While 88c is within the normal range specified by the manufacturer, you should take this info and toss it out because the manufacturer doesn't tell you that these T17 gears use a very low-quality solder which will break a lot sooner than you think, and thus, you want to try to run them as cool as possible to "hopefully" get them to hash longer.

This is similar to the multi-option firmware by Bitmain, I keep wondering why on earth won't these manufactures make an auto-tune firmware to tune every chip individually (at least the frequency for the miners that can't have different voltage), I mean BO+ does it, Vnish does it, and it works great, you get to utilize the "good" chips and let the "not so good" chips run at a different frequency that they like, and the results are.... just AWESOME.

The first thing I would have to tell you is that your prices are a bit off, if you don't trust your source 100% they are probably scammers,  trade assurance is B.S, and I don't know what PP will do if the seller proves they shipped you the miner you ordered which just so happened that it's in a miserable form, but, do what you think best.

Now, speaking of quality and only quality, the M21s is more robust than all the Antminer gears you mentioned by a few orders of magnitude, but then you talked about overclocking and what not, unfortunately, you won't be able to do that freely on whatsminer, whereby on all the Antminer/Bitmian gears you mentioned over clocking to extreme measures is possible using custom firmware, in fact, the S17 pro has a turbo mode which overclocks the miner using only the stock firmware.

 

57TH will make you a bout $21.5 a day before the power bill, since you posted this 2 days ago, i'll just worth with the highest number you could have had then which is $22.5 with btc price at $46,800.

This means, you think you will be paying $2.5 a day to run a  57TH gear, for that to be accurate you need to spend 10 cents an hour, based on your 16c kWh, you will need a gear that consumes 625w and puts out a 57TH, that means a gear that has an efficiency of 11w/th, which is about 30% of the consumption of the most effienct gear known to mankind, so no, your numbers are WRONG.

The best gear you could buy will be in the 30w/th range, which will be $6.5 for the power bill per day, and the cost will be in the 7k range, so as it stands it's  437.5 days to pay back for the machine, not sure what you did wrong, but your numbers are far off.

But just to answer your question, the numbers you see on whattomine.com and the other website are rather accurate (assuming you input the right power consumption which I think you didn't), it's not a guessing game, they take your "potential" hashrate and divide it by the total hashrate and since the block reward is constant, it's easy to accurately (not 100% accurate since luck and block fees vary) tell you how much you could have made today, but here is a recent incident, 2 days ago when you did the math, you could have got $22.5, now it shows $21.5, it goes out to show that it's very normal for this number to keep dropping.

While there is no way to confirm this, I strongly believe that the majority of this 8% comes from gears that were unprofitable at the 30k levels rather than the relocation of mining gears, of course, gears are being relocated and there is no doubt about it, I just highly doubt they are being moved this fast, if the price drops back to 30k in the next epoch, we should get some clarity regarding this.

I don't see why you can't, you will have two problems you need to fix.

1- The miner's fans, since you have to remove those for immersion cooling the control board won't allow the hash boards to hash, and thus, you have two solutions, running custom firmware like Vnish which has an option to disable the fan check, the other solution would be a fan simulator, it's something you can find online that "tricks" the control board by giving it a false signal similar to the one it would get from the actual fans.

2-The PSU's fans:

This will be a bit more tricky to solve, the PSU needs the fan to be spinning in order to function, so not sure how you would bypass that, however, I believe the control board controls the PSU, so my guess is that these custom firmware will also handle the PSU fans when you activate immersion cooling option, to confirm this, head to their official topis and ask there, check AwesomeMiner, Asic.to and BO+.

How did you figure out it was a control board issue? did you try a different control board? if you actually did, and it worked, it could be a different firmware, flash the same firmware on the "bad" control board and it should fix it.


Mixed hash boards can still work if they have the same product id/pcb, look for something like this in the kernel log


Another way to confirm if this is actually caused by mixing hash boards, run them 1 by 1, if they all work individually but not when mixed then you make sense, if they don't, then you have bigger problems to worry about.

