1 2011-05-21 00:00:02 <pfifo> phantomcircuit, for the first pass, the previous and merkel hashes, do i pass them as ascii or binary?
  2 2011-05-21 00:00:16 <phantomcircuit> pfifo, what?
  3 2011-05-21 00:00:23 <phantomcircuit> you mean hex encoded sha256 ?
  4 2011-05-21 00:00:26 <phantomcircuit> no it's binary
  5 2011-05-21 00:01:45 <pfifo> phantomcircuit, i mean, say i have a check sum that begind with 'a', do i pass the ascii value of a (92) or the hex value of a (10)
  6 2011-05-21 00:02:15 <luke-jr> &
  7 2011-05-21 00:02:26 <phantomcircuit> ok now i literally have no idea what you're trying to ask
  8 2011-05-21 00:02:27 <phantomcircuit> like
  9 2011-05-21 00:02:29 <phantomcircuit> at all
 10 2011-05-21 00:02:33 <jgarzik> me either
 11 2011-05-21 00:02:44 <jgarzik> maybe try ebcdic ;p
 12 2011-05-21 00:03:15 <luke-jr> lol
 13 2011-05-21 00:03:38 <pfifo> a checksum is the hexcidecmial representation of a number
 14 2011-05-21 00:04:04 <luke-jr> no
 15 2011-05-21 00:04:16 <luke-jr> the checksum IS the numbe
 16 2011-05-21 00:05:03 <pfifo> right, and should i pass the number, or the ascii
 17 2011-05-21 00:08:22 <pfifo> this comes into play on this page, at the top where their checksumming "hello" https://en.bitcoin.it/wiki/Protocol_specification  if you follow their example and checksum "hello" youll get the desired result, but for the second checksum, if you checksum "2cf24dba5fb0a30e26e83b2ac5b9e29e1b161e5c1fa7425e73043362938b9824" you will NOT get the second number listed there... this is because your checksumming the human readable form of the binary
 18 2011-05-21 00:09:15 <gjs278> ;;bc,mtgox
 19 2011-05-21 00:09:15 <gribble> {"ticker":{"high":7.1,"low":5.5772,"vol":46573,"buy":6.2009,"sell":6.39,"last":6.3993}}
 20 2011-05-21 00:10:54 <pfifo> pfifo@pfifo-desktop:~$ echo -n "hello" | sha256sum | sed 's/ *-.*//' | sha256sum
 21 2011-05-21 00:10:55 <pfifo> e62965c744c6b6df887dbcdc8349aa67ad79c762f2310b40dc444893719163fd  -
 22 2011-05-21 00:11:20 <pfifo> does that make sense now?
 23 2011-05-21 00:12:10 <jgarzik> pfifo: all of bitcoin is binary
 24 2011-05-21 00:12:23 <jgarzik> pfifo: it might get converted when output via HTTP JSON-RPC, but ignore that
 25 2011-05-21 00:12:58 <jgarzik> pfifo: you are hashing binary data, C++ data structures.  sha256(sha256(C++ 80-byte binary block header))
 26 2011-05-21 00:13:40 <pfifo> jgarzik, can you give me an example 80 byte block header and the checksum its supposed to generate
 27 2011-05-21 00:14:21 <jgarzik> pfifo: https://en.bitcoin.it/wiki/Block_hashing_algorithm
 28 2011-05-21 00:15:01 <jgarzik> pfifo: you can download and examine the data yourself.  block chain is fully open, as is the source code for bitcoin and all hashing miners (CPU and GPU).
 29 2011-05-21 00:15:31 <pfifo> jgarzik, yes that what im trying todo, but there isnt enough information in that page
 30 2011-05-21 00:16:31 <pfifo> i can write C, but not python, and im not intrested in a pool miner that uses a midstate but rater the from scratch block generation method, so I have yet to see a good example in code
 31 2011-05-21 00:17:14 <jgarzik> pfifo: see https://github.com/jgarzik/pyminer or https://github.com/jgarzik/cpuminer and especially https://github.com/bitcoin/bitcoin
 32 2011-05-21 00:18:32 <phantomcircuit> jgarzik, he said good and not python
 33 2011-05-21 00:18:36 <phantomcircuit> oh your cpuminer
 34 2011-05-21 00:18:37 <phantomcircuit> that works
 35 2011-05-21 00:20:09 <pfifo> jgarzik, look, ive check into cpuminer and bitcoin. I reserve IRC for when im desperate
 36 2011-05-21 00:22:51 <pfifo> if you have a piece of code that hasnt been optomized and performs only what im intrested in doing, that would be a great help, but threading and sse involved in thos clients makes it difficult to follow exactly what is happening
 37 2011-05-21 00:23:44 <jgarzik> pfifo: sha256_via.c
 38 2011-05-21 00:24:16 <jgarzik> pfifo: cpuminer is multi-threaded yes...  but each sha256 instance is single-threaded.
 39 2011-05-21 00:24:39 <oneman> jgarzik, tried to message u about something
 40 2011-05-21 00:24:42 <oneman> no reply
 41 2011-05-21 00:24:59 <jgarzik> oneman: yes, I saw the messages
 42 2011-05-21 00:25:05 <oneman> I think you would care
 43 2011-05-21 00:25:18 <jgarzik> oneman: why?
 44 2011-05-21 00:25:46 <oneman> I just thought you would
 45 2011-05-21 00:28:49 <pfifo> jgarzik, yeah that would be GREAT if I had a via to try it with
 46 2011-05-21 00:29:10 <jgarzik> pfifo: if you can read C source code, that shows you precisely what to do
 47 2011-05-21 00:29:15 <pfifo> jgarzik, as far as the code goes im doing EXACTLY that
 48 2011-05-21 00:29:33 <jgarzik> pfifo: replace "via_sha256" with your favorite sha256 algo
 49 2011-05-21 00:29:46 <jgarzik> pfifo: and make sure to get the stupid byteswapping right
 50 2011-05-21 00:29:47 <eamon> http://boards.4chan.org/g/res/17628295#17628462
 51 2011-05-21 00:29:53 <pfifo> jgarzik, the C algo uses a midstate
 52 2011-05-21 00:30:48 <jgarzik> pfifo: you asked for the simplest, straightforward, easy to read working example.  sha256_via.c.
 53 2011-05-21 00:30:52 <jgarzik> pfifo: midstate is optional
 54 2011-05-21 00:31:02 <jgarzik> pfifo: it saves time, that's all
 55 2011-05-21 00:32:07 <pfifo> jgarzik, that isnt a working example imo, wont run without the right hardware, so putting a printf() in there cant be done
 56 2011-05-21 00:32:30 <jgarzik> <jgarzik> pfifo: replace "via_sha256" with your favorite sha256 algo
 57 2011-05-21 00:32:30 <pfifo> "Illegal Instruction"
 58 2011-05-21 00:34:24 <pfifo> jgarzik, ohh i mis read, your telling me to swap out the actal function
 59 2011-05-21 00:41:04 <luke-jr> ;;bc,gen 85000
 60 2011-05-21 00:41:06 <gribble> The expected generation output, at 85000 Khps, given current difficulty of 244139.48158254 , is 0.350190849081 BTC per day and 0.0145912853784 BTC per hour.
 61 2011-05-21 00:43:34 <luke-jr> ;;bc,gen 196000
 62 2011-05-21 00:43:37 <gribble> The expected generation output, at 196000 Khps, given current difficulty of 244139.48158254 , is 0.807498899057 BTC per day and 0.0336457874607 BTC per hour.
 63 2011-05-21 00:48:12 <pfifo> jgarzik, I cant get it to work right
 64 2011-05-21 00:50:15 <phantomcircuit> pfifo, what are you trying to do...?
 65 2011-05-21 00:50:26 <vorlov> Diablo-D3: are you here?
 66 2011-05-21 00:51:33 <pfifo> phantomcircuit, compute the checksum of block# 125408, which is "0000000000003cc913fb9fd8f3970f00fb50ed22f4f533b07c5dbed9c2b0655d" but neither my C ode nor m PHP code will return the correct checksum
 67 2011-05-21 00:52:36 <oneman> If any coder here with opencl experience wants to work with me on my WebCL miner message me, or anyone who wants to donate to or invest in the project.
 68 2011-05-21 00:53:01 <billy_ran_away> Hey, I'm just sort of discovering bitcoin, is there any instructions that show to how to manually "solve" the Genesis block?  Like maybe in a high level language or bash or something?
 69 2011-05-21 00:53:09 <pfifo> IF someone could give me the 80 byte block header as its fed to the hasing alco, i would INSTANTLY know where i am going wrong
 70 2011-05-21 00:53:14 <phantomcircuit> pfifo, that's the hash of the block, not the block data
 71 2011-05-21 00:53:39 <pfifo> phantomcircuit, you misread
 72 2011-05-21 00:54:00 <luke-jr> anyone know how many MH I can get with a nvidia TNT 2?
 73 2011-05-21 00:54:18 <phantomcircuit> pfifo, im fairly certain i did not
 74 2011-05-21 00:54:48 <phantomcircuit> pfifo, you're trying to calculate the checksum in https://en.bitcoin.it/wiki/Protocol_specification#Message_structure
 75 2011-05-21 00:55:55 <pfifo> phantomcircuit, no you misread, im sure of it, i am not feeding THAT check sum to sha256(sha256())
 76 2011-05-21 00:56:14 <pfifo> i am feeding THIS "0100000000000000000031a9db7abffcdc972afd825c4d981e2aa35d72359edc239101ae897f6d19ee62909c079a7b3b23d14a280b97ed820c9d9116f74e7d269126f0be1411d74df2b9441a58857644"
 77 2011-05-21 00:57:24 <phantomcircuit> pfifo, as a string?
 78 2011-05-21 00:57:39 <phantomcircuit> or is that the hex of what you're sending to sha256
 79 2011-05-21 00:58:01 <pfifo> phantomcircuit, i converted the binary data into hex, i am feeding it binary data
 80 2011-05-21 00:58:16 <pfifo> my result is 085e2568b917bce65f32b3970939df55cd9649e596d70f6d913f915f3c6c10e8
 81 2011-05-21 01:00:02 <jgarzik> pfifo: are you byte- and word-swapping?
 82 2011-05-21 01:00:05 <pfifo> is that what the block header actually is? cause im 99% sure, im putting the header together wrong, either an endianess problem, or possibbly i have things in reverse order
 83 2011-05-21 01:00:13 <phantomcircuit> pfifo, the hash is little endian
 84 2011-05-21 01:01:07 <pfifo> can either of you give me the actual block header?
 85 2011-05-21 01:01:38 <alystair> I have an odd question
 86 2011-05-21 01:01:42 <eamon> [Tycho]: Do you make electronic music?
 87 2011-05-21 01:01:46 <alystair> why isn't pooled mining part of the normal client?
 88 2011-05-21 01:01:55 <phantomcircuit> pfifo, no but i can tell you that your prev_hash is wrong
 89 2011-05-21 01:02:06 <eamon> [Tycho]: If so, I bought one of your albums.
 90 2011-05-21 01:02:08 <alystair> it seems like the more beneficial thing to do
 91 2011-05-21 01:02:10 <pfifo> phantomcircuit, reversed?
