Bitcoin Forum
March 19, 2024, 06:21:33 AM *
News: Latest Bitcoin Core release: 26.0 [Torrent]
 
  Home Help Search Login Register More  
  Show Posts
Pages: « 1 ... 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 [258] 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 ... 800 »
5141  Bitcoin / Hardware / Re: Novec 7000 Project [immersive evaporating cooling] on: August 09, 2013, 07:57:40 PM
The issue the OP had with pressure is that the heatsink/peltier were insufficient to extract heat.  The FPGA was injecting x watts into the working fluid and the heatsink was removing y watts.  y<x so the amount of the fluid that was in gas form increased over time.  A larger heatsink, more powerful peltier, or an active compressor would improve stability.  Still one way to avoid it is like aTG said it is easier to just work at a higher temperature.

Yeah. Its not just about the total watts. At higher temperature delta more watts is going to get transferred. It is very possible that with same setup 50C liquid cools just fine...


I probably wasn't clear but that is what I was saying (or trying to).  Too bad the easiest change (working fluid with higher boiling point) isn't a cheap switch.  Still very cool experiment.
5142  Bitcoin / Hardware / Re: ► ► ►HashFast Endorsement on: August 09, 2013, 07:51:52 PM
And let me not forget one other significant thing; this unit has water cooling.  NO MORE NOISE.  There is a God!

That's a common misconception... You still need a noisy fan to cool down the radiator in a water cooling setup. Unless you have a fancy setup hooked to swimming pool or other large body of permanently cool water.

Well you need a fan but it doesn't need to be noisy.  However you are right in the longer context.  Water "cooling" is actually a misnomer.  The water is merely a heat transfer system.  Much like a copper heat sink is a heat transfer system.  Nobody would call that "copper cooling".  Note this isn't a dig at HashFast the term is "water cooling" it just happens to be inaccurate.

HOWEVER the advantage of water cooling is that water is an amazingly good heat transfer system (transfers about 25x the heat per volume of fluid).  So you can move the heat to a much larger areas like a radiator with 1000x the surface area of the chip and use fewer larger slower fans.  No noise is a mismoner but it can be much quieter.
5143  Bitcoin / Development & Technical Discussion / Re: SHA-2* family maybe broken in several years. on: August 09, 2013, 07:16:43 PM
Change the POW the day BFL ships to their last customer...   

Change it today (with a delayed start in 6 block months) as a test on how many people verify the code they are running before upgrading. Smiley
5144  Bitcoin / Development & Technical Discussion / Re: SHA-2* family maybe broken in several years. on: August 09, 2013, 07:08:17 PM
... such a back door, if one existed, would have to be so cleverly hidden that no programmer or mathematician or otherwise smart person would see it.

Yes. But we're talking about the NSA here. They've got more resources than you can shake a stick at. They've been working on this problem hardcore since the end of WWII, because that's what wins wars. If there was a backdoor, you'd have to be incredibly naive to think that they would make this information public.

Its not that they would make it public, it is that finding this backdoor would be the equivelent of the holy grail in cryptography.  It would instantly give you +1 quadrillion nerd points and make you a recognized name in the field.  You could get any job in cryptography anywhere in the world with your paper as your entire resume. Every cryptographer in the WORLD has been looking for flaws either intentional or merely unintended weakness for over a decade.   It has been extensively studied, analyzes, countless papers on theoretical vulnerabilities have been written, the basis for SHA-2 (DM construct) is lectured about in computer science and math courses.   It isn't just a couple cryptographers in academia looking for a flaw, we are talking about cryptographers working for foreign governments, cryptographers working for financial institutions.  You don't think if a researching working for the Russians or Chinese couldn't use this as a way to discredit the US they wouldn't. That is the power of open source.  The NSA has some smart people, but the backdoor would have to be soooooooo freaking amazingly good that the entire planet couldn't find it despite actively looking for it for the better part of the decade.  If nobody has found the backdoor then they might not for decades to come.  So why design the SHA-3 standard (which was selected in an open and transparent way) and kill your own backdoor?

