Bitcoin Forum
April 18, 2014, 09:05:46 AM *
News: Due to the OpenSSL heartbleed bug, changing your forum password is recommended.
 
  Home Help Search Donate Login Register  
  Show Posts
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 [49] 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 ... 726
961  Other / Beginners & Help / Re: Instant BTC to USD/PayPal on: January 26, 2014, 03:22:37 AM
https://BitSimple.com
962  Bitcoin / Bitcoin Discussion / Re: Has the NSA already broken bitcoin? on: January 24, 2014, 04:32:46 PM
The point he's making is... related to quantum computing vs bitcoin. Research it ..

Please tell me how quantum annealing can break cryptography.

Hint: it can't
963  Bitcoin / Development & Technical Discussion / Re: Idea for a next generation crypto distributed publicly by social network sign up on: January 23, 2014, 07:18:15 PM
If it isn't decentralized then what prevents the central authority form simply giving himself 100,000 coins.  Trust that the issuer will never do anything wrong?

If peers can't cryptographically verify that the issuance is valid then what prevents someone from hacking the client to simply tell the network "um yeah this client here witnessed 10,000 logins so lets generate him some coins".  Trust that no user would dare give themselves extra free money?

Well if you want a virtual currency where you need to have absolute trust in the issuer the USD is a better choice.

Simpler answer.
Don't start by thinking about how people will use it properly, start by thinking how will people break/hack/manipulate/steal/dupe this system.   If you don't have an answer for that beyond "trust and hope" you don't have anything.   I am not saying give up, I am saying start there. 
964  Bitcoin / Legal / Re: Filed a request for an administrative ruling with FinCEN this morning. on: January 23, 2014, 05:37:30 PM
yes, bumping this thread. DeathandTaxes, did you get a response?

Yes, but it was a request for clarification.  What they wanted clarified was clearly articulated in the original so my unsupported belief is that the request was merely a way to reset the clock and allow them to be compliant with their own rules.
965  Bitcoin / Development & Technical Discussion / Re: Idea for a next generation crypto distributed publicly by social network sign up on: January 23, 2014, 05:32:44 PM
How do you do this in a decentralized manner.  It is a non trivial problem.  That is why it hasn't been done.  There may not be a solution but if there is one it likely is radically outside the block. 

Saying something like "give someone x coins when they create a social media account" abstracts all the "problem" from the problem.  It would be like saying we need to go to the moon.  Step one we go to the moon. Step two we party because we went to the moon.

Making a centrally controlled virtual currency is trivial.  It is simply a ledger on a computer, and computers in the 1960s were good at crunching ledgers.  The fact that you put the ledger on multiple computers doesn't make it decentralized, if a single authority has the ability to change values in the ledger at will.   If you want a centralized virtual currency well just open a bank account.  Most of the dollars in the world exist only in cyberspace.  For all intents and purposes the dollar is a centralized virtual currency.

Now if you want a decentralized virtual currency well that is a lot harder.   When someone gets x coins for opening a social media account HOW EXACTLY do they get it?  HOW EXACTLY does the rest of the network verify that this issuance of new currency is valid?
966  Other / Off-topic / Re: What is the average processing fee for a credit card transaction? on: January 22, 2014, 04:54:52 AM
There is no standard rate, it varies significantly from one processor to another and one merchant risk profile to another.  Walmart (massive volume, card present=swiped, low fraud rate) probably secures custom deals for an interchange below 1%.

On the other hand an adult services company (low volume, card not present, high fraud rate) can end up paying 10% or more.  Throw in the cost of anti-fraud screening, and the cost of chargebacks (getting hit with $35 chargeback fee when your average purchase is <$35 is going to hurt your bottom line), the true cost can approach 15% or more.
967  Economy / Speculation / Re: Are wall street movies the best warning signal to sell on: January 21, 2014, 08:31:53 PM
So when there is a Bitcoin movie in theaters ...
968  Economy / Services / Re: Are there any press release distribution companies which accept Bitcoins? on: January 21, 2014, 05:53:15 PM
Thanks for the link.  I knew there had to be one.  The bad news is I already completed distribution but link definately saved for next time.
969  Other / Beginners & Help / Re: Using Bitcoin in Burger King... on: January 21, 2014, 12:56:52 AM
Sure, because when I use a credit card they have a fair degree of assurance that they know who I am, so they can come after me if I defraud them.