 92 2011-05-21 01:02:26 <alystair> as more and more cpu cycles get used for bitcoin
 93 2011-05-21 01:02:45 <alystair> where a single client normally wouldn't be able to do anything
 94 2011-05-21 01:02:45 <phantomcircuit> pfifo, no, simply wrong there is no block with that hash
 95 2011-05-21 01:03:00 <phantomcircuit> pfifo, http://blockexplorer.com/block/00000000000031a9db7abffcdc972afd825c4d981e2aa35d72359edc239101ae
 96 2011-05-21 01:03:02 <eamon> GPU is the new CPU!
 97 2011-05-21 01:03:19 <phantomcircuit> alystair, you shouldn't be using the mining in the mainline client anyways
 98 2011-05-21 01:03:33 <pfifo> phantomcircuit, http://blockexplorer.com/b/125407
 99 2011-05-21 01:04:50 <pfifo> phantomcircuit, ahh i see
100 2011-05-21 01:05:16 <alystair> phantomcircuit... most people don't know that!
101 2011-05-21 01:05:20 <alystair> once bitcoin gets public
102 2011-05-21 01:05:24 <alystair> they will just be blowing away cpu cycles :S
103 2011-05-21 01:05:35 <alystair> might as well make mainline client more practical for the 'general bitcoin consumer'
104 2011-05-21 01:06:18 <phantomcircuit> alystair, it's been removed from the newest client iirc
105 2011-05-21 01:06:26 <alystair> ah neat!
106 2011-05-21 01:06:35 <alystair> so it's a true 'wallet' system then :)
107 2011-05-21 01:06:51 <alystair> software rather
108 2011-05-21 01:06:51 <pfifo> phantomcircuit, ok i fixed that problem, but still dont get the right checksum :(   sha256(sha256(01000000000000000000139abda7fbcfcd79a2df28c5d489e1a23ad52753e9cd321910ea98f7d691ee2609c970a9b7b3321da482b079de28c0d919617fe4d76219620feb1411d74df2b9441a58857644)) = 516d2f9d6b6e073a46ce788c344053f8b3b923c7f91c8b854a43a3308236cef6
109 2011-05-21 01:07:09 <alystair> any  folks here that know the base algorithm?
110 2011-05-21 01:07:18 <alystair> I have a question regarding splitting work into even smaller chunks
111 2011-05-21 01:08:11 <phantomcircuit> alystair, the work unit is already split into very small pieces
112 2011-05-21 01:08:47 <alystair> oh, per hash?
113 2011-05-21 01:09:28 <phantomcircuit> pfifo, merkle root is wrong now
114 2011-05-21 01:10:01 <phantomcircuit> alystair, what?
115 2011-05-21 01:10:21 <alystair> phantomcircuit: where is there an explanation, I saw a javascript based system but was wondering how everything really works
116 2011-05-21 01:10:30 <alystair> I didn't read anything, really
117 2011-05-21 01:10:46 <phantomcircuit> alystair, lol
118 2011-05-21 01:10:47 <phantomcircuit> uh
119 2011-05-21 01:10:58 <phantomcircuit> bitcoin is a distributed notary basically
120 2011-05-21 01:11:17 <alystair> ok not that simple :P
121 2011-05-21 01:11:26 <alystair> I meant on the technical level if I wanted to make my own js based thing
122 2011-05-21 01:11:41 <alystair> in-browser-mining magic
123 2011-05-21 01:11:54 <phantomcircuit> alystair, uh it's 2 rounds of sha256 on the block header until the double sha256 is less than the target
124 2011-05-21 01:12:20 <phantomcircuit> alystair, also mining in a browser is a huge waste of time, you'll generate a block maybe once every billion years or so
125 2011-05-21 01:12:46 <phantomcircuit> pfifo, where are you getting this data from?
126 2011-05-21 01:12:54 <pfifo> phantomcircuit, no the merkel is right, are you sure your comparing it to the right block? if so then please be more specific about what is wrong with it
127 2011-05-21 01:12:59 <alystair> phantomcircuit: it becomes easily distributed tho'
128 2011-05-21 01:13:13 <pfifo> phantomcircuit, http://blockexplorer.com/rawblock/0000000000003cc913fb9fd8f3970f00fb50ed22f4f533b07c5dbed9c2b0655d
129 2011-05-21 01:13:14 <phantomcircuit> alystair, yeah it's also hilariously slow
130 2011-05-21 01:13:19 <alystair> http://lorinhalpert.com/ipoc/gamegossip/gamegossip-redo.jpg
131 2011-05-21 01:13:20 <alystair> whoops
132 2011-05-21 01:13:25 <ezl> https://github.com/ezl/mtgox
133 2011-05-21 01:13:28 <ezl> python wrapper for the mtgox api
134 2011-05-21 01:13:34 <alystair> stupid paste hotkey
135 2011-05-21 01:15:41 <ne0futur> ezl: great, thanks for the link !
136 2011-05-21 01:15:54 <ne0futur> there is still no open source trading bot ?
137 2011-05-21 01:16:06 <ne0futur> even with a minimal intelligence
138 2011-05-21 01:16:52 <ezl> thats my plan...
139 2011-05-21 01:17:01 <ezl> but need a reasonable source for market data
140 2011-05-21 01:17:16 <ezl> what would it take to make a better exchange?
141 2011-05-21 01:18:37 <ne0futur> there are much data on bitcoincharts.com
142 2011-05-21 01:18:45 <ne0futur> mtgox and more
143 2011-05-21 01:19:22 <ezl> seems like making the exchange is the cash cow though
144 2011-05-21 01:20:35 <pfifo> phantomcircuit, any ideas?
145 2011-05-21 01:20:48 <gjs278> ;;bc,mtgox
146 2011-05-21 01:20:48 <gribble> {"ticker":{"high":7.04,"low":5.5772,"vol":46895,"buy":6.0434,"sell":6.1498,"last":6.1872}}
147 2011-05-21 01:21:05 <ne0futur> ezl: talk with MagicalTux he ll sure hear your suggestions
148 2011-05-21 01:21:14 <luke-jr> lol
149 2011-05-21 01:21:18 <ne0futur> nop ?
150 2011-05-21 01:21:31 <luke-jr> ne0futur: ezl's suggestion is that ezl gets the trade fees :P
151 2011-05-21 01:22:01 <ne0futur> I was asking of making an open source  trading bot
152 2011-05-21 01:22:24 <ezl> haha i'd love that
153 2011-05-21 01:22:29 <phantomcircuit> pfifo, all your hashes are reversed
154 2011-05-21 01:22:40 <ezl> i really just want a way to listen on a pipe for tick data
155 2011-05-21 01:23:03 <ezl> but if it didin't exist, i was thinking "well shit, maybe i should see what it takes to make an exchange, so i could just HAVE all the tick data"
156 2011-05-21 01:23:08 <ezl> reinventing the wheel
157 2011-05-21 01:23:10 <phantomcircuit> pfifo, like i said the block hashes are little endian
158 2011-05-21 01:23:56 <ne0futur> using bitcoincharts.com to gather all the data from different markets and giving this tick could be an idea nop ?
159 2011-05-21 01:26:31 <ne0futur> ezl: talk with tcatm sure he could be interested
160 2011-05-21 01:26:37 <ne0futur> and sure MagicalTux could help
161 2011-05-21 01:26:38 <pfifo> phantomcircuit, so like this? sha256(sha256(00000001ae019123dc9e35725da32a1e984d5c82fd2a97dcfcbf7adba931000000000000bef02691267d4ef716919d0c82ed970b284ad1233b7b9a079c9062ee196d7f894dd711141a44b9f244768558)) = 48211f278088b797234bf948ed4cc081923e7f8d9a8d8c19debec895ff31c49a
162 2011-05-21 01:27:13 <ezl> cool
163 2011-05-21 01:27:18 <ezl> ne0futur: you a trader?
164 2011-05-21 01:27:31 <pfifo> phantomcircuit, NO YOU MEAN LIKE THIS!!!!! sha256(sha256(01000000ea101932cde95327d53aa2e189d4c528dfa279cdcffba7bd9a13000000000000eb0f621962d7e47f6119d9c028de79b082a41d32b3b7a970c90926ee91d6f7981411d74df2b9441a58857644)) = 5d65b0c2d9be5d7cb033f5f422ed50fb000f97f3d89ffb13c93c000000000000
165 2011-05-21 01:27:42 <billy_ran_away> Hey, I'm just sort of discovering bitcoin, is there any instructions that show to how to manually "solve" the Genesis block?  Like maybe in a high level language or bash or something?
166 2011-05-21 01:27:49 <luke-jr> &
167 2011-05-21 01:27:59 <luke-jr> that makes no sense
168 2011-05-21 01:28:14 <ne0futur> ezl: more or less, trying to help new users getting in the bitcoin wave
169 2011-05-21 01:28:46 <pfifo> phantomcircuit, thanks so much, Ive been working on this for like 8 hours now, ill drop this info in the forum at some point
170 2011-05-21 01:28:51 <ne0futur> ezl: i use mtgox and bitcoin market and #bitcoin-otc but rarely daytrader
171 2011-05-21 01:29:03 <ne0futur> (not enough time a bot could help me :p )
172 2011-05-21 01:29:05 <phantomcircuit> pfifo, put it in the wiki block_hashing_algorithm page
173 2011-05-21 01:29:23 <jfksir> heh this is a pretty cool nick considering i just bashed on the keyboad
174 2011-05-21 01:29:57 <currentB> lol
175 2011-05-21 01:30:30 <phantomcircuit> billy_ran_away, uh that doesn't make much sense
176 2011-05-21 01:30:44 <phantomcircuit> billy_ran_away, you mean to calculate the hash of a block header?
177 2011-05-21 01:30:53 <phantomcircuit> pfifo, what language are you doing this in btw?
178 2011-05-21 01:31:14 <billy_ran_away> phantomcircuit: Yea, I'm mining right now with poclbm, but I want to understand it better.
179 2011-05-21 01:31:47 <billy_ran_away> phantomcircuit: So I wanted to maybe just solve an easy one that had already been solved.
180 2011-05-21 01:31:48 <ne0futur> ezl: and iirc the two of them like python too ;)
181 2011-05-21 01:31:48 <phantomcircuit> billy_ran_away, what kind of background do you have?
182 2011-05-21 01:31:51 <billy_ran_away> phantomcircuit: Programmer.
183 2011-05-21 01:32:10 <pfifo> phantomcircuit, fist i started on C, and it failed, i deleted that code and tried to do a midstate version as a sanity check, when that failed, I reverted to PHP for simplicity
184 2011-05-21 01:32:27 <phantomcircuit> billy_ran_away, bitcoin is a distributed notary system
185 2011-05-21 01:32:30 <phantomcircuit> I reverted to PHP for simplicity
186 2011-05-21 01:32:33 <phantomcircuit> lol wat
187 2011-05-21 01:32:56 <pfifo> phantomcircuit, PHP for me is very very easy
188 2011-05-21 01:32:58 <phantomcircuit> billy_ran_away, so basically the way it works is that you broadcast your transaction onto the network and the miners include it in their block
189 2011-05-21 01:33:01 <billy_ran_away> phantomcircuit: I know that, I just wanted to "experience" or understand the specifics of mining.
190 2011-05-21 01:33:07 <pfifo> phantomcircuit, like BASIC easy
191 2011-05-21 01:33:17 <jfksir> so if i modify the client to send a transaction with no fee...will ppl use it in their block atm?
192 2011-05-21 01:33:19 <phantomcircuit> pfifo, except it's not
193 2011-05-21 01:33:38 <jfksir> i look at the blocks and it seems like lots of fairly large transactions get through with no fee
194 2011-05-21 01:33:59 <billy_ran_away> phantomcircuit: I understand how the transactions and trading bitcoin works.  I just wanted to see the math behind the hash blocks and how it relates to the actual coins, especially in relation to mining.