To put it into context there was a RNG based on ECC which had a backdoor (NIST still claims it is merely a vulnerability).  Despite not even being popular, almost nobody was even using it, most people hadn't even heard of it, the flaw was found and published by a cryptographer just a few months after NIST published the spec.  That was a nothing compared to the effort that has gone into validating SHA-2.  It is the cornerstone of the entire cryptographic world.  Everything from secure communication, to military codes, to the global financial system depends on SHA-2.  The potential damages could run into the trillions making the effect on Bitcoin look like a rounding error. Obviously disproving a negative is impossible, but it is precisely because of the amount of time spent studying SHA-2 trying to find a vulnerability ANY vulnerability and coming up empty for over a decade that SHA-2 is consider strong cryptography, not because anyone trusts the NSA.
5145  Bitcoin / Hardware / Re: Novec 7000 Project [immersive evaporating cooling] on: August 09, 2013, 06:06:39 PM
Why not just build a pressure chamber since you have to condense the working fluid and use R134a as its a fraction of the cost?  Would only take 80-100PSI to hit a similar range.

The cost of pressure chamber, the cost of compressor, the ongoing electrical cost. 
Still it is an interesting idea.  Essentially building an air conditioner around the heat source.  Would be a cool experiment.
5146  Bitcoin / Hardware / Re: Novec 7000 Project [immersive evaporating cooling] on: August 09, 2013, 05:48:56 PM
Hi OP, you could try to contact Allied Control (http://www.allied-control.com/index.php). I know they have built and run a 1T+ mining farm using over 6000+ spartan FPGA with immersion cooling. But they have dissolved the farm a few months ago because of the difficulty and the ASIC miner.

Yeah that is another option for working fluid.  IIRC they used Novec 649 (boiling point of 50C).
http://solutions.3mmagyar.hu/3MContentRetrievalAPI/BlobServlet?lmd=1351678101000&locale=hu_HU&assetType=MMM_Image&assetId=1319241050803&blobAttribute=ImageFile

It has a higher boiling point of 50C.   I think a fluid around 45C to 55C is probably the sweet spot for chip cooling. The lower the boiling point relative to ambient temp the more difficult it is to cool the gas to below the boiling point so it condenses back into a liquid.  The issue the OP had with pressure is that the heatsink/peltier were insufficient to extract heat.  The FPGA was injecting x watts into the working fluid and the heatsink was removing y watts.  y<x so the amount of the fluid that was in gas form increased over time.  A larger heatsink, more powerful peltier, or an active compressor would improve stability.  Still one way to avoid it is like aTG said it is easier to just work at a higher temperature.

The again you don't want too high of a boiling point because you start to run up against the thermal limits of ASICs and FPGAs.  If the working fluid reaches a max temp of 50C, then the surface of the chip package will be ~50C.  The internal die temp will higher.  How much higher depends on thing like the chip design, surface area, and chip package.  Using CPU and GPU as a data point they have thermal diodes as part of the die to record the internal temp and they run 10C to 20C higher than working fluid temp.  In a two phase system (liquid->gas->liquid cycle) the package surface is never going to cooler than the boiling point as the boiling is what "removes" the heat.    A different route would be using a much higher boiling point and some circulating pumps with a heat exchanger (radiator) and just keep the entire system in one phase (liquid).  The system would never boil and would use circulation to move the heat from the chip to the radiator, but where is the fun in that. Smiley  Anyone know what the upper safe limit is for various FPGA & ASIC designs?  If any have thermal throttling where does it kick in at? On edit: Looks like Avalons default is 60C and BFL is 65C.  Not sure if that can be pushed higher.

