Topic: [ANN] Spondoolies-Tech - carrier grade, data center ready mining rigs - page 15. (Read 1260202 times)

Well spoken! Pretty much what I was thinking but couldn't put in writing 

I mined BTC with GPU's from June 2011 to about July 2013 when I finally got my BFL 30 gh'er.The diff did grow,but EVERYONE still made coin.The earnings were still pretty good until Bitfury (I know they just sold to that guy mining in washington)& KnC,etc.....  started their mines.

Thanks for all the input guys!! 

Your delusional if you think GPU and ASIC mining is comparable.  ANYONE can buy a gpu and fire up a rig.  yeah maybe the big boys could get thier GPUs from the source maybe 50% cheaper than the regular joe but at least everyone has a shot.

The way the ASIC bs works now , is its propriety they ASIC makers basically mine the hell out of watever with the ASICs the develop and either sell the tattered masses chips that have been blasted to hell for months or old tech 55nano crap they  want to upgrade out of

so they get to keep out money AND mine at the same time its a total joke.  


its not like AMD and Nvidia are biting the hand that feeds them like the Asic companies do so i trust them far more.


Not to say that if all coins were GPU mined it would not be harder to the average joe to find blocks or watever but thats what pools are for.  Any sane home miner would rather go back to GPU mining than stay with ASICs becuase the whole ASIC game is doubly rigged

its not just the money they have over you its the hoarding of the Tech they get away with too or outright just ripping fools off lol , hello KNC, Black Arrow the list goes on.

 Never heard of AMD or Nvidia taking peoples money and not giving them what they paid for right ?  i didn't think so.

Well spoken!

A great time for PoC (and release of PoC2 in a matter of a month) to take over as preferred PoW style of mining. Requires almost no power, and keeps decentralization as it is most efficiently mined with drives that people already own. The consumer grade hardware will forever be king of BURST and PoC algorithm. Come along to the future of PoW style mining, and check out the way crypto was meant to be.

https://bitcointalksearch.org/topic/ann-new-burst-op-mine-any-free-space-hdd-mining-ats-ae-p2p-marketmore-1323657

But by then fees will be paying the bills. When diff changes make so little difference that a piece of gear's viability would be measured in years rather than months, the king will definitely be power cost and only the central farms will still have it.

At some point, either after enough halvings or if the price drops enough, it will no longer be cost effective to mine bitcoin on a large scale.  At that time all of the mega mines will close up shop and we will be back to small-time "home" miners securing the network.  I cannot see any other way bitcoin would ever go back to "the way it was".

I was literally typing up a response that is almost exactly the same as you just posted. No matter what happens, the participation incentive for all these coins is based on profit. When money is involved, companies will find a way no matter what changes are made to continue making a profit.

I keep seeing people say that coins mined with ASICs need to go back to GPU mining. If that happens all the datacenters running ASIC equipment will switch over to GPUs as we have seen with the current flavor of the month ETH. This would also make it easier to bounce around between coins if everything was mined on the same hardware. I also see people saying that transitioning to PoS will help. Problem with that is that profits from PoS are based on the amount of currency you control. So then companies will just buy up the coins and profit that way without the additional hardware/operating costs of running large datacenters.

There is just no way to keep things decentralized when money is involved. People just need to accept that and move on.

A thousand times this ^^

If there is any conceivable way to scale mining of a coin and revenues generated by mining outweigh the expenses, then 10/10 times mining power will be consolidated over time. You can pick any coin, any algorithm, any hardware specification, it doesn't matter...if it's profitable and it scales, it'll go in a datacenter. There hasn't been a feasible way around this, since really it's just economics.

That's what the Cryptonote protocol's cryptonight PoW does.

https://en.wikipedia.org/wiki/CryptoNote#Egalitarian_proof_of_work

It uses the AES-NI acceleration built into modern CPUs, while occupying 2MB of L2/L3 cache per mining thread.