195 2011-05-21 01:33:59 <phantomcircuit> jfksir, you'd just have to find a miner who would include free transactions
196 2011-05-21 01:34:11 <pfifo> phantomcircuit, http://fpaste.org/d2Nk/
197 2011-05-21 01:34:22 <phantomcircuit> billy_ran_away, you run sha256(sha256(block)) until it's < target
198 2011-05-21 01:34:40 <phantomcircuit> billy_ran_away, it's a lottery
199 2011-05-21 01:35:07 <billy_ran_away> phantomcircuit: Right now in my head I have an analogy that mining is like having your computer do scratch off's and as more computer do scratch offs the more little things your computer has to scratch off to see if it's current ticket is a winner.
200 2011-05-21 01:35:13 <phantomcircuit> pfifo, yeah so? that's slow as hell and about twice as long as my python implementation
201 2011-05-21 01:35:37 <billy_ran_away> And the faster your computer can scratch off the ticket the more times it can check in relation to the other computers also scratching...
202 2011-05-21 01:35:43 <phantomcircuit> billy_ran_away, more or less that's an accurate analogy
203 2011-05-21 01:35:44 <pfifo> phantomcircuit, speed has no relevance in learning how something works
204 2011-05-21 01:35:55 <phantomcircuit> pfifo, ok then
205 2011-05-21 01:36:24 <pfifo> phantomcircuit, and i dont know python, even examples escape me there
206 2011-05-21 01:36:41 <billy_ran_away> phantomcircuit: So I guess I wanted to walk though the easy scratch off myself... So I could see process for myself.  And I figured the Genesis block was the easiest.
207 2011-05-21 01:36:57 <phantomcircuit> billy_ran_away, uh you cant do the math by hand really
208 2011-05-21 01:37:02 <phantomcircuit> billy_ran_away, it would take you weeks
209 2011-05-21 01:37:21 <billy_ran_away> I know, by hand I meant program something myself either in Java, or bash, or ruby or perl
210 2011-05-21 01:37:45 <phantomcircuit> billy_ran_away, take the block data (which was 100% arbitrary) then hash it twice with sha256
211 2011-05-21 01:37:55 <_Maru_> when I make a transaction does it show up right away? or it takes 10 min before it shows up...
212 2011-05-21 01:38:18 <billy_ran_away> Okay cool
213 2011-05-21 01:38:33 <phantomcircuit> _Maru_, it will show up right away (if all goes well), but will be listed as unconfirmed
214 2011-05-21 01:39:01 <_Maru_> what are the chances of an unconfirmed transaction turn to a fail?
215 2011-05-21 01:39:26 <billy_ran_away> phantomcircuit: Which field is the block data in this? https://en.bitcoin.it/wiki/Genesis_block
216 2011-05-21 01:39:29 <phantomcircuit> _Maru_, probably not high, but it's happened
217 2011-05-21 01:40:32 <phantomcircuit> billy_ran_away, https://en.bitcoin.it/wiki/Block_hashing_algorithm
218 2011-05-21 01:40:48 <phantomcircuit> _Maru_, i'd make the choice based on how much money it is
219 2011-05-21 01:41:31 <Diablo-D3> [10:50:27] <vorlov> Diablo-D3: are you here?
220 2011-05-21 01:41:33 <Diablo-D3> yes
221 2011-05-21 01:41:47 <phantomcircuit> lies
222 2011-05-21 01:42:00 <pfifo> billy_ran_away, the block header and checksum for block 0 sha256(sha256(0100000000000000000000000000000000000000000000000000000000000000000000003ba3edfd7a7b12b27ac72c3e67768f617fc81bc3888a51323a9fb8aa4b1e5e4a29ab5f49ffff001d1dac2b7c)) = 000000000019d6689c085ae165831e934ff763ae46a2a6c172b3f1b60a8ce26f
223 2011-05-21 01:42:00 <vorlov> Diablo-D3: what can you use to overpower an ATI 5870 in linux
224 2011-05-21 01:42:11 <gjs278> radeon bios editor
225 2011-05-21 01:42:17 <billy_ran_away> pfifo: Awesome thanks.
226 2011-05-21 01:43:11 <pfifo> billy_ran_away, (in human readable ascii representation of course)
227 2011-05-21 01:43:48 <vorlov> and that i understand is only available in windows?
228 2011-05-21 01:44:09 <billy_ran_away> So in bash it's sha256sum `sha256sum `echo "0100000000000000000000000000000000000000000000000000000000000000000000003ba3edfd7a7b12b27ac72c3e67768f617fc81bc3888a51323a9fb8aa4b1e5e4a29ab5f49ffff001d1dac2b7c"``
229 2011-05-21 01:44:10 <billy_ran_away> ?
230 2011-05-21 01:44:21 <phantomcircuit> billy_ran_away, no
231 2011-05-21 01:44:33 <phantomcircuit> billy_ran_away, he gave you a hex encoded raw block
232 2011-05-21 01:44:40 <billy_ran_away> ohhhh
233 2011-05-21 01:44:58 <pfifo> billy_ran_away, nope wont work the command `sha256sum` outputs human readable ascii, you need rwa binary format
234 2011-05-21 01:45:04 <F4C3> god damn magento is a bitch if a framework to develop on
235 2011-05-21 01:45:51 <phantomcircuit> bitch of
236 2011-05-21 01:45:56 <phantomcircuit> it's a bitch of a framework
237 2011-05-21 01:46:01 <F4C3> thanks
238 2011-05-21 01:46:02 <phantomcircuit> some people...
239 2011-05-21 01:46:37 <pfifo> billy_ran_away, awk can convers ascii to binary, but i forget how
240 2011-05-21 01:46:39 <jfksir> error:14090086:SSL routines:SSL3_GET_SERVER_CERTIFICATE:certificate verify failed while accessing https://github.com/bitcoin/bitcoin.git/info/refs
241 2011-05-21 01:46:46 <phantomcircuit> haha bye time to cause mayhem
242 2011-05-21 01:46:48 <phantomcircuit> XDXD
243 2011-05-21 01:55:34 <vorlov> what do u guys normally get out of a 5870 ? in terms of mhash/s
244 2011-05-21 02:00:39 <gjs278> 401
245 2011-05-21 02:00:41 <gjs278> 925 core clock
246 2011-05-21 02:00:43 <gjs278> 300 mem
247 2011-05-21 02:00:49 <gjs278> latest poclbm
248 2011-05-21 02:01:11 <sm00th> how do you undclock mem on linux
249 2011-05-21 02:01:30 <sm00th> i  tried the atconfig --od stuff and it wont let me do it
250 2011-05-21 02:02:53 <gjs278> you can't do it on that
251 2011-05-21 02:03:03 <gjs278> you need to use AMDOverdriveCtl
252 2011-05-21 02:03:29 <vorlov> is there a way to respawn in gnome the way upstart respawns dying processes?
253 2011-05-21 02:09:56 <vorlov> will amdoverdrivectl let u do all that without reflashing the bios of a 5870
254 2011-05-21 02:09:57 <vorlov> ?
255 2011-05-21 02:10:33 <vorlov> i have kindof a noob question here... i just installed the deb package of amdoverdrivectl... whats the name of the binary to execute?
256 2011-05-21 02:12:19 <oneman> ill answer all your questions about AMD overdrive control for a few BTC
257 2011-05-21 02:12:22 <oneman> :D
258 2011-05-21 02:23:08 <vorlov> how were u able to set 925mgz with amdoverdrivectl
259 2011-05-21 02:23:09 <vorlov> ?
260 2011-05-21 02:23:12 <vorlov> i can only get 900
261 2011-05-21 02:23:53 <Diablo-D3> [11:42:00] <vorlov> Diablo-D3: what can you use to overpower an ATI 5870 in linux
262 2011-05-21 02:24:05 <Diablo-D3> other than what amdoverdrivectrl does?
263 2011-05-21 02:24:09 <Diablo-D3> nothing much
264 2011-05-21 02:24:14 <Diablo-D3> we dont have a bios editor yet
265 2011-05-21 02:24:29 <vorlov> gjs278 mentioned getting 925mhz on his clock
266 2011-05-21 02:24:30 <vorlov> in linux
267 2011-05-21 02:24:33 <vorlov> and im trying to figure out how
268 2011-05-21 02:25:37 <Diablo-D3> without a bios editor, different manufs have different bioses
269 2011-05-21 02:25:46 <erbs> hows it gooong
270 2011-05-21 02:25:50 <Diablo-D3> they're all one of AMD's revisions, but they have their clock/volt tables modified
271 2011-05-21 02:26:09 <Diablo-D3> swapping in random firmwares isnt recommended
272 2011-05-21 02:26:19 <Diablo-D3> but you can edit your existing one with an editor
273 2011-05-21 02:26:26 <Diablo-D3> but they're in windows only
274 2011-05-21 02:27:59 <vorlov> i c
275 2011-05-21 02:28:15 <vorlov> so basically im set with what i have in amdoverdrivectrl
276 2011-05-21 02:28:16 <vorlov> which really isn't much diff than what aticonfig lets me do
277 2011-05-21 02:28:16 <XX01XX> What's the usual rule for showing proof-of-work in a pool?
278 2011-05-21 02:28:54 <Diablo-D3> vorlov: well for one you can memory underclock
279 2011-05-21 02:28:56 <jfksir> process_begin: CreateProcess(NULL, -c "if not exist gcc_mswud mkdir gcc_mswud", ...) failed. make (e=2): The system cannot find the file specified.
280 2011-05-21 02:28:59 <Diablo-D3> XX01XX: I dont understand your question
281 2011-05-21 02:29:00 <jfksir> when i do wxWidgets-2.9.1uildmsw>mingw32-make -f makefile.gcc on windoze
282 2011-05-21 02:29:30 <XX01XX> What's the usual way pools decide who gets what share?
283 2011-05-21 02:29:38 <erbs> suuuuuppppppppppp
284 2011-05-21 02:29:48 <jfksir> am i supposed to be running inside some ming shell or something?
285 2011-05-21 02:29:51 <Diablo-D3> XX01XX: you have a login and passsword
286 2011-05-21 02:29:59 <erbs> XX01XX: poolmaster picks a number out of the air and divides everyones share by that
287 2011-05-21 02:30:14 <jfksir> hmm no is says DOS Shell...
288 2011-05-21 02:30:15 <erbs> (but he assures everyone the total # isn't inflated)
289 2011-05-21 02:30:17 <XX01XX> very scientific, erbs
290 2011-05-21 02:30:26 <Diablo-D3> XX01XX: your equestion is going to need to be fleshed out
291 2011-05-21 02:30:28 <jfksir> has anyone tried this out?  https://github.com/bitcoin/bitcoin/blob/master/doc/build-msw.txt
292 2011-05-21 02:30:48 <vorlov> Diablo-D3: i just tried that... got up to 400.... can u explain to me why underclocking does that???
293 2011-05-21 02:31:06 <XX01XX> How is it tracked how much work people in the pool have done?
294 2011-05-21 02:31:11 <Diablo-D3> vorlov: underclocking memory gives you a bigger power and thermal budget
295 2011-05-21 02:31:23 <jfksir> XX01XX: it doesn't
296 2011-05-21 02:31:29 <vorlov> Diablo-D3: layman terms....?