Semi-OT but my rant on the excessively high cost limiting what a lot of potential innovation
Novec 649 is still ~$300 a gallon though.  Actually I don't know of any working fluid that has a high Dielectric Strength that isn't $300 per gallon or more.  Pricing is also tightly controlled so no free market at work.  3M is the only major supplier, it holds a lot of patents which it aggresively defends (so chance of startup making a Novec-649 "compatible" fluid is essentially nil) and it doesn't allow discounting.  Distributors or retailers that get caught risk being blacklisted and with only one supplier that means you are out.    It is as shame 3M has never been able (or willing) to bring the cost down.  I know it is a highly profitable niche market and I am sure 3M margins are something like 90%+ but imagine how many more applications would be economical at $20 or even $50 a gallon.  Part of the problem is 3M doesn't really want there to be a cheap solution.  For example one commercial application is for the cooling of lasers.  It kinda is the same scenario as cooling a CPU/GPU/FGPA/ASIC but it is a lot more energy (like 1000W or more) in a much smaller space.  The laser isn't immersed the laser shell is filled with the fluid.  This means much less is used.  A typical laser might only use 250CC (~1/16th of a gallon).  The system is completely sealed so they only replacement fluid is what is lost during maintenance.  Trust me the stuff is treated like liquid gold but you usually need to top the laser back up (maybe 20CC or less).  Maybe someday another company without the vested interest to keep prices high will design and patent a much cheaper working fluid.  If they do the cynic in me says 3M will buy them out and jack the price up 1000%.
5147  Bitcoin / Mining speculation / Re: Useless ASICs on: August 09, 2013, 04:24:40 PM
Ok so theoretically, won't we reach a point where so many people have ASICs that they will generate fewer in BTC than they consume in electricity? Won't we reach a parity of sorts at which point people become unmotivated to mine in any way? This would be just like what has happened with the GPUs and FPGAs, but without a possible improvement of mining technology.

Obviously, should they become worth 1000$ each then it remains worth it for longer.

Simple answer is yes but then again mining is generally a low margin relatively boring endevor.  The change of technology (CPU->GPU GPU->ASIC) creates a short term "mania" but eventually things settle down.  Yes difficulty will keep rising in the near future but eventually the margins on new rigs will be so low that people will stop buying them (or buy less of them) that will slow the curve.  At higher difficulty the least efficient rigs will operate at a loss so the miners will idle them and that will lower difficulty (or at least flatline).  Then it becomes a more mundane business of replacement.  With difficulty high new ASICs sales will be relatively low.  It will be the most efficient units being sold and as they raise difficulty the least efficient miners will be forced out.

Remember GPU was about 100x as efficient as a CPU yet eventually the market adapted.  Difficulty didn't keep doubling every month for 2 years continually.  The reality is most people will never get rich mining.  Yes if you get the right tech at the right time you might make a fortune but really that is more luck than anything else.  Those that picked Avalon1 over BFL "won" but those who picked Avalon3 or bulk chips not so much. 

Quote
I sort of imagine a future where bitcoin mining is done philanthropically by those who can afford to, in order to keep money honest, but not as something you are materially rewarded for. Maybe even by a *shudder* elected body.

Highly unlikely.  The margin will be compressed to a point where the cost to mine a BTC is slightly less than the cost to buy one.  If the return goes negative well people can always click the off switch and when they do the difficulty relative to price will drop.  If it drops enough then people will buy or turn on rigs and difficulty will rise.  If it rises to much the least efficient miners will be forced out again and difficulty declines .... etc ... etc ... etc.  This is essentially the cycle that GPU mining has been in for two years now.  Of course even there many people will still mine.  Not everyone is irrational into thinking mining will make them a millionaire in a year, and a billionaire in a decade.  Mining allows anyone to convert electricity into BTC.  It is a way to "buy" BTC without any need for an exchange, or counterparty.
5148  Bitcoin / Mining speculation / Re: Useless ASICs on: August 09, 2013, 04:15:20 PM
At the end of the day, it seems like the only people making any money are early adopters and people selling shovels/running pools, which is a truly sad state of affairs.

So, just like every other gold rush in history?  Not sure why you think bitcoin is somehow unique.  The signs were obvious a year ago that the long term profits were going to be in selling the miners.

Well why on earth would anyone new get into Bitcoin?  The value prop is disappearing.  And the people selling the hardware pretty much know they are screwing folks on the receiving end because of the big black void of difficulty vs. BTC value.