Unless you have better fabs than Intel and AMD, you can't (economically) built an ASIC with superior specs.

Of course this is a typically thankless accomplishment, as the peanut gallery simply stops whining about ASICs and starts whining about botnets.   

How about an algorithm, that (currently) relies on specific features of specific hardware? For example, the current generation of Intel or AMD processors for desktops or servers, the amount of cache or RAM they have or must have, and certain instruction sets built in to those chips, such as AES or SS-something.

It will last for a few years or a couple of generations, but ,.. eventually GPUs will have more power than CPUs, and FPGAs will get built, and then ASICs.

In the short term, you may have a coin that can only be CPU mined, because whatever algo it uses is best run on a particular architecture, such as the latest Xeons or Core i7s, because it simply won't run on a GPU that doesn't have the same amount of L1, L2 or L3 cache and RAM. The problem then is that to mine this coin would require a computer where you reserve that much RAM for it, or possibly even dedicate the whole machine to it.

I can see desktops today that have 8 GB to 16 GB of RAM on them. And I see servers with 36 core processors and 2 TB of RAM, where you could virtualize your miners to run a whole bunch of them.

I don't see GPUs with that much RAM. I definitely don't see FPGAs or ASICs with that much RAM.

It's much better or more practical to just get that particular hardware.

Even in this case, centralization will happen, because someone will just buy a whole bunch of machines and stuff them in a warehouse. Yes, everyone that can afford a desktop unit can be a miner, but the big guys will do the same thing. And there will always be pools.

I don't know how mining is going to turn out for bitcoin, but at the very least I can see the technology can exist way into the future and usage of the coin grow.

As for you guys with large warehouses and cheap power, you may or may not help the little guys hosting their stuff, but for sure, it would make sense to you to also run your own stuff, because eventually the little guys will either pull out or just stop.

But don't underestimate the hundreds or thousands of altruistic miners. They're the same guys looking for digits of pi, looking for prime numbers, looking for ruler thingies, cracking RSA, looking for aliens in outer space, finding a cure for cancer, and a bunch of other things, all for no pay.

And don't forget the small pockets of "rogue" or opportunistic miners, people who stick their little boxes in their office (for free power) or dorm rooms (for students).

As long as a manufacturer stays in the game and sells end products to consumers, someone will buy them and run them. It's a pity that this thread's manufacturer is gone, they had some good miners during their time.

I think the altcoin scene will soak up a lot of the hardware that was used for mining BTC initially, but only temporarily before being put on the market. With electricity prices in my country being so high it's simply not worthwhile mining BTC, but there are a few altcoins that are worthwhile mining if the rental price is right.

For sure, the company with the most efficient miner is going to stay in the game so long as it is profitable.  However, that profit margin will be too thin to continue the arms race of spending millions on next generation miners.  One company will win with the most efficient hardware and lowest power cost.  Centralization is inevitable. 

One thing about the halving both good and bad is all the farms which will be turning off. The good is, diff might drop which will help compensate the loss of revenue (though this will stabilize to a new equilibrium); the bad is, now a lot of gear will hit the secondhand market and might be scooped up by a few farms with cripplingly low power cost and extra space on the shelf. So the diff drop from those machines is only temporary, and the new owner is now more powerful than before.

The entire Bitcoin economy was fostered on the backs of miners who profit from the creation of coins.  If the coins are profitable, financially backed enterprises will adopt it and mine it, taking a greater share of the profits until it is no longer profitable to do so.

I think the halving is supposed to help prevent this.  While there is thought that the price of Bitcoin will increase when it halves, it likely won't just double in value.  Primarily because there are whales in the harbor waiting to dump coins when the value reaches some threshold that they've agreed to with their investors.  That's my guess anyway.  But such a reduction on the thin margins that Bitcoin is earning miners currently will result in more than a few major organizations to switch off their mines.  Something I believe Bitmain is seeing as well, and so they are now diversifying into alternative technologies.