No they have a fair assurance who the cardholder is, which for millions of transactions totalling billions of dollars a year isn't the "purchaser".

Say a given merchant has a loss rate (due to CC fraud both third party fraud and so called friendly fraud) of 1% and pays 2% in CC fees.  If the merchant loses money to Bitcoin fraud but the loss rate is <3% the merchant's bottom line has improved.   Chasing a goal of zero fraud no matter how intrusive it is for the consumer isn't something merchants do with credit cards, why would they for Bitcoin?
970  Bitcoin / Bitcoin Discussion / Pretty cool Bitcoin promotion on "On The Morning Edition, CBC" (radio). on: January 20, 2014, 09:27:21 PM
Quote
On The Morning Edition, CBC Kitchener-Waterloo was able to successfully transmit Bitcoin over radio waves.

This makes what is believed to be the first known transmission of the digital currency by a public radio station.

A series of beeps were played over the air, and listeners were asked to use an app known as chirp.io to decipher a code produced by the sound.

Chris Skory of Rockland County, New York was the winning recipient, and unlocked 0.05 Bitcoin worth about $40. The Bitcoin was donated by Waterloo start-up Tinkercoin and a local Bitcoin enthusiast. 

"I did not think that I was going to get it at all," said Skory. "I figured there was going to be some pretty stiff competition out there."

http://www.cbc.ca/news/canada/kitchener-waterloo/cbc-kw-sends-bitcoin-over-the-airwaves-1.2503580
971  Bitcoin / Bitcoin Discussion / Re: Bitcoin Code access on: January 20, 2014, 08:34:36 PM
There is a button labeled "fork".  Fork the QT client repo to one you control and you can do whatever you want with it.  

Now if you are asking how to you convince millions of people to download your client over the one they already know and trust ... well that is a different non-technical problem.  Convince people you are trust worthy.
972  Economy / Services / Are there any press release distribution companies which accept Bitcoins? on: January 20, 2014, 05:06:40 PM
Are there any press release distribution companies which accept Bitcoins?

I always like to spend real money when I can.  I doubt it but I thought I would ask before using a company which only accepts dead presidents.
973  Other / Meta / Re: The forum needs to be divided: Software, Hardware/EE, Physics, Number Theory, .. on: January 20, 2014, 04:57:07 PM
How could number theory be considered off-topic?

That assertion is no less than insane.

Number theory as it relates to Bitcoin = on topic.
Number theory in general = off topic.

Pretty easy to understand.
974  Other / Meta / Re: The forum needs to be divided: Software, Hardware/EE, Physics, Number Theory, .. on: January 20, 2014, 04:53:30 PM
"This" happens on all boards of any size.  The reality is most people are lazy.  They don't take the time to educate themselves on where threads should go.  The forum probably needs more moderators.  The post volume is at an all time high.   

An interesting experiment would be a fine/reward system in micro amounts.  Post in the wrong forum and it has to be moved you are charged 2,000 satoshis.  Help a moderator out by reporting a violation you receive 500 satoshi reward.  Waste the moderator's time by reporting something valid lose 1,500 satoshis.  Something like that.

Of course you did help the mods out and clicked the report to moderator link on those posts which made you angry?  Right?
975  Bitcoin / Bitcoin Discussion / This is why Bitcoin will win ... the speed of credit cards with the cost of cash on: January 19, 2014, 11:15:47 PM
Just one example.  http://shinybars.com/t/paymentoptions

Simplified you have three options:
CC = fast + 3% more expensive
"Bank stuff" = slow + 3% cheaper
Bitcoin = fast + 3% cheaper
976  Other / CPU/GPU Bitcoin mining hardware / Re: Residential Limit 15amp or 20amp? on: January 19, 2014, 06:09:44 PM
********************************************************
THE DEATH&TAXES PROTOCOL :
30A breaker, 10AWG branch line, and a NEMA L6-30R outlet (240V, 30A, locking).  Cheap used datacenter PDU with a NEMA L6-30P plug.  "Good for a solid 5.76KW."

We're looking at setting up 2-3 server racks worth of GPU rigs, and powering them all off 220. Each rack will be using ~8kW (possible more), which will be 35+A if we use 220.