"Estimated mining power by October:

Minimum: 100TH AM + 100TH Avalon + 300TH Avalon chips + 200TH Bitfury + 300TH BFL = 1PH
Realistic: 300TH AM + 100TH Avalon + 500TH Avalon chips + 500TH Bitfury + 1PH BFL + 500TH KNC = 3PH (actually up to 5 or 6 with all upcoming hardware)

Revenue per GH per day: from 1 to 4 mBTC

Lifetime revenue:
The very best case (totally unrealistic) scenario: 1PH by October and 10x growth per year after that: 0.6 BTC/GH
Much more realistic case: 3PH by October and 100x growth per year after that: 0.1 BTC/GH
The worst case: 6PH by October and 1000x growth per year after that: 0.03 BTC/GH"

OK so seems like you've got some educated opinions on BTC and mining.
So I ask again:
How can someone new to BTC makes sense of these numbers posted above for the GB:

Your numbers are well simplistic.  While your Oct numbers may be realistic assumming that growth will continue up to 1000x year over year is "kinda" (and I use that term to be nice) unrealistic.  Within 6 months into 2014 Bitcoin mining would use more computing power then the entire rest of the world combined.  By the end of 2014 Bitcoin mining would use nearly $1B a year in electricity.  By the end of 2015 Bitcoin mining would nearly 30% of the energy used by the human race.  By the end of 2016 it would use enough energy to send a small probe to the nearest star.

Obviously difficulty growth can't remain exponential forever.   The curve will level off it is a mathematical certainty.  The only question is what is the shape of the curve. Look at GPUs, when they hit the scene they were up to 100x as powerful as a CPU.  Not only that it became possible to use 3, 4, even up to 8 per rig.  Thus difficulty/price exploded very rapidly.  If you extrapolated that trend in the first 3 months out over the next 2 years you would have predicted that difficulty would hit 8,000 BEFORE the first ASIC was produced.  Obviously that didn't happen.  So in any estimate you need to look at what is the break even point for each technology.  Start w/ Avalon because they are the least energy efficient.  At what difficulty would the average Avalon miner be mining (just looking at electrical cost) at a loss, assume $0.10 per kWh.  Once difficulty hits that point those miners will idle their rigs.  Long before that point nobody will buy new Avalon's because it simply no longer makes sense.
5149  Other / Beginners & Help / Re: new bitcoin miner 7GH speed on: August 09, 2013, 08:18:05 AM
I am glad it has support for dual HDMI.  I have been saying what is missing from current standalone miners is the ability to connect not just one but two monitors.
5150  Bitcoin / Hardware / Re: Novec 7000 Project [immersive evaporating cooling] on: August 09, 2013, 08:15:47 AM
http://en.wikipedia.org/wiki/FC-72

Boiling Point: 56C
5151  Bitcoin / Press / Re: 2013-08-08 Bloomberg: The SEC Shows Why Bitcoin Is Doomed on: August 09, 2013, 07:43:01 AM
Can you imagine the headlines if the judge had ruled the other way.  Does anyone think this guy would be saying Bitcoin now has a chance because a judge has ruled they are so worthless that you can't even commit fraud with them?
5152  Bitcoin / Group buys / Re: [OPEN] Hash Fast pre-order units mining shares on: August 09, 2013, 06:53:58 AM
Don't forget the 1% to cover expenses. Wink
5153  Bitcoin / Legal / Re: Do all bitcoin businesses have to register as MSBs as per FinCIN guidance? on: August 09, 2013, 05:04:31 AM
They registered with FinCEN as a money transmitter.  Still don't take the above as written in stone.  Personally I believe FinCEN guidance is on rather weak ground.  I guess it all depends on how far your legal counsel is willing to push things.
5154  Other / Off-topic / Re: Bootstrap or Foundation? on: August 09, 2013, 04:13:00 AM
if you really want to do something custom foundation is a good starting point. 
If you are looking fro rapid application development which gets out the door quick bootstrap.
5155  Bitcoin / Hardware / Re: HashFast launches sales of the Baby Jet on: August 09, 2013, 03:02:03 AM
Why does BitPay use such a low figure for USD/BTC? Also, I wish local pickup was an option.