Check this thread


You need a bit more than just that, this is a brief explanation of the things you are going to be doing > https://www.youtube.com/watch?v=5WH7g61d90w&t=89s.



It's what's under the heatsinks that matters, the solder that keeps the chip in place goes bad or the chip itself is burned, try to identify chip 25 on your chain 2 and see if you can see any physical damage like solder balls between the chip legs, refer to the image below


Source: https://www.zeusbtc.com/NewsDetails.asp?ID=187


Perfectly fine.

Not at all, it doesn't.

Did you buy this new or was it owned by someone else? the reason why I ask is that because some custom firmware will f the PIC and cause it to show 0xff, if you are lucky, a different firmware might fix this, it's highly unlikely that this is a hardware issue.


It will be something very close to this


My advice is, don't take the hardware path before trying the software, if different firmware doesn't fix it, order pickit3 for $18 and extract the hex from another hash board, if that does not fix it, then replace the actual PIC IC.

There you go.


The T17 chips are connected in series if all 30 Asics are not connected the board won't hash, your chain 1 shows 0 Asics which means either the first chip is bad or it has some other issues, chain 2 shows 24 Asics which means the signal is interrupted at chip number 25.

It's very, very, very common for T17 to lose chips this way, the solder bitmain used is of terrible quality, the reported failure rate for these gears was well over 30% in the first a few months, it has now been 2 years and without a doubt, the failure rate has gone up by quite a bit.

The bad news is that these are not easy to fix, you are going to need some tools to confirm which chips need replacement, and then you are going to need to buy the chips from China, and replacing them isn't something everyone can do, if you don't have the tools and skills needed, you will have to send it somewhere to fix it, where are you located?

Well, if your temps are fine, voltage is good, the PSU's screws are well tightened - you have no choice but to move to the software level.



I am not the best to answer this question, as the vast majority of gears I owned were Bitmain, only the S9/S9i and S9j did me good, the 17 series was the worst for me, however, as it stands I would still recommend Whatsminer and Avalon over Bitmain, any day and every day, I have tested the M21s and they are great gears, I just got an M20s which I should set up in the next a few days and see if it performs anywhere near the M21s.

Yes, it will keep going up until you start paying more for the power bill than you earn.

See, this rather simple math, your profit is decided mainly by two factors, price, and difficulty.

You want higher prices and lower difficulty for more profit, the opposite = less profit and potentially a loss.

The facts are, the price can go both ways, the difficulty only goes up, of course, it doesn't go up in a straight line, but if you google "bitcoin difficulty chart" you will understand that difficulty tends to go up a lot more often than the opposite direction, as a result, at one point into the future, your mining gear will become obsolete, that heavily depends on your power rate, someone who pays 5 cents can mine a lot longer with the same mining gear than someone who pays 10 cents, so with 12 cents kWh, you are going to be forced to shut down a lot faster than many other miners.

The above doesn't mean you won't be able to ROI and make some profit on top of that before having to shut down your gears, it just means your chances are a lot lower than those who have free power or pay only 4 cents, and given that mining is essentially a competition between miners, you might want to avoid getting into a battle knowing that the other opponents have more chances than you do.

alas, Bitmain just can't make a perfect gear again, it has been the case for years, the 17 series had a huge failure rate due to bad solder, and now the 19 series has a high PSU failure rate, you are not alone in this, I see many folks on telegram complain about these PSUs, including those who have perfect electricity setup and thus it isn't an issue with electricity whatsoever, Bitmain must have messed up with something in that PSU, they probably used a cheap component somewhere to save a few bucks resulting in the failure of mining gears which they sold for a few thousand dollars.

This is either a temperature issue or a power issue, make sure the miner isn't running too hot or too cold, try not to set a static fan speed to allow the miner to get to the temp it finds best, also, if your voltage is below 220v it could cause some issues, the last thing you should try would be a different firmware.