The overwhelming portion of the hashrate being Chinese pools is another problem heavily related to corporate greed. When the whole world sends its manufacturing to a single nation whose dual-weilding of government corruption and civil rights abuses keeps costs low, we're paying evil for the right to exploit us. But the root of that problem is outside the scope of the bitcoin economy to deal with, and is more a by-product of the "greedy bastard" quality inherent to some extent in all human nature.

You are so correct. People need to accept this and, while this is also not desirable in the long term, they need to think that the low number of miners can bring us some advantages that a huge amount of miners will not. For example it is much easier to recover from fee mistakes. The fact that there is only a handful of miners make it alsmost impossible to keep the incorrect huge fee, while having a bigger number of miners will make it almost impossible to recover fees in the order of bitcoins. I do believe that in time the number of miners will grow and will reach a healthy decentralization while also having a couple of big players in the field for various services.

Anything that can get mined by any hardware is going to get mined almost entirely by a few large players anyway, because when you already have an assload of money people do you favors and allow you to get even further ahead. If someone can buy a thousand GPUs for 40% discount compared to the guy that can buy one, and can get wholesale electric for 1/3 the cost of average residential power, the game is always stacked in his favor. There is no hardware fight that the rich guys can't win; ASIC versus GPU is about how accessible the table scraps are to us peasants.

I'm all for the home miner, the little guy. That's been a consistent stance since I first showed up on here two or three years ago. My hosting is set up with no prepay and a per-customer MAX (rather than a MIN like a lot of places) power draw because I want to help small guys be able to compete. I started selling cheap and efficient server PSU hardware direct to home users because I wanted to help them narrow the head start the big farms had on 'em. Everything I've done for the last couple years has been to help keep small guys in the game because, even though I know it's a losing battle, it's still a battle worth fighting. Could I take my 100KW of 6-cent power and self-mine for my own profit? Sure, and I guarantee I'd come out farther ahead than I do with hosting. But that's not the point. The community has to help itself because everyone else is out to screw us over.

So if a GPU coin shows up that everyone can mine, who's going to be buying up all the GPUs? And if it's worth it, who's going to design an FPGA around it? And an ASIC, even if it's difficult?
And if the coin switches to proof of stake (which, if I understand correctly, pretty much earns you money based on how much money you already have?), it's still the same fight - the guy with the most money keeps winning more money.

There is no financial game where the guy with the most money doesn't win. But the line has to be drawn somewhere. Continually jumping from one easy game to another, and cashing out just before it caves in, is a great way to make short-term profits but in a zero-sum game every profit for you is a loss for someone else. I've hated the altcoin market since the beginning for that very fact. Until a coin shows up that miners in general treat like it's going to stick around a while with actual marketable value (based on actual features of a currency, like stability and utility) there's really no point. If Ethereum proves to do that, I'll get behind it. But for now, even if it's hard to compete with the large entities, bitcoin mining is it. We've been competing with the large entities since the beginning. Back then we used to work together, and people have been talking for over a year about what to do about the growing power of a dwindling number of manufacturers - but all it's been is talk. Maybe instead of giving up we should have actually done something instead of whining on the sidelines and walking away. Just throwing that out there.

Well,any coin that can be mined by ASICs is going to get mined solely by corp entities eventually,not by the masses.Is this what Bitcoin is about??

I kinda thought it was about power to the people kind if crap....guess I know nothing about Bitcoin after all 

Sorta like our US,"We the corporation" not "We the people" is going.

There is no algorithm you can't build dedicated hardware for, except one that changes frequently and unpredictably. But can you trust or rely on something that changes frequently and unpredictably? Is how it changes inherent to the protocol, or based on the whim of a central authority?

If it goes with an algo that can only be made by specific individuals in limited amounts & limited sales,yes 

GPU's & CPU's can be bought almost anywhere at anytime,so it can be farm mined too,but at least there is no extreme advantage like ASICs have to GPU's