297 2011-05-21 02:31:34 <Diablo-D3> vorlov: if you're hitting VRM max temp already and your GPU is auto-throttling, it can lead to this
298 2011-05-21 02:31:39 <jfksir> XX01XX: it just tracks the number of blocks created
299 2011-05-21 02:31:46 <jfksir> within a given timeframe
300 2011-05-21 02:31:49 <Diablo-D3> vorlov: but you'd have to exceed 85c on most cards, although 5970s may or may not hit it earlier
301 2011-05-21 02:32:00 <erbs> XX01XX: the poolmaster tells everyone else what he says is the amount of work done by the pool
302 2011-05-21 02:32:05 <jfksir> if it's running at more than 2016 per 2 weeks, difficulty goes up
303 2011-05-21 02:32:15 <Diablo-D3> XX01XX: if you're asking how do you calculate prop payout, you count all shares since the last time a pool made a block
304 2011-05-21 02:32:21 <Diablo-D3> XX01XX: and then you divide accordinglyu
305 2011-05-21 02:32:34 <XX01XX> How are shares proven is the question.
306 2011-05-21 02:32:35 <Diablo-D3> with pps, its a flat 50/current diff
307 2011-05-21 02:32:40 <erbs> it could all be a lie
308 2011-05-21 02:32:41 <Diablo-D3> XX01XX: shares are diff 1 blocks
309 2011-05-21 02:32:43 <erbs> no one knows for sure
310 2011-05-21 02:32:54 <jfksir> oh i guess i answered the wrong qn there :/
311 2011-05-21 02:33:05 <Diablo-D3> XX01XX: they're proven exactly the same way normal blocks are
312 2011-05-21 02:33:19 <Diablo-D3> XX01XX: the pool just changes the requested diff value on the getwork response
313 2011-05-21 02:33:44 <XX01XX> Huh... seems like that would spew a bunch of crap across the network
314 2011-05-21 02:33:49 <Diablo-D3> it does
315 2011-05-21 02:33:55 <erbs> the algorithm could be: payout = your_shares / (total_shares * fudge_factor) * total_payout
316 2011-05-21 02:34:14 <Diablo-D3> XX01XX: pools are being overloaded with the number of "blocks" found
317 2011-05-21 02:34:38 <XX01XX> How about having people save their most difficult block created in a given blk interval?
318 2011-05-21 02:34:47 <Diablo-D3> XX01XX: its not a new network connection because miners pool connections for reuse typically (even though bitcoind itself doesnt support this)
319 2011-05-21 02:34:48 <erbs> that would be good
320 2011-05-21 02:34:51 <Diablo-D3> XX01XX: cant
321 2011-05-21 02:34:56 <XX01XX> why?
322 2011-05-21 02:34:58 <Diablo-D3> a miner is given the header for a block
323 2011-05-21 02:35:04 <Diablo-D3> you cant reverse engineer the block
324 2011-05-21 02:35:20 <XX01XX> Eh?
325 2011-05-21 02:35:25 <Diablo-D3> the block header contains, among other things, a merkle hash of the list of transactions in the block
326 2011-05-21 02:35:44 <Diablo-D3> the first tx on that list is a gen tx sending 50 btc to the pool.
327 2011-05-21 02:35:51 <jfksir> ahh i got it...had cygwin earlier in my path
328 2011-05-21 02:35:53 <vorlov> Diablo-D3: so let me get it right... if im running it right now at 400mhash (900clock and 300memory) and i can maintain a decent temp (<85c) with my fan at about 60-75... thats alright?
329 2011-05-21 02:35:55 <jfksir> building fine now
330 2011-05-21 02:35:59 <vorlov> or am i gonna fry the card leaving it like that?
331 2011-05-21 02:36:03 <XX01XX> No, I mean when a BLK announcement goes out, everyone turns in their best hash for that particular block... and then starts hashing on the new block.
332 2011-05-21 02:36:10 <Diablo-D3> vorlov: you shouldnt hit vrm max at 85c, but like I said, some cards suck there
333 2011-05-21 02:36:13 <XX01XX> Rahterh than spewing Diff 1 BLKs constantly
334 2011-05-21 02:36:20 <Diablo-D3> vorlov: but yes, the lower your temp, the more abuse you can give it
335 2011-05-21 02:36:26 <vorlov> wonderful
336 2011-05-21 02:36:33 <Diablo-D3> XX01XX: cant be done
337 2011-05-21 02:36:36 <vorlov> Diablo-D3: how do i discover what and am i hitting vrm max?
338 2011-05-21 02:36:38 <XX01XX> Why not?
339 2011-05-21 02:36:53 <erbs> good idea XX01XX
340 2011-05-21 02:36:54 <Diablo-D3> vorlov: you can only check VRM temps in windows on cards that support it
341 2011-05-21 02:37:00 <vorlov> i c
342 2011-05-21 02:37:06 <Diablo-D3> XX01XX: because its a proof of actual work in this case
343 2011-05-21 02:37:09 <XX01XX> If they can send in Diff 1 blks they can save blks and send in their blk with the highest difficulty
344 2011-05-21 02:37:17 <vorlov> well currently card tells me 59.50C at 60% fan
345 2011-05-21 02:37:19 <Diablo-D3> XX01XX: it needs to be a consistent low goal for all users
346 2011-05-21 02:37:20 <vorlov> and im getting 400mhash out of it
347 2011-05-21 02:37:25 <vorlov> am in the "safe zone"?
348 2011-05-21 02:37:36 <jfksir> 60c is fine
349 2011-05-21 02:37:40 <Diablo-D3> XX01XX: uh, dude, I just explained to you why they cannot withhold high difficiulty blocks
350 2011-05-21 02:38:12 <Diablo-D3> XX01XX: they CANNOT be submitted by the miner on it's own... the miner is NOT given the whole block and the miner cant change the block even if he wanted to
351 2011-05-21 02:38:15 <jfksir> what i don't get is what prevents pool participants from just saying they are doing work
352 2011-05-21 02:38:28 <XX01XX> diablo.. someone with little hash power would turn in a 1 diff block because that's all they could find in 10 minutes... someone with more hash power could find something more difficult.
353 2011-05-21 02:38:30 <jfksir> like, can' i write a program that just pretends to mine in a large pool
354 2011-05-21 02:38:33 <XX01XX> statistically speaking
355 2011-05-21 02:38:50 <Diablo-D3> XX01XX: you mean just require different diffs for more powerful users? I plan on doing that
356 2011-05-21 02:39:07 <erbs> is Diablo-D3 on the crack
357 2011-05-21 02:39:20 <Diablo-D3> XX01XX: you can say, make everyone do diff 10 blocks
358 2011-05-21 02:39:40 <XX01XX> Diablo... not REQUIRE... let people hash normally, and they  keep their most difficult block saved... when a BLK announcment goes out they send in their most difficult block.
359 2011-05-21 02:39:43 <Diablo-D3> jfksir: btw, the pool software solves those diff 1 blocks
360 2011-05-21 02:40:07 <Diablo-D3> jfksir: it knows what headers it sent you, and it also does sha256(sha256()) on the header your miner sent back
361 2011-05-21 02:40:17 <XX01XX> then you award share based on the difficulty of the blocks turned in.
362 2011-05-21 02:40:18 <Diablo-D3> jfksir: you cant lie precisesly by the same rules that make bitcoin function
363 2011-05-21 02:40:33 <Diablo-D3> XX01XX: thats nonsensical
364 2011-05-21 02:40:47 <jfksir> huh?
365 2011-05-21 02:40:49 <XX01XX> How so?
366 2011-05-21 02:40:55 <Diablo-D3> XX01XX: it makes mining far more variable for one
367 2011-05-21 02:41:18 <Diablo-D3> pool mining is supposed to make the btc output of mining more smooth for small users
368 2011-05-21 02:41:19 <jfksir> how could the pool managing software confirm i did in fact hash all the work it gave me?
369 2011-05-21 02:41:27 <jfksir> wouldn't it have to also have done that work?
370 2011-05-21 02:41:29 <Diablo-D3> jfksir: do you understand how mining works?
371 2011-05-21 02:41:53 <XX01XX> Statistically it should level out on a timeline, no?
372 2011-05-21 02:42:00 <sm00th> how much harder is difficulty 2 compared to difficulty 1?
373 2011-05-21 02:42:01 <jfksir> yes...take a number from network, make some transactions, figure out how to hash that so it start with a certain pattern
374 2011-05-21 02:42:04 <jfksir> something like that
375 2011-05-21 02:42:13 <x6763> https://en.bitcoin.it/wiki/Block_hashing_algorithm
376 2011-05-21 02:42:20 <jfksir> i don't mean fooling the network
377 2011-05-21 02:42:26 <jfksir> i just mean fooling the pool oprator
378 2011-05-21 02:42:41 <jfksir> what prevents me from just pretending
379 2011-05-21 02:42:48 <jfksir> and never finding a solution'
380 2011-05-21 02:43:06 <Diablo-D3> jfksir: kind of. you're given a header, you can change exactly one number in it, and you keep guessing until you either run out of numbers or you find a share of appropriate diff
381 2011-05-21 02:43:22 <sm00th> x6763 ty
382 2011-05-21 02:43:27 <sm00th> its linear thats what i thougbht
383 2011-05-21 02:43:31 <Diablo-D3> jfksir: that is at least 2^33 attempts between diff 1 blocks on average
384 2011-05-21 02:43:42 <Diablo-D3> jfksir: thats over 4 billion
385 2011-05-21 02:43:43 <jfksir> right...and i get that
386 2011-05-21 02:43:54 <Diablo-D3> to VERIFY it, you only need to do it _once_.
387 2011-05-21 02:43:55 <jfksir> but when i mine in deepbit, i never find a block
388 2011-05-21 02:43:58 <jfksir> or i haven't yet
389 2011-05-21 02:44:03 <sm00th> so its actually not that hard to find difficulty 2, 3, etc. blocks
390 2011-05-21 02:44:07 <jfksir> so how does it verify i'm doing anything at all?
391 2011-05-21 02:44:10 <sm00th> I think XX01XX's idea is pretty good
392 2011-05-21 02:44:23 <Diablo-D3> sm00th: diff ratings are how many diff 1 blocks does it take to do this
393 2011-05-21 02:44:43 <Diablo-D3> sm00th: we're at like 150k diff... 1 in 150k diff 1s will be a canidate for the next block
394 2011-05-21 02:45:03 <Diablo-D3> or 150k * 2^33 hash attempts
395 2011-05-21 02:45:06 <XX01XX> ;;bc,stats
396 2011-05-21 02:45:08 <gribble> Current Blocks: 125442 | Current Difficulty: 244139.48158254 | Next Difficulty At Block: 127007 | Next Difficulty In: 1565 blocks | Next Difficulty In About: 1 week, 0 days, 21 hours, 58 minutes, and 25 seconds | Next Difficulty Estimate: 335707.62753285
397 2011-05-21 02:45:16 <sm00th> right, so if somone is able to find a difficulty 1 block in a round, they'll probably find some 2s, 3s, etc.
398 2011-05-21 02:45:19 <Diablo-D3> jesus, we're that high already?
399 2011-05-21 02:45:24 <Diablo-D3> s/150k/244k/
400 2011-05-21 02:45:31 <Diablo-D3> sm00th: yes.