So lets say we use 9kW per rack. We'll just do two of your setups? 6x 30A 220 breakers. 10ga wire run ~100ft to our room, and terminate to 6x L6-30R outlets. 2x Compaq 24A PDUs with L6-30R plugs per rack. Then each PDU will power 3 rigs, with 3 of these, and 3 of these. Each rig will use an ATX PSU, and a custom server PSU.

The problem I see is that the breaker is 30A, but is being limited by the PDU to 24A, which if run at 80% load is only ~20A. Is it a better idea to install a single 50A 220 breaker per rack, run 8awg wire to 2x L6-30P, and then run both of the 24A PDUs per rack to those? The actual draw through the PDUs wouldn't change, would still be at ~20A each, but now you've got 8awg wire carrying 40A to a 50A breaker, rather than 2x 10awg carrying 20A each to 2x 30A breakers. Does this make sense?

Although I suppose if we did that, we're really limited to 40A per rack. If we went with the 2x 30A, we could upgrade the 2x 24A PDUs with 4x 16A PDUs. As long as each of the 16A PDUs were kept below 13A, the draw per breaker would be ~25A, and the total draw per rack could be increased to 52A, or 11kW.

So if we're only pulling 8-9kW per rack, we could get away with a 50A breaker and 2x 24A PDUs? But if we wanted to pull more, splitting it up between multiple breakers would give us more headroom?

I also realize that we're going to have to run another breaker for our AC equipment, but that's a topic for another day. And I'm not really an electrician, I'm just trying to figure this all out. Any suggestions welcome!


Edit: you may already know some of this the post became more of a general guide as I kept type.

So a couple things. The first is that in the use despite the common usage of "110" or "220" nominal voltage is 120 & 240.  You should however stick a multimeter into an outlet to find out.  You can test a 120V outlet, the 240V is going to be double that.  Unless you are at the far end of distribution the measured voltage at the outlet is probably closer to 120/240 then 110 /120.  Still you should adjust the power capacity numbers based off your actual voltage.

The second is that the 24A per PDU already includes the derate.  The circuit (30A breaker, 10AWG wiring, NEMA L6-30R outlet) is rated for 30A peak load.  Continual load is derated to 80% of that so 30A * 0.8 = 24A.  A code compliant circuit using a NEMA L6-30R receptacle (proper name for "outlet", outlets are actually on the PDU) is good for 24A continual or 30A peak.  That is why the PDU lists 24A, you don't have to further derate that.  Putting it together with 240V means each PDU is good for 5.76KW of power per pdu.

Third.  I am sure you have double checked but most PDU don't use C19 connectors.  I assume your server PDU is "beefy" and pulls more than 12A?  Just don't want someone ready briefly to be confused.  Almost all PSU have a C13 connector.  PDU have a C14 connector so the cable most people will need is one C13 to C14 cable per device. 

Are there other options? 
Well there are but I think you find any other option is either more expensive or not up to code. 240V, 30A circuit just comes down to the most economical choice.  The used PDUs are dirt cheap, the outlets are nothing "exotic" (even home depot probably carries a NEMA L6-30R is look hard enough) and while 10AWG isn't "fun" to work with it is still a lot easier to manage then the larger cables.  For example the standard connector for a 50A circuit is a NEMA L6-50R which doesn't exist except on paper, as safer pin and sleeve solutions are used. One common alternative is the California standard (become widely used in movie industry) http://www.stayonline.com/detail.aspx?id=6272 

Yeah that is the plug only.  $70.  A matching outlet is another $50 to $70.  To make a custom cable which converts a 50A connector into 2x NEMAL L6-30 plus the 50A outlet is probably going to run you $250+. Smiley  So I am not saying "the 30A route" is the only possible way to deliver power, I am just saying trust me I did the research and cost per KW you aren't going to find a cheaper, code compliant solution then using 240V 30A circuits and 30A (derated to 24A) used PDUs. 

So if you need <5.76KW of power.  Install one 30A breaker, make one 10AWG run, install one NEM L6-30R outlet, and get one matching PDU*.
So if you need <11.52KW of power.  Install two 30A breakers, make two 10AWG run, install two NEM L6-30R outlets, and get two matching PDUs*.
So if you need <17.28KW of power.  Install three 30A breakers, make three 10AWG run, install three NEM L6-30R outlets, and get three matching PDUs*.
So if you need <23.04KW of power.  Install four 30A breakers, make four 10AWG run, install four NEM L6-30R outlets, and get four matching PDUs*.
...

