|
Ja. Really, the one major difference between mines and general-purpose DC's is the business model behind them. A data center takes all comers and know that anyone who needs servers can use their services. Even more to the point is that typical DC customers are not concerned about day-to-day fiat exchange rates.
Mines of any sort have just 1 type of customer and that means it is more risky because those customers are at the mercy of crypto-to-fiat exchange rates. When exchange margins get tighter many miner customers start scaling back and even drop out of the game.
|
|
|
|
|
Well to start, the reason the utility wants to know the address is so they can determine how close the nearest substation is located to where you want to setup the equipment and be sure that their local grid can support what you want. They also need to verify that the area is zoned as either commercial or light/medium/heavy industrial to allow your operation there. An added 'gotcha' is that in the past couple years some utility companies located where power is/was dirt cheap (mainly in Oregon and Washington state) have become very unfriendly to mining operations so that may or may not factor into things.
That said, for loads less than a few MW usually the utility provides the main distribution xmfr that steps down to either 208 or 480VAC 3-phase wye (3 lines, 1 neutral) along with connecting the transformer to their grid. Lines from the transformer to the main incoming low voltage service disconnect and from there to your breaker panels is paid for and done by you.
Things the utility will want to know are:
Desired voltage (208V or 480V 3-phase w/neutral)
Type of load (lighting, motors for compressors, data-center)
Desired service capacity load
Expected average load
Maximum short-term load on startup (mainly applies to large refrigeration or A/C compressors)
Building plans detailing your power distribution at least up to your main breaker panel so they can be certain it is a safely designed setup that they are connecting to
|
|
|
|
|
That is a bunch of malarkey and marketing doublespeak. Aside from sheer power density of the equipment in the racks there is no difference between a 'normal' data center and ones built for dedicated purposes such as mining or protein folding.
All the same design rules apply to both of them: The need for power and network stability/backup, good efficient cooling and comprehensive operations monitoring is the same. Again, the only thing different is that the increased power density translates into much higher power & cooling requirements.
|
|
|
|
|
I have to add that if the holes I see in the pads do not also go through the PCB then DO NOT be tempted to drill some holes through the PCB! There is no (easy) way to know where there are signal traces or power planes inside the boards (they can have up to 3 internal layers) and you cannot take the risk of cutting through them and/or shorting something out..
|
|
|
|
|
A couple points about this:
1) The OP mentioned a data cable getting burnt. If the terminal that came off was the negative one, a burnt cable tells me that w/o the high-current connection to the PSU the chips were probably trying to complete the circuit by trying to connect through the returns in the data cable. Not good! There is a chance that other parts may have been damaged by that (hopefully not).
2) The bus terminals are very substantial chunks of metal connecting to an equally substantial pad on the PCB. That means you will need a rather hefty soldering iron - at least 50W and even higher would be better. In fact, one of those old transformer-type soldering guns - usually >100w - would be very handy...
The pics show what looks to be bolt holes in both the PCB and terminal. Wonder why Canaan did not reinforce the connection by using a bolt as well as just solder?
|
|
|
|
Of course, if all exchange had acted faster and initiated those measures in time the 14 customers wouldn't have succeeded in sending the bitcoins. Probably ten more customers would have been stopped from completing the transaction.
That's not something we want to be relying on, though. We tell people about the importance of holding their own keys and not relying on exchanges and web-wallets, but so many of them clearly can't be trusted with that responsibility. I don't see a way around it. Blacklisting is something that will *never* be implemented at protocol level, yet we're applauding a centralized service for effectively doing that to prevent their eejit customers from giving money to scammers. Welcome to the Modern World where the Human part of the equation is being rubbed out from many facets of life. Apparently, Humans can no longer be relied upon to pay attention to what they are doing and when something happens they (or their survivors) then love to whine about it because "it's not their fault!" so, systems are being built to think for them.  What Coinbase is doing is no different than what auto makers are doing with their "safety" features like lane warning/guidance, back up cameras, vehicle next to you sensors etc. Then there is this: Because many (idiots) are killed/injured each around the world because they are walking while texting, decide to cross the road without a care or thought about anything and get hit by a vehicle, cities are installing crosswalk warning systems designed specifically get the twits attention. Of course, that just furthers degrades incentives to pay attention in the 1st place...  Case in point, a quote from the linked article (emphasis mine) We developed a cell phone app that warns pedestrians when they initiate unsafe crossings and tested the app in our stateof-the-art pedestrian simulator. The project expands on our first SAFER-SIM grant that investigated permissive alerts (ones that indicate when it is safe to cross). We found that texting pedestrians who were given permissive alerts took safer gaps than those without these alerts. However, they also paid much less attention to the traffic, relying on the alert system to identify when it was safe to cross. Safe bet the same applies to exchanges trying to automajikly prevent folks from send coin to scammers. Alfred E. Newman comes to mind with his "What? Me worry?" quip. |
|
|
|
|
Nope, not joking. That ^^ kind of stunts is why I stopped using BM gear years ago. Much prefer MicroBT (Whatsminers) and Canaan (Avalons). MicroBT are more efficient but the Avalons are better built.
|
|
|
|
As far as I understand, yes it's not really possible with that ASIC implementation.
Correct. The reason is that miners do not solve the complete hash (and never have). They only look for results that match diff so only do 1/2 the calcs needed to actually solve the equation. AFAIK this has always been the case since even CPU mining days. |
|
|
|
I wouldn't wait for investigations to be complete. Instead, I would take part in it and provide any information required. As you haven't done anything wrong, it would help if you liaised with the bank and the police to get to the root of this. Get actively involved.