401 2011-05-21 02:45:34 <sm00th> maybe not every single round, but if you find 1 diff1 per round, you'll find a few diff2s per hour
402 2011-05-21 02:45:46 <Diablo-D3> sm00th: remember how fast your card is
403 2011-05-21 02:45:50 <jfksir> hmm i think i know what they must do
404 2011-05-21 02:46:01 <sm00th> so the mining payouts based on max difficulty found would smooth out pretty fast
405 2011-05-21 02:46:02 <Diablo-D3> it takes my 5850 doing 300mhash about 12 and a half minutes to find 50 shares.
406 2011-05-21 02:46:13 <Diablo-D3> sm00th: well, it would lopside payout
407 2011-05-21 02:46:16 <jfksir> when i'm looking for the really difficult block, i probably remember the hardest solution i've found so far
408 2011-05-21 02:46:24 <Diablo-D3> it'd basically count it as multiple shares
409 2011-05-21 02:46:29 <jfksir> and send that up to the pool operator as proof of work
410 2011-05-21 02:46:35 <jfksir> and it can verify in a single hash
411 2011-05-21 02:46:35 <XX01XX> jfksir... that's what I've been saying
412 2011-05-21 02:46:42 <transpi> hmmm. sell or hold?
413 2011-05-21 02:46:48 <Diablo-D3> except thats NOT proof of work in the pool sense
414 2011-05-21 02:46:50 <sm00th> right, but on average you only need to find 2 shares to find a difficulty 2
415 2011-05-21 02:46:52 <XX01XX> transpi... sell, they're on another slump
416 2011-05-21 02:46:58 <jfksir> XX01XX: ok what you were sayign sounded like you were just talking about the network and not my pool
417 2011-05-21 02:47:02 <transpi> thanks
418 2011-05-21 02:47:08 <Diablo-D3> what if I find 500 diff 1 blocks and not a diff 2?
419 2011-05-21 02:47:13 <jfksir> i didn't understand
420 2011-05-21 02:47:18 <Diablo-D3> I tell the pool, under XX01XX's scheme, 1.
421 2011-05-21 02:47:20 <sm00th> so if you can submit a difficulty 2, thats (statically speaking) proof-of-work that you have done two shares
422 2011-05-21 02:47:23 <Diablo-D3> how am I winning here?
423 2011-05-21 02:47:37 <Diablo-D3> or what if I did 0 work and got a diff 100k share?
424 2011-05-21 02:47:42 <Diablo-D3> do I get 100k shares here?
425 2011-05-21 02:47:43 <sm00th> if you can submit a difficulty 3, thats POW that you did 4 shares
426 2011-05-21 02:47:51 <Diablo-D3> sm00th: no, thats proof of 3
427 2011-05-21 02:47:55 <sm00th> right, sorry
428 2011-05-21 02:47:57 <sm00th> typo
429 2011-05-21 02:48:04 <erbs> it avgs out, even if each individual payout isn't exactly the amt of work you did
430 2011-05-21 02:48:07 <sm00th> but you dont actually need the network traffic for all 3
431 2011-05-21 02:48:13 <Diablo-D3> but like I said, I did 2^33 hashes total, and found a diff 100k
432 2011-05-21 02:48:19 <Diablo-D3> does that prove I did 100k?
433 2011-05-21 02:48:20 <Diablo-D3> no
434 2011-05-21 02:48:22 <Diablo-D3> it doesnt
435 2011-05-21 02:48:23 <jfksir> but the pool operator only wants the block solution, right?
436 2011-05-21 02:48:25 <Diablo-D3> it proves I did 1.
437 2011-05-21 02:48:37 <Diablo-D3> jfksir: he wants all solutions above the specified diff
438 2011-05-21 02:48:37 <jfksir> like, if he trusted everyone
439 2011-05-21 02:48:37 <sm00th> statistically its a lot more than 1
440 2011-05-21 02:48:43 <Diablo-D3> all pools currently do 1.
441 2011-05-21 02:48:47 <XX01XX> Sure...  but that's not likely to happen consistently for a given user unless they have the hardware for it.
442 2011-05-21 02:49:05 <Diablo-D3> the whole entire point of pools is supposed to smooth out payments.
443 2011-05-21 02:49:24 <XX01XX> Sure... and most of the time your proof-of-work would be representative.
444 2011-05-21 02:49:30 <XX01XX> roughly speaking
445 2011-05-21 02:49:36 <Diablo-D3> imo its not
446 2011-05-21 02:50:00 <sm00th> lol, XX01XX i think we need to give up
447 2011-05-21 02:50:05 <erbs> without auditing your work is useless anyway
448 2011-05-21 02:50:14 <sm00th> XX01XX but I think its a good idea
449 2011-05-21 02:50:30 <jfksir> man this wx thing takes forever to build
450 2011-05-21 02:50:31 <sm00th> i have a bunch of ideas for a pool, but no time to work on it...would love to use that one though
451 2011-05-21 02:50:33 <erbs> so you have to pick the right difficulty level to make it possible for mutual auditing
452 2011-05-21 02:51:52 <sm00th> btw i think its best if you send the hardest difficulty found right away
453 2011-05-21 02:52:07 <sm00th> otherwise you have to deal with synchronizing it at the end
454 2011-05-21 02:52:20 <jfksir> someone needs to figure out how to make gcc use ati cards to speed up compilation
455 2011-05-21 02:52:41 <erbs> dudesup
456 2011-05-21 02:52:52 <sm00th> XX01XX you see what I mean?
457 2011-05-21 02:52:55 <transpi> Is there a vairable to make the difficulty easier? I have a netbook and I have been mining for 5 days with nothing.
458 2011-05-21 02:53:07 <sm00th> transpi, CPU mining is worthless
459 2011-05-21 02:53:17 <sm00th> but if you want to play with it, you need to join a pool
460 2011-05-21 02:53:32 <XX01XX> sm00th... eh... you'd still get a burst of traffic at a new interval because everyone would announce their first block as their hardest.
461 2011-05-21 02:53:45 <XX01XX> and then more each time they found a new one.
462 2011-05-21 02:53:49 <sm00th> okay, you could spaces those out
463 2011-05-21 02:54:12 <sm00th> the new ones would be much slower, factor of 2 at each level
464 2011-05-21 02:54:24 <erbs> pool skimming is probably very common
465 2011-05-21 02:54:32 <erbs> the financial incentive is huge
466 2011-05-21 02:54:41 <transpi> for mining?
467 2011-05-21 02:54:57 <sm00th> transpi, yes join a pool for mining
468 2011-05-21 02:55:01 <transpi> I have been mining for 5 days with nothing!
469 2011-05-21 02:55:16 <sm00th> http://eligius.st
470 2011-05-21 02:55:17 <transpi> full power 24-7
471 2011-05-21 02:55:30 <XX01XX> transpi... how many khash ?
472 2011-05-21 02:55:41 <transpi> what is a kash?
473 2011-05-21 02:55:42 <sm00th> you're still not going to get much but at least you will see something happening
474 2011-05-21 02:55:48 <erbs> brosto
475 2011-05-21 02:56:02 <XX01XX> transpi... in the lower left corner is a number...
476 2011-05-21 02:56:06 <transpi> oh
477 2011-05-21 02:56:09 <XX01XX> khash/s
478 2011-05-21 02:56:15 <XX01XX> What is it?
479 2011-05-21 02:56:16 <transpi> It is a acer netbook?!
480 2011-05-21 02:56:35 <XX01XX> OF THE BITCOIN WINDOW
481 2011-05-21 02:56:39 <sm00th> XX01XX you can have pool members delay their first submission some random time over the first 2 minutes
482 2011-05-21 02:56:42 <transpi> oh sorry
483 2011-05-21 02:56:44 <transpi> wait
484 2011-05-21 02:57:01 <Namegduf> transpi: Mining on a netbook will not be worth the electricity
485 2011-05-21 02:57:01 <transpi> 177
486 2011-05-21 02:57:08 <transpi> what?
487 2011-05-21 02:57:12 <sm00th> Namegduf of course not
488 2011-05-21 02:57:13 <XX01XX> ;;bc,calc 177
489 2011-05-21 02:57:14 <gribble> The average time to generate a block at 177 Khps, given current difficulty of 244139.48158254 , is 187 years, 44 weeks, 3 days, 7 hours, 47 minutes, and 26 seconds
490 2011-05-21 02:57:19 <sm00th> but he's doing it for fun
491 2011-05-21 02:57:22 <transpi> I have been doing this for 5 days straight
492 2011-05-21 02:57:27 <Diablo-D3> [12:50:05] <erbs> without auditing your work is useless anyway
493 2011-05-21 02:57:27 <jfksir> apparnetly you need perl 5 installed to build bitcoin
494 2011-05-21 02:57:28 <Diablo-D3> I agree
495 2011-05-21 02:57:29 <XX01XX> transpi... keep mining for another two centuries and you'll get your bitcoins
496 2011-05-21 02:57:30 <transpi> nothing
497 2011-05-21 02:57:31 <transpi> 0.000
498 2011-05-21 02:57:37 <jfksir> would be nice if that was mentioned in the windows build readme
499 2011-05-21 02:57:43 <Diablo-D3> essentially XX01XX's method is random payout
500 2011-05-21 02:57:48 <Namegduf> transpi: That's not at alls surprising.
501 2011-05-21 02:57:49 <sm00th> he might get lucky
502 2011-05-21 02:57:57 <Diablo-D3> if I wanted random payout, I'd solo mine
503 2011-05-21 02:58:01 <sm00th> true story: with 500Mh I got two blocks in my first two days
504 2011-05-21 02:58:01 <transpi> when will it be a single bit coin?
505 2011-05-21 02:58:05 <transpi> 2 weeks?
506 2011-05-21 02:58:06 <Namegduf> At 150MH/s, it's 11 weeks.
507 2011-05-21 02:58:10 <Namegduf> At... what?
508 2011-05-21 02:58:12 <transpi> ok
509 2011-05-21 02:58:16 <Namegduf> Under 1MH/s
510 2011-05-21 02:58:20 <transpi> 177k
511 2011-05-21 02:58:27 <Namegduf> 177K
512 2011-05-21 02:58:30 <transpi> that is fast no?
513 2011-05-21 02:58:33 <XX01XX> Diablo-D3... On a short timescale, sure.   On a long timescale it balances out.
514 2011-05-21 02:58:34 <Namegduf> No, god no.
515 2011-05-21 02:58:44 <transpi> it is a netbook by acer...
516 2011-05-21 02:58:54 <XX01XX> Most people will report a fair share most of the time.
517 2011-05-21 02:58:55 <transpi> it loads firefox pretty quickly
518 2011-05-21 02:59:03 <sm00th> Diablo-D3 what about proportional payout? that's still random right?
519 2011-05-21 02:59:06 <Namegduf> transpi: It will take you, on average, 179 years
520 2011-05-21 02:59:09 <Namegduf> To generate a block
521 2011-05-21 02:59:13 <transpi> oh no
522 2011-05-21 02:59:27 <Namegduf> 1000 times yours is not really "good"
523 2011-05-21 02:59:27 <transpi> i wont even have this netbook in 179 years
524 2011-05-21 02:59:30 <Namegduf> It's a netbook
525 2011-05-21 02:59:31 <sm00th> transpi just go to the site I told you and set that up
526 2011-05-21 02:59:36 <Namegduf> Netbooks are not known for speed
527 2011-05-21 02:59:43 <transpi> 179 years = 1 bit coin?