Because MtGox price is for USD you can't withdrawal.  Not really useful for anything other than saying "look at the high number".  I am pretty sure bitpay price is ~= bitstamps price.
5156  Bitcoin / Hardware / Re: ► ► ►HashFast Endorsement on: August 09, 2013, 12:22:45 AM
The most likely path of a hashrate graph is that it will trace a sigmoid function.



It will start accelerating, then start to slow down when people stop increasing their hashrate as it becomes less profitable.  There is no way that hashrate will ever go down unless the price of bitcoin also colapses, so that mining ceases to even cover electricity costs. (Of course it'll be superimposed on another, slower growth caused by More's law as transistor density increases, but we can ignore that for now)

Right now, the difficulty will have to go up something like 100x before it starts to get close to power costs. At those rates a 400Gh/s device will only make a few dollars a day.

It is important to consider that not all units (either delivered or proposed) have similar electrical efficiency.  In a curve like indicated above one is in a better position if their personal "hash efficiency" (chip efficiency, PSU efficiency, and individual electrical rates combined to produce operating cost per hash) is superior to the "average Joe".  As difficulty climbs, gross revenue per unit of hashing power will decline and thus net revenue will get squeezed.  The marginal miner (higher electrical cost in USD per hash) will be crushed first.  Those miners will go negative ROI% and once they are sure that price/difficulty isn't going to improve (give up home) eventually idle their rigs.  That means it is possible for hashrate to decline independent of price.  Of course any decline will be short lived as miners contemplate higher ROI% on new units and place orders.  Unless shipping time is 1 day though there will be a lag (we have seen it in the GPU era even with supplies essentially unconstrained) and those higher efficiency units will be able to power through.

Still there are always risks involved, the odds are nobody is going to make a fortune mining anymore.  I would be happy to just generate some coins at a small margin and reduce the needs to deal with awful exchanges. Smiley
5157  Bitcoin / Hardware / Re: ► ► ►HashFast Endorsement on: August 09, 2013, 12:16:15 AM
If they are so sure of themselves then send me unit and I will pay in BTC when it is hashing. 

Why would ANYONE do that.  If someone has a unit which is currently hashing why not just keep using it themselves rather than sell it to you?  I am not saying you should buy one but thinking that someone will produce a unit, which is operational, and profitable and then turn it off just so they can send it to you and collect some BTC after the fact?  Would you?
5158  Other / Beginners & Help / Re: BTC a real Currency!!! Why the price drop on: August 09, 2013, 12:05:29 AM
Correlation is not causation.  The sun also rose this morning so maybe the sun going up makes Bitcoin prices go down.  Bitcoins enjoy the night life.
5159  Bitcoin / Bitcoin Discussion / Re: Do you think people pulling out of mining is good for Bitcoin? on: August 09, 2013, 12:00:26 AM
The more minors the better.

If there is a flaw in Bitcoin ASICS have exposed it... concentrating too much control in too few hands.

Someone needs to create a Bitcoin fork that solves this problem... insuring mining stays widely distributed over as many users as possible.


But they are selling the ASICs to the hands of many.  They can make more money selling to rubes than mining, and there goes your central distribution theory.

This.  As evident by how low ASICMiner was able to drop the price of USBMiners there is a lot of markup.  ASICMiner simply sold them that high because well everyone else tripped over their dick and left ASICMiner with no significant competition.  BFL jumping the gun and promising ASICs about a year before it was feasible started this whole staggered start run multi-year pre-order bonanza.  In time though the network will sort it out.  

There will be raw ASIC chips made by at least 3-4 manufacturers (not all of them are going to make it, but at least some will).  I also think eventually most chip builders will take the route that Avalon (and yes AMD) have done and not offer complete miners for sell, instead getting out of the retail business and selling chips to OEMs (yeah that "AMD video card isn't made or sold by AMD, just the chip is").  Which means you probably will see at least a dozen mining OEMs all of the world making end units in a variety of sizes and price points.  