* I recommend using the APC AP9571.  There is nothing magical about it but APC made millions of them so there are tons of used ones on ebay for dirt cheap.


An ebay listing may say 208V but it works fine at 240V as well (datacenters are usually 208V which is why it lists that).
30A, 24A derated (5.76KW usable)
NEMA L6-30P plug to match your economical NAM L6-30R recptacle.
Has twelve C14 outlets.
Just get some C14 to C13 cables from amazon or monoprice (<$5 ea) to connect from the PDU to each PSU.

If the distance from your breaker panel to where your rigs (and NEMA L6-30R receptacles) is long you may want to consider a subpanel.  How this would work is that in your main panel one or more high current breakers is used and a single run with large gauge wire is run to a sub panel located near the rigs.   The sub panel then has say 6 30A breakers and short 10 AWG runs to NEMA L6-30R receptacles.

If you go another route just remember you can pull the full rated current of the PDU (as it has already been derated).  However the circuit (receptacle, wiring, and breaker) has to be designed such that the continual load is only 80% of the circuit peak capacity. Some PDU labels will list both capacities (30A, 24A derated).  This is why you will find PDUs generally are 12, 16, 24, or 40 Amps.

12A PDU is 80% of a 15A circuit (15A outlet, wiring, and breaker)
16A PDU is 80% of a 20A circuit (20A outlet, wiring, and breaker)
24A PDU is 80% of a 30A circuit (30A outlet, wiring, and breaker)  <- The sweet spot in terms of $/KW installed
40A PDU is 80% of a 50A circuit (50A outlet, wiring, and breaker)

Another piece of advice is to start small (i.e. start with enough gear to fill one 5.76KW "unit").  Speaking from personal experience when you start getting into the tens of KW of power range everything becomes more of a challenge especially cooling.  It is much easier to design a single efficient layout (say half a rack) and then scale out, rather than buying gear needing 30KW+ and then realizing the engineering, power, and cooling challenges are no joke.




977  Other / CPU/GPU Bitcoin mining hardware / Re: Residential Limit 15amp or 20amp? on: January 19, 2014, 04:50:21 AM
DeathAndTaxes!!! Nice link .. cheap too!!

Thanks for your contribution to this thread! Where is your btc tip jar..!! I wanna by ya beer!
Ill hunt you down from your site if i dont hear from you! Thanks again.

I don't need a tip (I got a coin or two in the ole wallet.dat), glad to help for free.  My miner days are likely coming to a close so I will need to mine vicariously.

<shamless plug mode>
If you absolutely feel the need to help ...  consider https://bitSimple.com next time you need to sell some coins to pay that electric bill. Smiley
</shameless plug mode>
978  Bitcoin / Development & Technical Discussion / Re: Simple Transaction Signing Question on: January 19, 2014, 04:46:32 AM
The handling of tx fee is a pretty ingenious solution.  The tx fee and the miner reward are loosely coupled.

There is one potential concern, since tx fees are implied not explicitly defined, a bug in a client could result in a loss of coins for a user.

For example say a flawed client created this tx:
TxIn0 (value 100 BTC)
TxOut0 (value 1 BTC)
TxOut1 (value 9 BTC)

You just paid 90 BTC as a fee to miners.  Likewise a flawed miner could actually "erase" coins permanently.  If the sum of all txs in a block is 1 BTC and the block reward is 25 BTC but due to a bug the miner makes the coinbase tx 25.1 BTC then the money supply would be reduced by 0.9 BTC permanently.

The major clients are at this point pretty extensively tested so the risk is more in custom implementations or using the rawtx RPC call (which is very low level and will let you do lots of dangerous things).

Bitcoin likely can't be changed at this point however for an altcoin the only (hindsight being 20/20) changes I would recommend would be:
a) make the tx fee explicitly defined in the tx header and thus a tx is invalid if the sum(inputs) =/= sum(outputs) + txfee
b) make a block invalid if coinbase =/= sum(tx_fees) + current_block_subsidy

979  Other / CPU/GPU Bitcoin mining hardware / Re: Residential Limit 15amp or 20amp? on: January 19, 2014, 01:15:17 AM
I had the same question on adapters and got some of these http://amzn.com/B004OC579E and these http://amzn.com/B00066HQ50. They both seem to do the trick, though I don't have my outlet installed yet, so haven't fired everything up. They're pretty inexpensive, especially if you have Prime.