528 2011-05-21 02:59:47 <sm00th> you will get some tiny fraction of a bitcoin pretty fast
529 2011-05-21 02:59:54 <transpi> ok
530 2011-05-21 02:59:57 <transpi> wow
531 2011-05-21 03:00:02 <Namegduf> 50, but you get them in a block of 50 and nothing until then.
532 2011-05-21 03:00:08 <sm00th> transpi, no its 187 years for 50 BTC
533 2011-05-21 03:00:17 <XX01XX> sm00th... you could have people delay reporting just their hardest block, too and spread the traffic out.
534 2011-05-21 03:00:21 <jfksir> and then only if you're not unlucky
535 2011-05-21 03:00:41 <transpi> Where is the button to make it mine for a netbook?
536 2011-05-21 03:00:44 <Diablo-D3> sm00th: no
537 2011-05-21 03:00:47 <transpi> to make it easier?!
538 2011-05-21 03:00:49 <sm00th> XX01XX well, how do you know it was done in time?
539 2011-05-21 03:00:51 <Diablo-D3> its just shares since last block find
540 2011-05-21 03:00:53 <Namegduf> transpi: There is no way to "make it easier"
541 2011-05-21 03:01:08 <Namegduf> The difficulty is shared, globally, and there's no way it can be easier for anyone.
542 2011-05-21 03:01:12 <transpi> Sorry, the difficulty rating ranges
543 2011-05-21 03:01:13 <sm00th> Diablo-D3 how long does it take to find a block?
544 2011-05-21 03:01:21 <Diablo-D3> sm00th: depends on how big the pool is
545 2011-05-21 03:01:24 <Namegduf> transpi: That's nothing different.
546 2011-05-21 03:01:27 <Diablo-D3> but the bigger the pool the smoother it is
547 2011-05-21 03:01:40 <sm00th> Diablo-D3 regardless of the size of the block, its still random
548 2011-05-21 03:01:41 <XX01XX> smooth.... the txns.   If the share blk they're turning in and the BLK announced aren't similar you know they've been hashing an old block
549 2011-05-21 03:01:47 <sm00th> sorry size of POOL
550 2011-05-21 03:02:31 <transpi> I have a dell pentium 4 will that work better?
551 2011-05-21 03:02:38 <Diablo-D3> sm00th: every 10 minutes on average isnt random
552 2011-05-21 03:02:50 <Namegduf> transpi: CPU mining will not get you a significant amount in any reasonable length of time.
553 2011-05-21 03:02:57 <XX01XX> every 10 minutes "on average" is just as random as the crap you've been complaining about
554 2011-05-21 03:02:59 <Diablo-D3> transpi: they dont make cpus fast enough
555 2011-05-21 03:02:59 <sm00th> XX01XX any work done after someone else finds the block is worthless (and will be gamed)
556 2011-05-21 03:03:07 <transpi> ok
557 2011-05-21 03:03:15 <Namegduf> A fairly decent graphics card will give you about 4 a week
558 2011-05-21 03:03:16 <XX01XX> I've seen block intervals as short as a few seconds
559 2011-05-21 03:03:21 <sm00th> Diablo-D3 no individual pool finds a block every 10 minutes
560 2011-05-21 03:03:23 <Namegduf> If it's ATI
561 2011-05-21 03:03:24 <Diablo-D3> yes, and Ive seen a 4 hour block
562 2011-05-21 03:03:35 <Diablo-D3> sm00th: no, but deepbit I swear is at 20 minutes now
563 2011-05-21 03:03:46 <Namegduf> transpi: You just can't get free money easily, sorry.
564 2011-05-21 03:03:54 <XX01XX> sm00th... then you'd have to require people to do their SHR anouncement as soon as they recieved the BLK announcement.
565 2011-05-21 03:03:55 <sm00th> Diablo-D3, sure, and deepbit is ruining bitcoin
566 2011-05-21 03:04:04 <transpi> Wait what kind of laptop do you use to do this?
567 2011-05-21 03:04:07 <erbs> sup dudeachoo
568 2011-05-21 03:04:07 <sm00th> XX01XX: then you get the flood of traffic
569 2011-05-21 03:04:19 <XX01XX> Which would cause a burst of traffic... but it probably isn't any worse than having a mega miner spew Diff1's at you
570 2011-05-21 03:04:23 <Diablo-D3> transpi: do NOT mine with a laptop
571 2011-05-21 03:04:49 <sm00th> XX01XX no but its worse than better solutions
572 2011-05-21 03:04:55 <jfksir> transpi: probably is fine if you're not paying the power bill, but don't expect much
573 2011-05-21 03:05:10 <sm00th> XX01XX actually it may be worse than diff 1s, at least those are random, somewhat
574 2011-05-21 03:05:14 <jfksir> you need an ATI graphics card for any serious results
575 2011-05-21 03:05:22 <erbs> just make the min share diff-2's, boom, half the traffic
576 2011-05-21 03:05:45 <Diablo-D3> erbs: yes but
577 2011-05-21 03:05:49 <Diablo-D3> I'd go straight to 8 I think
578 2011-05-21 03:05:53 <Diablo-D3> er
579 2011-05-21 03:05:54 <Diablo-D3> 16
580 2011-05-21 03:05:55 <sm00th> erbs: true, or adjust the difficulty to the worker
581 2011-05-21 03:06:13 <sm00th> erbs: target getting a few results per block on average
582 2011-05-21 03:06:18 <Diablo-D3> the hex text currently looks like eight 0s, four Fs, and the rest 0s
583 2011-05-21 03:06:18 <XX01XX> well, we've got enough crypto laying around... make them sign it when they get the BLK announcement.
584 2011-05-21 03:06:34 <Diablo-D3> diff 16 just makes that 9 0s.
585 2011-05-21 03:07:46 <sm00th> XX01XX how does that prove they had done the work before the block announcement?  it only proves they sent the reply after (which is obvious)
586 2011-05-21 03:08:26 <phantomcircuit> Diablo-D3, did you see all the people on the forum freaking out about some guy talking about fpgas?
587 2011-05-21 03:08:32 <XX01XX> sm00th... signatures usually incorporate the time/date it was signed, no?
588 2011-05-21 03:08:36 <XX01XX> bah, that could be faked...
589 2011-05-21 03:08:43 <Diablo-D3> phantomcircuit: dude, Ive been laughing at fpgas for like 9 months now
590 2011-05-21 03:08:45 <sm00th> XX01XX :)
591 2011-05-21 03:08:46 <Diablo-D3> maybe a whole year
592 2011-05-21 03:09:03 <Diablo-D3> BUT IT USES 2W.... it costs like $200+ you fucking dipshit
593 2011-05-21 03:09:11 <phantomcircuit> Diablo-D3, it's like people dont understand the size of an unrolled sha256 operation or something
594 2011-05-21 03:09:26 <Diablo-D3> phantomcircuit: its not hard, just look at my fucking miner of goddamned doom
595 2011-05-21 03:09:33 <sm00th> The FPGA people don't understand that there are always going to be people with free electricity
596 2011-05-21 03:09:42 <sm00th> 2W is too much
597 2011-05-21 03:09:51 <sm00th> if you are competing on energy costs
598 2011-05-21 03:10:01 <Diablo-D3> the FPGA people dont understand that there is always going to be a massively parallel DSP that does the task a shitload faster
599 2011-05-21 03:10:09 <phantomcircuit> Diablo-D3, no i meant the on chip size, an unrolled sha256 takes up a shit ton of fpga space...
600 2011-05-21 03:10:21 <Diablo-D3> phantomcircuit: yeah but you can get bigger fpgas
601 2011-05-21 03:10:32 <phantomcircuit> yeah but they cost a lot more than 200USD
602 2011-05-21 03:10:32 <sm00th> well, yeah for more $$
603 2011-05-21 03:10:33 <phantomcircuit> :P
604 2011-05-21 03:10:36 <Diablo-D3> fpgas win massively on power efficiency.... but rape you on costs
605 2011-05-21 03:10:43 <Diablo-D3> even what Art did is basically bullshit
606 2011-05-21 03:10:43 <sm00th> power is free
607 2011-05-21 03:11:05 <sm00th> Art has some alternate agenda
608 2011-05-21 03:11:12 <Diablo-D3> structured ASICs cost too much (although are a shitload cheaper than FPGAs, thats for sure)
609 2011-05-21 03:11:14 <Namegduf> You mean he works for the CIA?
610 2011-05-21 03:11:16 <Namegduf> Of course
611 2011-05-21 03:11:19 <Namegduf> It all makes sense now
612 2011-05-21 03:11:27 <Diablo-D3> I mean, if I wanted to beat bitcoin
613 2011-05-21 03:11:28 <sm00th> No I don't mean that
614 2011-05-21 03:11:33 <Namegduf> (You can say that as an alternative to an actual argument for how it makes sense)
615 2011-05-21 03:11:38 <Namegduf> (It works)
616 2011-05-21 03:11:41 <phantomcircuit> Diablo-D3, you'd only need like a million bucks
617 2011-05-21 03:11:47 <Diablo-D3> I'd give Art several million dollars and have him go make shit for like 45 or 65nm, not structured
618 2011-05-21 03:11:51 <phantomcircuit> which for financial institutions is like nothing
619 2011-05-21 03:11:51 <XX01XX> sm00th... I suppose you'd have to see in practice if people could consistently inflate their share with another two minutes of hashing after the BLK... after all they're losing time to contribute a share to the next block, anyway...
620 2011-05-21 03:12:01 <Diablo-D3> and then have him run off like a million cores
621 2011-05-21 03:12:18 <sm00th> XX01XX, yes but they'd know whether the previous block paid off or not
622 2011-05-21 03:12:24 <Diablo-D3> it'd merely double the hash power of bitcoi.
623 2011-05-21 03:12:27 <Diablo-D3> because
624 2011-05-21 03:12:28 <phantomcircuit> Diablo-D3, sure but you'd have chinese competitiors in a matter of days ;)
625 2011-05-21 03:12:32 <Diablo-D3> GPUs still fucking win here
626 2011-05-21 03:12:46 <Diablo-D3> phantomcircuit: so what
627 2011-05-21 03:12:49 <Diablo-D3> let them
628 2011-05-21 03:13:00 <sm00th> GPUs are going to get quite a bit more power efficient anyway
629 2011-05-21 03:13:01 <XX01XX> sm00th... I don't understand why that's significant...
630 2011-05-21 03:13:15 <Diablo-D3> 5xxx is about 2mhash/watt
631 2011-05-21 03:13:17 <sm00th> XX01XX because you know whether to work on the current block or work on "faking" the previoius block
632 2011-05-21 03:13:28 <XX01XX> sm00th... you'd send in SHR after each BLK regardless of who won it...
633 2011-05-21 03:13:33 <Diablo-D3> art's structred asics are like 10mhash/watt
634 2011-05-21 03:13:48 <Diablo-D3> and you measure efficiency in magnitudes to get useful comparisons
635 2011-05-21 03:13:58 <sm00th> XX01XX: but would you keep trying to improve your result on the previous block or not?
636 2011-05-21 03:14:04 <Diablo-D3> *structured
637 2011-05-21 03:14:26 <phantomcircuit> Diablo-D3, only? i thought he had like 2Ghash/s running on like 10W
638 2011-05-21 03:14:53 <Diablo-D3> no
639 2011-05-21 03:15:01 <XX01XX> sm00th... You'd have a window to submit the SHR... I suppose you could keep hashing on a previous block... but you're cannibalizing your time to hash ont he next block.
