in deep trouble, so the failure rate is really near impossible to predict, but based on my experience and what I hear from many other miners, I believe 20% should be used as a general rule of thumb.

Keep in mind that the size of your farm is also very important because even when the global failure rate is 2-3% x model, it could be 50% for you and of course vice versa, I think a few folks around here got lucky with the 17 series, I personally had close to 100% failure rate in the first few months, a large Chinese miner (at that time) reported 40-60% failure rate in the first 6 months (Bitmain's CEO himself confirmed that high failure rate).

If you are a large player who places orders in the tens of millions, that might not be a huge issue, I remember how Bitmain offered to send their technicians to those large players in China to have their 17 series fixed on the spot, and yet they screwed the majority of other clients by forcing them to send a whole miner to get 1 board fixed, and then some people had to wait forever to get the gears back, some got back broken miners, so it's different for everyone, you might be lucky, but when you are running the numbers for a business you should not count on luck and -- instead, you should use the worst-case scenario numbers.


GPUs are a different story, those companies have been building them for decades, those companies know they will be making those GPUs forever, they also operate in countries where the law protects the client to some degree, the warranty gets honored despite the small failure rate, on the other hand, with ASIC miners, you are dealing with Chinese companies that have poor quality control, only a couple of years of experience, demand to supply ratio is unbelievable, and of course, they know their business won't last forever.

Another factor is that ASIC makers are under severe pressure, they rush to make miners and sell them to the public before doing enough testing, a great example is the entire 17 series, 90% of failures were caused by bad solder/soldering, Issues like that appear in the first month or two of running, which means nearly zero testing was done on them.

Now you have the S19 series, the PSUs suck and have a high failure rate, another proof that it didn't undergo enough testing.

A bulk Chinese reseller whom I trust confirmed that the new Avalons PSUs have up to 80% failure rate, another proof that Canaan did close to nothing of testing.

MicroBT put out a new firmware that can't be reverted, and guess what? once you flash it the old PSU is no longer compatible, and thus you will have to either order a new PSU or a replacement hash board, which is out of stock most of the time, not sure if they did it intentionally or they were just careless, but the outcome remains the same for the poor folks who had to go through that trouble, and it gives you a good idea of how bad these companies manage their products.

I can go on forever, but I think I need not, it's plain simple, GPUs are more reliable than ASICs by an order of magnitude, nobody in their right mind should assume that the quality of both is similar in any way, shape or form.

I admire your intention, but you can't do anything about it, in fact, you don't even need to worry about the strength of bitcoin because it's already too strong, with mining there is always a self-interest in protecting the network, the more hashrate you have means the more profit you make which also means the more you have to worry about keeping bitcoin safe.

despite the fact that bitcoin mining is on the brink of being controlled by a few large entities -- the network is more secured than ever, if we are talking about decentralization then ya that's a whole different topic, but as far as security is concerned, there is exactly nothing the average joe can do, especially not with a raspi, it's like a drop of water in the desert.

في الواقع اخي يوجد طريقة لمعرفة نوع العنواين التي تبتدا برقم 3, ولكن بشرط ان يكون هناك transaction صادر من ذلك العنوان, الشرح المفصل كنت قد كتبته في هدا الموضوع https://bitcointalk.org/index.php?topic=5133475.0

الملخص هوا انه يمكنك معرفة نوع العنوان عن طريق الاطلاع على input scripts, فأن كانت Segwit ستكون ضاهرة على انها P2SH او Witness وعدم وجود هده الرموز يعني انه العنوان عبارة عن Multisig.

you are right about the quality recent miners of Bitmain (only Bitmain) has improved compared to the previous 17 series, but other companies like Avalon are now making worse gears, I did not pull the 20% out of no where, people have reported 80% failure rate on new Avalon gears, 25% on S19's PSUs, 10% on MicroBT 20-30 series, it's nothing but wishful thinking if you run the numbers with the assumption that failure rate would be below 10%.

The reason why I said 20% was to account for

1-The downtime
2-Repair Cost
3-Shippment fees

If you want to use 10% or even 8% then you need to take those factors into account as well, you need to consider that waiting for 6 months to get your miners back isn't something unusual, or sending a whole miner that has only 1 dead hash board and then having to wait for a month or two to get it back, many things can go wrong and using anything below 10%+ the three points above is honestly being unrealistic. 



Not sure why you think that keeping a hardware wallet safe without anyone knowing you have it is harder than maintaining a mining farm that's worth a few million and sucking thousands of amps of the grid.



This is a valid point but it's a bit of cherry-picking IMO, so this model of "using tax to your advantage" isn't fit for everyone, if I don't have unrealized tax gains that are worth 32% of whatever I am willing to spend on mining -- it's no longer 32% is it? it could be 5% or even 1%? please elaborate more on this subject without assuming that everyone who enters mining has some sort of massive unrealized tax gains.


He deserves them, a very detailed explanation of how the tax stuff works in the U.S.



the miner's lifespan is more important than the power rate, I have free power but I have hundreds of dead hash boards  S9ks, S9, S9se, D3s, S17 pro, S17,S17 +, T17, L3s and you name the rest, recently I fired up 40 Avalon 841 miners which run like shit, glad I didn't buy more.

I corrected the word for you based on your last post.

Also, you are confusing yourself, the amount or number of shares means EXACTLY nothing to finding a block, a bitcoin block is just a hash equal or below target ( in other words, = or > than difficulty).

The protocol does not care about the number of hashes you submit, if the current difficulty is 1000 and you get 1 billion shares and none of them is above 1000, then you get no blocks, if you submit only one share that is = or > than 1000, you get a block reward, it's that simple.

But!  your chances of getting a hash that is above 1000 will INCREASE with the more hashes you hash in x amount of time(the more work your miner does).

Imagine you are rolling a dice and the goal is to get the number 6, your hands are slow you can only roll it once every minute, someone else rolls the dice 100 times a minute (they have 100x more chances than you do), but still you can get 6 at the first trial and that person doesn't get any 6 for 6 days straight, but what are odds of that happening!

8% is being over-optimistic IMO, it would be more realistic to run the number with 20% with the assumption that you won't pay for fixing using cash but rather in hashrate, so no extra cost added for repair, also to take into account that downtime and whatnot.

Also, the lost wallet or hacks isn't something to take into consideration here, unless you apply that to mining as well, bad security practices will cost you a lot whether you mine or hodl, someone got a virus in their farms that disabled the bootloader and had to replace over 600 control boards, with the downtime and labor that was a deadly cost, besides, taking care of one hardware wallet is a lot easier than safely maintaining a larger mining farm, a thousand MORE things can go wrong while mining as opposed to silently holding bitcoin, so really, it makes more sense to use this argument against mining.



So it's 661k + extra 92 miners vs 646 + no extra miners?


Now, let's compare these two against buying bitcoin instead, bitcoin has an average of 275% gain every year, and if we want to ignore anything before 2018 the mean annual gain is 93.8%.

There is also something I don't understand here, why do you use $723,520 instead of $1,064,000, why do you deduct the 32% tax upfront, I am confused here, who will give you the 32%, how and when?

Your initial investment should be $1,064,000 and the tax you pay on income later will be reduced, isn't that how it works? please elaborate.


Great post by the way, gave you some merit for the detailed info.