The margins will shrink (probably go negative for a while) and Bitcoin will reflect the relatively boring, low margin commodity business that it is.  The risk takers and speculators will move on to the next shiny and a large number of miners will grind out a small (very small) profit between their capital costs plus electrical costs and the the coins minted.  Many will probably see it not as a way to get rich but as an indirect way to buy Bitcoins in a decentralized fashion.  Input so many hours, plus so much computing power plus so much burnt electricity and out pops a shiny Bitcoin.  
5160  Bitcoin / Legal / [No] Did FinCEN overstate the money transmitter definition in their guidance? on: August 08, 2013, 11:46:18 PM
On edit:  DOH.  I realized somehow I had window open with outdated (2009) version of the regs.  Well prior to 2011 when the reg was updated it would have been a decent argument.  The latest version of the regs however are symmetric so no joy there.

Quote
5) Money transmitter —(i) In general. (A) A person that provides money transmission services. The term “money transmission services” means the acceptance of currency, funds, or other value that substitutes for currency from one person and the transmission of currency, funds, or other value that substitutes for currency to another location or person by any means. “Any means” includes, but is not limited to, through a financial agency or institution; a Federal Reserve Bank or other facility of one or more Federal Reserve Banks, the Board of Governors of the Federal Reserve System, or both; an electronic funds transfer network; or an informal value transfer system

Since you can't delete anything from this forum I will leave the original incorrect post below so everyone can see my mistake.  
Quote
I was rereading the Bitcoin Foundation's response to DFI and the section dealing with the asymmetry of the MT law in CA.  At first it caused me to first say "man I wish the federal law was written similarly" and then made me check to see what the wording actually was.

In the guidance dated FinCEN states
Quote
The term "money transmission services" means "the acceptance of currency, funds, or other value that substitutes for currency from one person and the transmission of currency, funds, or other value that substitutes for currency to another location or person by any means."11

11)
31 CFR § 1010.100(ff)(5)(i)(A).

http://www.fincen.gov/statutes_regs/guidance/html/FIN-2013-G001.html

However what does the actual regulate states.

Quote
31 C.F.R. 103.11(uu)(5)

(5) Money transmitter— (i) In general. Money transmitter:
(A) Any person, whether or not licensed or required to be licensed, who engages as a business in accepting currency, or funds denominated in currency, and transmits the currency or funds, or the value of the currency or funds, by any means through a financial agency or institution, a Federal Reserve Bank or other facility of one or more Federal Reserve Banks, the Board of Governors of the Federal Reserve System, or both, or an electronic funds transfer network;

http://www.ecfr.gov/cgi-bin/retrieveECFR?gp=&SID=a7e1bef2c8dbac0d6893186f0c01390c&n=31y3.1.6.1.2&r=PART&ty=HTML#31:3.1.6.1.2.1.3.1


Removing the aspects not in dispute it becomes:
Quote
accepting currency, or funds denominated in currency, and transmits the currency or funds, or the value of the currency or funds

The actual definition doesn't provide three criteria for accepting and transmitting as the guidance indicates but rather two criteria for accepting and a third criteria for transmitting.

Accepting:
a) currency
b) funds denominated in currency

AND

Transmitting
a) currency
b) funds
c) the value of the currency or funds

Note that despite the statement in the guidance the actual reg does not define a money transmitter as one who accepts "the value of the currency or funds" that is only included in the definition for transmission.  One could argue this is merely an oversight by FinCEN however this particular reg has been updated three times.  If the reg was intended to state the following as FinCEN claims they why doesn't it?

Quote
Any person, whether or not licensed or required to be licensed, who engages as a business in accepting currency, or funds denominated in currency, or other value that substitutes for currency and transmits the currency or funds, or the value of the currency or funds, by any means through a financial agency or institution, a Federal Reserve Bank or other facility of one or more Federal Reserve Banks, the Board of Governors of the Federal Reserve System, or both, or an electronic funds transfer network

Reg 31 CFR § 1010.100(ff)(5)(i)(A) as implied by FinCEN's guidance (the bolded portion is absent from the actual regulation).
Pages: « 1 ... 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 [258] 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 ... 800 »
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!