640 2011-05-21 03:15:23 <Diablo-D3> phantomcircuit: each chip I think is like 4w, and it does like a hundred mhash (I forget the exact numbers)
641 2011-05-21 03:15:35 <sm00th> XX01XX: if the your pool found the previous block, then you want to try to increase your share of it. its like pool hopping but worse
642 2011-05-21 03:15:50 <sm00th> I'm going with the maximize h/$ theory
643 2011-05-21 03:15:58 <sm00th> power is free
644 2011-05-21 03:15:59 <Diablo-D3> btw, XX01XX's idea fails for one reason
645 2011-05-21 03:16:01 <XX01XX> sm00th... nah, the SHR keep track whether your pool has won the block or not.
646 2011-05-21 03:16:02 <Diablo-D3> its difficult to implement well
647 2011-05-21 03:16:08 <Diablo-D3> simplicity is king.
648 2011-05-21 03:16:54 <sm00th> XX01XX I'm sorry I haven't worked with the code, so I don't know what SHR means
649 2011-05-21 03:17:16 <Diablo-D3> I think he means share
650 2011-05-21 03:17:32 <XX01XX> Sorry... the hardest block you've found in a given interval.
651 2011-05-21 03:17:46 <sm00th> but if its something you have to send back after the block, you can decide after the fact whether to try to inflate it
652 2011-05-21 03:17:50 <sm00th> that's no good
653 2011-05-21 03:17:58 <Diablo-D3> sm00th: no, now that you cant do
654 2011-05-21 03:18:00 <XX01XX> You can't inflate it unless you can do the hashing
655 2011-05-21 03:18:05 <Diablo-D3> the block still has to be provable
656 2011-05-21 03:18:10 <sm00th> XX01XX: right, you keep going
657 2011-05-21 03:18:14 <Diablo-D3> XX01XX's plan just fails
658 2011-05-21 03:18:26 <Diablo-D3> I sure as hell wont implement it, and I have the most used miner ever.
659 2011-05-21 03:18:27 <sm00th> for whatever the window is
660 2011-05-21 03:18:58 <sm00th> I don't see how it fails though
661 2011-05-21 03:19:01 <XX01XX> sm00th... you could, but that cannibalizes your share of the next block
662 2011-05-21 03:19:21 <sm00th> XX01XX: my point is, the next block might be worthless, the previous block is known not to be
663 2011-05-21 03:19:25 <Diablo-D3> theres no such thing as a free lunch, people.
664 2011-05-21 03:19:55 <XX01XX> sm00th... this is to keep track of your pool contribution... it's tracked whether the pool you're in finds the block or not.
665 2011-05-21 03:20:23 <sm00th> Let's say your pool found the last block after 1 block.  shares in that block are worth a LOT
666 2011-05-21 03:20:49 <sm00th> so you have an incentive to keep going on that block, rather than work on the next one which is worth less.  even for deepbit they're worth 50% on average
667 2011-05-21 03:20:53 <sm00th> for any other pool its a lot less
668 2011-05-21 03:21:19 <sm00th> Diablo-D3 I don't understand why it fails
669 2011-05-21 03:21:47 <Diablo-D3> sm00th: why his idea fails?
670 2011-05-21 03:22:01 <Diablo-D3> because he basically requires modification to miner software that does not improve the end user experience
671 2011-05-21 03:22:07 <Diablo-D3> its dead in the water right there and then
672 2011-05-21 03:22:25 <sm00th> It reduces the end user's network usage for mining, that's worth something
673 2011-05-21 03:22:37 <XX01XX> Ah... I see what you're saying.   I suppose it depends how much you could inflate your share by continuing to mine during the window
674 2011-05-21 03:22:48 <Diablo-D3> sm00th: yeah, so what? I can change my pool up to diff 16.
675 2011-05-21 03:22:54 <Diablo-D3> 16 times reduction.
676 2011-05-21 03:23:05 <sm00th> Diablo-D3 fair enough
677 2011-05-21 03:23:08 <XX01XX> ;;bc,calcd 700 16
678 2011-05-21 03:23:09 <gribble> The average time to generate a block at 700 Khps, given the supplied difficulty of 16, is 1 day, 3 hours, 16 minutes, and 10 seconds
679 2011-05-21 03:23:21 <Diablo-D3> and the big problem isnt network usage on the user side
680 2011-05-21 03:23:24 <Diablo-D3> its network usage on the pool side
681 2011-05-21 03:23:36 <Diablo-D3> deepbit is effectively under a constant ddos
682 2011-05-21 03:23:45 <Diablo-D3> not from some evil hacker, but from all the users mining
683 2011-05-21 03:23:50 <sm00th> Diablo-D3 well again that does improve the user's experience because they'll have less trouble connecting to the pool, etc.
684 2011-05-21 03:24:01 <Diablo-D3> sm00th: btw
685 2011-05-21 03:24:06 <Diablo-D3> that doesnt fix the nonce saturation issue
686 2011-05-21 03:24:18 <Diablo-D3> to fix this shit you need a three fold plan
687 2011-05-21 03:24:29 <Diablo-D3> 1) stop miners from keep updating their getwork (LP does this)
688 2011-05-21 03:24:48 <Diablo-D3> 2) stop miners from obsessively sending work (higher diff does this)
689 2011-05-21 03:25:23 <Diablo-D3> 3) allow miners to also molest the time field to stop nonce saturation (theres a proposal for this)
690 2011-05-21 03:25:30 <phantomcircuit> 2 could be accomplished with batching the updates aswell
691 2011-05-21 03:25:52 <phantomcircuit> and 3 is totally doable
692 2011-05-21 03:25:57 <Diablo-D3> phantomcircuit: technically its already batched
693 2011-05-21 03:26:04 <Diablo-D3> miners support TCP keepalive
694 2011-05-21 03:26:44 <Diablo-D3> the cost of opening a new TCP connection is obliterated, everything else is effectively free (or costs the same in true batched)
695 2011-05-21 03:26:44 <phantomcircuit> Diablo-D3, i more the miners knowing the real diff and only sending blocks with the real diff or groups of like 100 shares
696 2011-05-21 03:26:53 <phantomcircuit> you could compress 100 shares pretty well
697 2011-05-21 03:26:58 <Diablo-D3> phantomcircuit: yes but
698 2011-05-21 03:27:12 <Diablo-D3> that requires major modifications of miners
699 2011-05-21 03:27:16 <phantomcircuit> obviously that doesn't work with instant payments
700 2011-05-21 03:27:22 <phantomcircuit> Diablo-D3, yeah im aware
701 2011-05-21 03:27:40 <Diablo-D3> it also doesnt work with, say, looking into your crystal ball with LP.
702 2011-05-21 03:27:55 <Diablo-D3> since submitting shares after a block is over is considered rude.
703 2011-05-21 03:28:26 <phantomcircuit> yeah
704 2011-05-21 03:28:40 <phantomcircuit> you'd need a grace period for submitting shares after the block has been found
705 2011-05-21 03:28:47 <jfksir> building boost...last lib then showtime
706 2011-05-21 03:28:52 <Diablo-D3> its just added complexity without a real gain
707 2011-05-21 03:29:00 <phantomcircuit> i guess
708 2011-05-21 03:29:03 <Diablo-D3> if I want 100 less shares, I up diff 100
709 2011-05-21 03:29:24 <pfifo> phantomcircuit, http://pastebin.com/Ya3604J0
710 2011-05-21 03:29:25 <jfksir> Diablo-D3: do you run one of these pools?
711 2011-05-21 03:29:25 <phantomcircuit> yeah
712 2011-05-21 03:29:36 <Diablo-D3> jfksir: no, but Im writing pool software
713 2011-05-21 03:29:56 <sm00th> Diablo-D3 why?
714 2011-05-21 03:30:08 <phantomcircuit> Diablo-D3, that removes a major reason to use pools though, regular payment
715 2011-05-21 03:30:13 <jfksir> i'm not surprised how well deepbit is doing i haven't had even a single issue
716 2011-05-21 03:30:27 <jfksir> on bitcoinpool is was a comedy of problems
717 2011-05-21 03:30:47 <Diablo-D3> phantomcircuit: well, in the past few hours Ive created almost 1200 shares
718 2011-05-21 03:31:05 <phantomcircuit> pfifo, rofl @ the compile instructions :P
719 2011-05-21 03:31:06 <Diablo-D3> 1200/16 is still a usefully large number wrt payouts
720 2011-05-21 03:31:13 <phantomcircuit> Diablo-D3, yeah i guess
721 2011-05-21 03:31:33 <phantomcircuit> pfifo, ps use uint32_t from stdint.h
722 2011-05-21 03:31:56 <Diablo-D3> btw, Im not saying my pool with be 100% perfect gold on one day here
723 2011-05-21 03:32:07 <Diablo-D3> but Im aiming for a slush/deepbit type experience
724 2011-05-21 03:32:16 <pfifo> phantomcircuit, i know, uint32 is proper, but thats not really a concern for an example
725 2011-05-21 03:32:19 <_ape> i havent had any complaints about btcmine so far
726 2011-05-21 03:32:22 <_ape> no fees either
727 2011-05-21 03:32:30 <_ape> just seems stable as hell
728 2011-05-21 03:32:31 <Diablo-D3> worst case is I go down in flames a few times
729 2011-05-21 03:32:33 <phantomcircuit> pfifo, well that code wouldn't work on my system
730 2011-05-21 03:32:36 <_ape> they probably had a lot earlier on though
731 2011-05-21 03:32:37 <phantomcircuit> pfifo, so it kind of is
732 2011-05-21 03:33:05 <pfifo> phantomcircuit, adding a -m32 would fix it
733 2011-05-21 03:33:48 <phantomcircuit> pfifo, im aware ;)
734 2011-05-21 03:33:51 <pfifo> unless you on like a mips64 os something weird
735 2011-05-21 03:33:56 <erbs> yeah the solution is to put mining into the protocol which i outlined earlier. pretty easy to do and tamper-proof
736 2011-05-21 03:34:09 <phantomcircuit> pfifo, it would also require me to have the 32 bit glibc installed
737 2011-05-21 03:34:11 <phantomcircuit> which i dont
738 2011-05-21 03:34:17 <jfksir> i kind of like the fee
739 2011-05-21 03:34:25 <jfksir> gives the owner more of an incentive
740 2011-05-21 03:34:30 <phantomcircuit> erbs, what?
741 2011-05-21 03:34:35 <jfksir> to make sure it operates %100
742 2011-05-21 03:34:39 <jfksir> and the means, too
743 2011-05-21 03:34:48 <erbs> pooling rather
744 2011-05-21 03:35:45 <pfifo> phantomcircuit, good thing most people who code know about the difference between 32 and 64 bit architectures
745 2011-05-21 03:35:48 <Namegduf> I'm going to switch away from pool work
746 2011-05-21 03:35:49 <phantomcircuit> erbs, what protocol?
747 2011-05-21 03:35:56 <Namegduf> A lottery is more fun
748 2011-05-21 03:35:57 <phantomcircuit> pfifo, you'd be surprised....
749 2011-05-21 03:36:03 <Diablo-D3> pfifo: bwhahahaha
750 2011-05-21 03:36:04 <Diablo-D3> dude
751 2011-05-21 03:36:15 <phantomcircuit> im going to write an anonymous poker game over tor using bitcoins
752 2011-05-21 03:36:17 <phantomcircuit> LIKE A BOSS
753 2011-05-21 03:36:18 <Diablo-D3> most people, including $150k/yr "professionals" fail even at shit like that
754 2011-05-21 03:36:20 <erbs> jump in the pool, mind the sharks tho
755 2011-05-21 03:36:24 <Diablo-D3> phantomcircuit: ... goddamnit.