While that will work they make cords which have the correct connectors on each end.  As pointed out above you just need a C13 to C14 cord.  It is pretty common in datacenters.   Almost all electronics (like your ATX PSU) have C14 inlet, and PDU have C13 outlet.  So a single cord is used for everything.  Servers, switches, routers, firewalls, etc.

For amazon:
http://www.amazon.com/Tripp-Lite-P004-004-18AWG-Connectors/dp/B003MG9F78/ref=sr_1_2?s=electronics&ie=UTF8&qid=1390094044&sr=1-2&keywords=C13+to+C14

This is just an example I typed "C13 to C14" there may be (and likely is a cheaper, better version) but the cable one needs to go from PDU to PSU without any adapters is a C13 to C14 cable.


980  Bitcoin / Development & Technical Discussion / Re: Simple Transaction Signing Question on: January 19, 2014, 12:42:58 AM
It is complicated and probably better to look at the code directly as trying to explain it in English at a high level abstracts some of it away so while this gets you closer it may still leave some parts ambiguous.  The second image is closer to the correct scenario however remember a tx can have multiple inputs as well.

A hash of the simplified transaction consisted of a SINGLE INPUT and ALL outputs is created, this is then signed by the private key for that input and stored in the "sig & pubkey" portion of that tx input.  
See here: https://en.bitcoin.it/w/images/en/e/e1/TxBinaryMap.png  

Bitcoin has no concept of "owners" or "wallets" so the change output isn't special, it is an output just like any other.  If you are going to create an illustration for explaining then creating one with only a single input is a poor choice because the next questions becomes how to dI handle n inputs where n >1.  An example with 2 inputs is better because 3, 20, or 4,000 inputs are handled exactly the same as 2.  

The other thing you need to divorce yourself from is you aren't sending funds from a public key, you are using a SPECIFIC unspent output of a prior tx as an input.  Yes in conversation we may say "he paid me from address 123..." but when trying to explain Bitcoin's working that falls apart and trying to understand anything becomes impossible.  An input is a specific unspent output.  It is referred to (in the Tx input) by the tx hash (of the prior tx) and index (which output of that tx is being used here).

So in more general form.   Lets assume you have a tx which consists of 3 unspent outputs (which become inputs for this tx) A, B, C and two new outputs Y & Z (it doesn't matter if one of these is a change address or you just happened to pay two different receivers with the exact amount of coins, an inputs is an input and an output is an output).

First the output scripts are created.  The "normal" Bitcoin tx is a PayToPubKeyHash, so likely the user provided you a Bitcoin Address.  This is reversed into the PubKeyHash and a script which locks that output of a specific value to a specific PubKeyHash is created.  The outputs are arranged in a sequence randomly (to avoid leaking which one is the "change").  Lets say Z become TxOut0 and Y becomes TxOut1.  The inputs are then arranged in a sequence randomly.  Lets say A becomes TxIn0, B becomes TxIn1, and C becomes TxIn2.

We now have a tx which looks like this:
TxIn0
Txin1
TxIn2
TxOut0
TxOut1


A simplified version of the tx is created (TxIn0, TxOut0, TxOut1) and hashed.  The hash is signed by the private key for A and stored in the Sig & PubKey section for TxIn0.
A simplified version of the tx is created (TxIn1, TxOut0, TxOut1)and hashed.  The hash is signed by the private key for B and stored in the Sig & PubKey section for TxIn1.
A simplified version of the tx is created (TxIn2, TxOut0, TxOut1) and hashed.  The hash is signed by the private key for C and stored in the Sig & PubKey section for TxIn2.

One last thing which may not be obvious.  There is no 1:1 relationship between inputs and private keys.  For example if you have two unspent outputs to the same address (say A & C) then they share the same private key however they are still unique and discrete inputs so the tx above would still have 3 inputs.

I likely am overlooking or misstating something here so I would double check all this against the code.



Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 [49] 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 ... 726
Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!