I agree 100%. Folks under investigation who immediately go on the offensive are just putting a bigger target on their back. Assuming that you have done nothing wrong or even questionable, showing that you are not only willing but also eager to assist the investigation is always a good thing. |
|
|
|
|
It is not just an age related issue, ergo me changing the subject line...
Considering just how many folks here in the Forum seem to think that any cryptocoin is "Bitcoin" a good place to start would be making it clear that only BTC is 'Bitcoin'. All others are alts and of course most (but not all) are pretty worthless...
|
|
|
|
But seizing funds without enough proof is also against the law right Depends on the country. In the US, civil forfeiture laws not only DO allow seizure of assets on suspicion of a crime but also in many states DO NOT require that assets be returned if investigations end without prosecution! While fairly rare, there are numerous instances of that happening here in the US every year. When it does happen the only recourse is to file a civil suit against the authorities involved to recover the assets. Because the rules of evidence in civil suits are far more loose regarding what is allowed or not allowed the outcome of that is not always in your favor... |
|
|
|
^^ Proper place to post is the Awesomeminer thread...Requested the mods move it there. Oh, while you are there DO make sure Patrike knows about the bad git so he can try and do something about it or at least warn folks! AM does not have you dl from a git. Official site is https://www.awesomeminer.com/download |
|
|
|
er..I know folk with PSU's ...not sure if that is what you mean..lots and lots....
brad
edit: unless you mean no 'workaround' for the built in psu's and single bus?
Ja. The problem lies with the built-in ones. Old way was to feed in 12vdc and then go through a Vcore regulator on each board to feed the strings. Problem is that final regulator gives ~2-5% efficiency hit per-board. Solution is to use the PSU itself to do the regulation by making it programmable. That in it itself was responsible for a large part of better miner efficiency these days. That also let mfgrs make longer strings of chips because a custom PSU can put out whatever voltage is spec'd. Gee - shades of Bitfury who used to use boards with strings long enough to take either 24 or 48v feeds... As I recall the Bitmains range from 15 to over 20vdc depending on the model. As always, using higher voltage (to feed longer strings) while pulling the same current per-chip is a win-win. Only main glitch to that is that all the different PSU's respond differently to the signal the control board provides to set the output voltage. Should be a piece of cake for a mfgr to standardize that across their models but noooooo.... |
|
|
|
How do we know for sure if we hit a block given the last fiasco? If the miner hadn't noticed he hit a block NO ONE would have known, and we're renting hash so how do we know? Only 'backup' notification I can think of would be if NH lets the person who is running the rental know they found a block via a stats page or whatever. I've never seen a reason to rent so I have no idea if NH does that. |
|
|
|
For the past several years all of the major miner manufacturers have shown zero interest in making small, quiet miners. Only Sidehack's R606 fits that category and it tops out at 1THs. That said, Sidehack is working on a new faster (and still quiet) miner but so far has published no specs on it aside from having a target power consumption of around 500w vs the R606's sub-100w. |
|
|
|
|
Bet it is the top board that's dead, right? That is the one I had trouble with on 3 of them. Only 1 still running and has 2 boards working only because I salvaged a replacement board from one of the others.
I've had R4's from batch-1 through their last (6). None have that fan so it has to be a mod and is located to blow over the PCIe power plugs & Vcore regulator areas. Not a bad idea, but I've never had any problems related to that area getting too hot.
|
|
|
|
 Sounds like these will be quite nice! Also expanded on earlier reply
|
|
|
|
Heh heh! Actually no, quite the opposite. The substrates (heat spreading ceramic parts of them under the emitters) of around 90% of all high power LED's produced in the world are produced in Taiwan under our patents on the laser drilling process used to put vias in them.  My company SLI introduced the process to the world in late 2007. In 2006 I was in Taiwan at THEIL (Tong Hsing Electronics Ltd.) installing some new equipment when Management requested I sit in on a strategy meeting with one of their customers (LumiLEDS). LL wanted to know if there was a way to make smaller vias in ceramic to increase emitter density on the chips. I though for a couple seconds then answered with a qualified, "yeeesssss, but I need to do some testing on an idea for doing it..." You could almost hear the wallets snapping open.... When I got back home we got dev money from THEIL and the rest is history  https://www.synchronlaser.com/processing-ceramic-substrates-2/https://www.synchronlaser.com/fiber-lasers-scribing-of-ceramic-wafers/Just so happens that industrial fiber lasers were starting to enter the market with enough power and near perfect Gaussian beam shape that could do the trick provided there was a way around the problem of the fiber laser wavelength (same as YAG) having very poor absorption in Al 2O 3 and AlN. As in over 85% of the energy just passing through with little to no effect on the materials. I solved that and pattern density increased 3-6x, SLI got the patent on it, in return for their funding development we gave THEIL exclusive rights to use it for 2 years with systems we built for them (to-date now over 45 systems at THEIL alone). Since they also provide services for Cree and all the other major LED mfgrs in the world within a year HP LED's were popping up everywhere with prices dropping like a stone. |
|
|
|
Are all miners on the controller the same type? Must be all 841's. Maybe the others are 821's? You cannot mix them on 1 controller. As for troubleshooting - um, you *did* look at page-1 here right? |
|
|
|
|
Show Posts