No, we won't, everyone's chances will be exactly the same, unless you think that we are all sending the same hashes aka we are overlapping one another and thus wasting resources which isn't the case, every individual miner receives a completely different job and submits completely different hashes so joining forces changes nonthing.

Here is a simplified explanation.

Miner A, B, and C, each receive different work from the pool so

A submits shares:

1,2,3,4,5.

B submits shares: 

6,7,8,9,10

C submits shares:

11,12,13,14,15

Notice how they are different, and how they total 15, it doesn't matter if they use the same stratum connection or not, each of them has 5 chances/shots, if they join forces they will have a total of 15, 3*5 is the same as 15 as far as the probability of finding a block is concerned.

If you master the use of AM or any other monitoring software, then you probably won't need an instant/mobile notification of any kind, the software itself can do all you can do at night, it can reboot the miner, change mining profile, increase/decrease fan speed and basically everything you can perform remotely.

But if you still need email notification while your phone is in silent mode then you can use "Tasker" App, I believe you can make an exception for a single email address (sender) to override/bypass the silent mode.

Your PIC fw version shows 0xff, it has to be 0xb9 like so


I believe you must be doing something wrong in the process, did you compare the hex and made sure you have the exact one extracted from a working hash board?

هدا المنشور عبارة عن هرطقات وكلام فارغ, الهدف منه اما تنصيب تروجان او فايروس على حاسوبك او فقط لاخد المال منك مقابل برنامج لا يسمن ولا يغني من جوع, اي شخص يمتلك ولو 1% من "المنطق" سيصل الى هده الخلاصة.

Let's go with the assumption that he missed the word "hour" in his 25 megawatts, which will be 25000KwH a year, or 68.5Khw a day, so that's a 13.7KW solar system with an average of 5 sun hrs a day.

If he really means 25 megawatts solar system (as in, he mistakenly added the word "year") then I'll call b.s.

You need a certified /experienced electrician to set up the electricity infrastructure, but you don't need one 24/7, once a while a fuse or an MCB will need to be changed, but that should be on-call basis, there is no point hiring someone just for that if the set up is perfect you might not need an electrician for years, if ever.

Since the place is secured, you need someone who can access the farm "physically" in the shortest time possible, if everything runs fine and the farm is in a passive state you will only need someone to go there to clean the miners maybe once a month or quarterly (it will depend on how clean the air is), but problems like a dead PSU or a miner that needs hard reboot are ever going, and there is really no way for you to know how frequent you are going to need that someone.

In other words, let's say you pay that person $100 per visit/task vs $2000 a month for full-time attendance at the farm, so if you have 50 miners, then paying per visit would make more sense, if you have to say 500 miners then the number of visits will probably be a lot higher and thus paying $2000 a month would be better, this depends entirely on what options you have and how many miners you have because the more miners you have the more likely you face issues.

The person you need to hire doesn't need to have an IT certificate, basic IT knowledge is enough, and of course, he needs to know how to use a screwdriver , most things he will be doing are:

- Changing a dead PSU
- Replacing a dead fan
- Changing a LAN cable

You will need to monitor everything from home and tell him for example, change fan.2 on miner number 111, but if you want someone to fully operate the farm then you might want a full-time employee.

You should also use monitoring software such as AwesomeMiner, learn how to use it (it isn't all that simple), once you master it, it will be as good as having 2-3 employees who are watching every gear 24/7, the software itself will reboot the miner if needed, control the hash rate, reduce it if the miner gets hot, increase it if it's cold, send you an email/telegram messages when a miner goes offline and etc., this drastically reduces the need of someone to be in the farm, and thus saves you a lot of money.


Wrong. there is a loss in monitoring the individual miner performance, if the hashrate drops at the pool, you will need to manually check which miner is having a problem, also there is a single point of failure if the hardware that hosts the proxy has any issues the whole farm will go offline, so you will need a few back-ups to escape that, if bandwidth isn't an issue, the use of proxy has more disadvantages than advantages, also, most open source proxy projects are abandoned and buggy as hell.