756 2011-05-21 03:36:36 <phantomcircuit> Diablo-D3, im going to destroy pokerstars.net
757 2011-05-21 03:36:37 <phantomcircuit> rofl
758 2011-05-21 03:36:41 <Diablo-D3> that isnt it
759 2011-05-21 03:36:49 <Diablo-D3> you do realize what I am now forced to do, right?
760 2011-05-21 03:36:57 <phantomcircuit> implement it yourself?
761 2011-05-21 03:37:05 <Diablo-D3> no, listen to aformentioned song.
762 2011-05-21 03:37:14 <phantomcircuit> lolololololoolol
763 2011-05-21 03:37:29 <phantomcircuit> Diablo-D3, TAKE A SHIT LIKE A BOSS
764 2011-05-21 03:37:36 <phantomcircuit> GET KILLED LIKE A BOSS
765 2011-05-21 03:37:37 <pfifo> you better make a iphone/android version too
766 2011-05-21 03:38:09 <_ape> <Namegduf> A lottery is more fun <-- combine them
767 2011-05-21 03:38:18 <Diablo-D3> fly into the the sun (like a boss)
768 2011-05-21 03:38:40 <Namegduf> _ape: I was thinking to just switch off the pooling, it's like a lottery.
769 2011-05-21 03:38:48 <phantomcircuit> shit ill even throw in some awesome crypto shit liek player shuffling
770 2011-05-21 03:38:53 <XX01XX> ape... y'know... the "hardest block" think for share work provides a bit of both.
771 2011-05-21 03:38:54 <Namegduf> Rarely a high value payout instead of consistent low reward.
772 2011-05-21 03:38:56 <phantomcircuit> LIKE IM ON A BOAT MOTHER FUCKA
773 2011-05-21 03:39:14 <Diablo-D3> phantomcircuit jizzes in his pants.
774 2011-05-21 03:39:20 <phantomcircuit> lololol
775 2011-05-21 03:39:21 <jfksir> oh another qn i had...
776 2011-05-21 03:39:26 <_ape> yeah
777 2011-05-21 03:39:32 <_ape> well if you take donations
778 2011-05-21 03:39:35 <phantomcircuit> also
779 2011-05-21 03:39:37 <_ape> you could put a % of them towards a lottery pool
780 2011-05-21 03:39:42 <_ape> which pays out when somebody solves a block or something
781 2011-05-21 03:39:45 <phantomcircuit> zomg interwebz is so fast here
782 2011-05-21 03:39:46 <_ape> as extra incentive to donate
783 2011-05-21 03:39:50 <jfksir> what prevents a worker in a pool from solving a block and taking the 50 BTC for themself?
784 2011-05-21 03:39:56 <Diablo-D3> ..........
785 2011-05-21 03:39:57 <_ape> make your shares in the lottery go up with donation %
786 2011-05-21 03:39:59 <phantomcircuit> my hdd is having trouble keeping up with these torrents
787 2011-05-21 03:40:00 <phantomcircuit> rofl
788 2011-05-21 03:40:04 <Diablo-D3> do I have to crowbar people?
789 2011-05-21 03:40:07 <jfksir> is it that the receiving addrss of the bitpool manager is in the hash?
790 2011-05-21 03:40:09 <phantomcircuit> jfksir, magical elfs
791 2011-05-21 03:40:18 <phantomcircuit> jfksir, or do you want a real answer?
792 2011-05-21 03:40:19 <Diablo-D3> jfksir: I know you were here when I explained this.
793 2011-05-21 03:40:22 <XX01XX> could have a pool that pays the txn fees to whatever user actually hashed the winning block.
794 2011-05-21 03:40:30 <_ape> haha
795 2011-05-21 03:40:35 <phantomcircuit> XX01XX, actually you cant
796 2011-05-21 03:40:43 <phantomcircuit> you can pay them with the next block though
797 2011-05-21 03:40:45 <phantomcircuit> LIKE A BOSS
798 2011-05-21 03:40:46 <Diablo-D3> phantomcircuit: actually you CAN
799 2011-05-21 03:40:48 <jfksir> Diablo-D3: i haven't been watching this whole time
800 2011-05-21 03:40:52 <Diablo-D3> its just difficult as fuck
801 2011-05-21 03:41:05 <jfksir> phantomcircuit: and yes, i want the real answer lol
802 2011-05-21 03:41:06 <Diablo-D3> you'd have to take the block apart inside bitcoin itself
803 2011-05-21 03:41:22 <phantomcircuit> Diablo-D3, well you already have to account for shares/user, so why not also wins/user
804 2011-05-21 03:41:37 <_ape> jfksir: its posted on the forum like 100 times
805 2011-05-21 03:41:40 <_ape> in every pool thread somebody asks
806 2011-05-21 03:41:47 <_ape> also the logs of this channel probably have it 50 times :X
807 2011-05-21 03:41:49 <Diablo-D3> phantomcircuit: yes but
808 2011-05-21 03:41:54 <phantomcircuit> jfksir, because the block the miners in a pool hash contains a generator transaction sending the 50 BTC to the pool
809 2011-05-21 03:42:00 <Diablo-D3> Im also going to have a dick size measurement system
810 2011-05-21 03:42:00 <XX01XX> probably a good idea to figure out how to distirbute the fees, as that's going to become the main value as the bounty goes down.
811 2011-05-21 03:42:03 <jfksir> so...what i said
812 2011-05-21 03:42:06 <jfksir> ok
813 2011-05-21 03:42:08 <phantomcircuit> jfksir, yes
814 2011-05-21 03:42:11 <jfksir> ty
815 2011-05-21 03:42:18 <phantomcircuit> Diablo-D3, you would lose sir
816 2011-05-21 03:42:22 <Diablo-D3> seriously, a dick size hall of fame
817 2011-05-21 03:42:33 <phantomcircuit> Diablo-D3, my cock is significantly over sized for such a white man
818 2011-05-21 03:42:37 <jfksir> _ape: thanks for being so friendly and helpful
819 2011-05-21 03:42:45 <Diablo-D3> phantomcircuit: mine is measured in astronomical units.
820 2011-05-21 03:43:42 <sm00th> What you know as the universe is actually contained within one skin cell of my dick
821 2011-05-21 03:44:02 <jfksir> heh you know that makes me wonder
822 2011-05-21 03:44:15 <XX01XX> sm00th... what you know as your dick...   is contained within one skin cell of MY dick.
823 2011-05-21 03:44:16 <Diablo-D3> are planets just the sperm in God's ballsack
824 2011-05-21 03:44:35 <sm00th> don't worry im ending it all today
825 2011-05-21 03:44:39 <XX01XX> It's dicks all the way down.
826 2011-05-21 03:44:41 <jfksir> will we have little molecular biology computers making bitcoin blocks at some point?
827 2011-05-21 03:44:54 <Diablo-D3> dude
828 2011-05-21 03:44:58 <Diablo-D3> we'll be kidnapping chinese kids
829 2011-05-21 03:45:01 <jfksir> maybe the skinn cell of you dick _could_ do hashing
830 2011-05-21 03:45:03 <Diablo-D3> and plugging them into a miner
831 2011-05-21 03:45:18 <Diablo-D3> and we'll have tom cruise standing in the background laughing manically
832 2011-05-21 03:45:24 <sm00th> nah FPGAs gonna take over
833 2011-05-21 03:45:27 <XX01XX> He does that anyway
834 2011-05-21 03:45:33 <_ape> i'd pay cash money fora dick hasher
835 2011-05-21 03:45:33 <Diablo-D3> XX01XX: true
836 2011-05-21 03:45:40 <sm00th> chinese kids can't compete with FPGAs
837 2011-05-21 03:45:48 <phantomcircuit> Diablo-D3, rofl
838 2011-05-21 03:45:48 <XX01XX> ape... use a cheese grater... same effect
839 2011-05-21 03:45:50 <Diablo-D3> _ape: we call those "eletric pencil sharpeners"
840 2011-05-21 03:45:52 <phantomcircuit> sm00th, RAIST
841 2011-05-21 03:46:08 <_ape> lol
842 2011-05-21 03:46:12 <phantomcircuit> lol 3000 KB/s
843 2011-05-21 03:46:17 <phantomcircuit> I LOVE THIS PLACE
844 2011-05-21 03:46:22 <jfksir> i dunno seems like eventually craig venter will make a bacteria that does hashing
845 2011-05-21 03:46:23 <_ape> i actually have an electric pencil sharpener like 2ft from my mining rigs
846 2011-05-21 03:46:24 <_ape> go figure
847 2011-05-21 03:46:25 <phantomcircuit> they dont have clean anything
848 2011-05-21 03:46:32 <phantomcircuit> but they have 3000 KB/s internet
849 2011-05-21 03:46:33 <phantomcircuit> lololol
850 2011-05-21 03:46:46 <Diablo-D3> thats only like
851 2011-05-21 03:46:48 <Diablo-D3> 24mbit/sec
852 2011-05-21 03:47:13 <Diablo-D3> you can get 1gibt/sec in the bay area.
853 2011-05-21 03:47:15 <phantomcircuit> i get 3 mbps dsl in san francisco
854 2011-05-21 03:47:16 <phantomcircuit> so
855 2011-05-21 03:47:18 <XX01XX> "only"
856 2011-05-21 03:47:23 <Diablo-D3> phantomcircuit: HAH
857 2011-05-21 03:47:39 <_ape> you cna grow ears on their backs at the same time
858 2011-05-21 03:47:39 <sm00th> i have a colo box on 1gb connection to every backbone in the universe
859 2011-05-21 03:47:40 <_ape> double your money
860 2011-05-21 03:47:40 <Diablo-D3> [01:47:12] <Diablo-D3> you can get 1gibt/sec in the bay area.
861 2011-05-21 03:47:41 <Diablo-D3> pwned.
862 2011-05-21 03:47:43 <phantomcircuit> Diablo-D3, lol you can only get 1 gbps downtown on specific streets in specific buildings
863 2011-05-21 03:48:02 <phantomcircuit> and it's like $400/month
864 2011-05-21 03:48:08 <Diablo-D3> hey Namegduf
865 2011-05-21 03:48:09 <sm00th> phantomcircuit, yes that's what i have
866 2011-05-21 03:48:15 <jfksir> i wonder if ppl with pet hamsters ever hook their wheels into the power grid
867 2011-05-21 03:48:24 <Diablo-D3> wasnt nenolod going to move to san francisco just to get gigabit?
868 2011-05-21 03:48:33 <XX01XX> and you can get 1gbps on your phone in Korea, probably.
869 2011-05-21 03:48:40 <phantomcircuit> Diablo-D3, that would be a seriously bad move
870 2011-05-21 03:48:48 <phantomcircuit> Diablo-D3, sf is a monopoly town
871 2011-05-21 03:48:57 <phantomcircuit> ATT/comcast have split the town in half
872 2011-05-21 03:48:58 <Diablo-D3> phantomcircuit: its nenolod, he makes like a million dollars a year
873 2011-05-21 03:49:01 <phantomcircuit> it's a fucking scam
874 2011-05-21 03:49:09 <Diablo-D3> hes thinking about having a commercial line put into his house
875 2011-05-21 03:49:10 <XX01XX> jfksir... hope they don't hook it up wrong so it's a motor rather than a generator.